Authors - Dhavasironmani R.R, Maria Joel. J, Siddharth S.V, Ajith Sundaram Abstract - Marketing research is essential to get the correct information about the consumers’ needs and their changing preferences. The evaluation of the Consumer Behaviour, attitude, perception and satisfaction level has been the subject of the market research very frequently. Health Drinks indeed are essential for every individual. The quantity of intake may vary according to the age, occupation, income level, size of the family, but everyone accepts that in order to cope up with the energy demands of the day-to-day life, and to defend oneself from the polluted environment, one should definitely consume any health drink supplementary to the food intake. Preferences get converted into a habit which is hard to change. It is evidenced from the study that certain health drinks are being consumed through generations that the customers develop a high degree of brand loyalty towards that brand.
Authors - Bhuvaneswari Perumal, Vaishnavi Moorthy, Gladius Jennifer H Abstract - According to UNICEF-India, 46% of maternal fatalities and 40% of neonatal fatalities transpire during labor or within the initial 24 hours post-delivery. Antepartum care includes routine surveillance, assessment of risks, and appropriate actions to enhance the health of the mother and fetus. Intrapartum care provides for safe labour and delivery with surveillance and appropriate management of complications by skilled personnel.This study aims at identifying the importance of holistic care in these stages and how it helps in preventing complications through the identification of high risk pregnancies which will be useful in avoiding the development of severe problems in future. The study uses analytical tools and Machine Learning models to analyze the health data and risk factors of pregnancy. In existing state of art they have inadequate early risk prediction with poor personalization. So the collected data includes several risk factors identified and classified based on their level of risk. The results of the attempts of applying various machine learning models and EDA methods to define the most important risk factors. This is a very large reduction and in line with the United Nations Sustainable Development Goals for the year 2030.The aim is to reduce maternal and neonatal morbidity and mortality. Lack of adequate management of intrapartum care can lead to postpartum problems to a large extent and thus affect the prenatal and fetal well-being.
Authors - Yash Sharma, Bramhansh Agarwal, Sindhu Chandra Sekharan, C. Kavitha, S. Umamaheswari Abstract - In recent years, interior design has played an increasingly important role in improving the look and usability of residential and commercial spaces. Although professional designers are often employed for this purpose, the process can be time-consuming and costly, with limited flexibility for personalized input. To address these limitations, an AI-assisted solution has been developed. This system employs Conditional Generative Adversarial Networks to analyze photographs of indoor environments alongside text descriptions that reflect user preferences such as desired furniture style, color schemes, and spatial arrangements. After processing the information, the tool provides a range of design suggestions tailored to the user’s specific needs. This method eliminates the need for repeated consultations and allows for rapid generation of unique, realistic interior layouts. The approach supports a more inclusive and affordable design experience, enabling individuals to explore personalized decor ideas efficiently. By merging visual data with linguistic inputs, the system presents a novel pathway for intuitive and responsive interior design support.
Authors - Puja Cholke, Om Yogesh Suhagir, Maroof Mustaq Mohammed Gadiwale, Srushti Pancham Mane, Sanika Suresh Mohite, Shreya Ramesh Phalke Abstract - Snake bites pose a severe public health risk, especially in rural and tropical regions, where delayed treatment often leads to fatalities. Existing systems struggle to classify snakes accurately based on symptoms, causing delays in administering the correct antidote. To address this issue, BiteSage (Snake Bite Antidote Suggester) utilizes data science and machine learning to classify snake bites as venomous or non-venomous based on user-reported symptoms and recommend the appropriate antidote. A chatbot interface assists users in symptom formulation and provides real-time counseling. Additionally, the system offers visualization tools to analyze global trends in snake bites, enhancing awareness and preparedness. The model ensures high precision in bite classification and antidote recommendations, backed by comprehensive data analytics. This research benefits medical professionals in remote areas and educates the public, helping to reduce fatalities and improve emergency response. By integrating AI-driven analysis, real-time assistance, and data visualization, BiteSage enhances medical decision-making and public awareness, ultimately saving lives.
Authors - Kruthiga S, Sindhu Chandra Sekharan, H.Summia Parveen, C. Kavitha, S. Umamaheswari Abstract - The transformative potential of deep learning techniques to revolutionize the landscape of medical image analysis, enabling accurate and efficient multi-disease prediction across a spectrum of critical health conditions. This work provides a solution to the early detection challenge of disease through prediction for Tuberculosis, Pneumonia, Glaucoma, and Brain Tumors using deep learning methods. By leveraging the expressive power of convolutional neural networks and transfer learning strategies, we have developed a robust framework capable of learning intricate patterns and subtle features indicative of diseases such as brain tumor, glaucoma, pneumonia, and tuberculosis. Through meticulous data preprocessing, model selection, and rigorous training and validation procedures, our approach ensures the reliability and generalizability of disease predictions, offering clinicians a powerful tool for early diagnosis and personalized treatment planning. The integration of Python programming language facilitates seamless implementation and deployment of our framework, making it accessible to healthcare practitioners and researchers alike. Overall, our study represents a significant advancement in the field of medical image analysis, with the potential to improve patient outcomes and revolutionize healthcare delivery on a global scale.
Authors - Pranay Meshram, Prakash Prasad Abstract - In the rapidly evolving digital landscape, data security has become paramount, necessitating innovative encryption techniques that balance computational efficiency with robust protection. This research introduces the New Efficient Selective Encryption Algorithm (DSEA), a novel approach to selective text encryption that addresses critical challenges in current cryptographic methods. By leveraging intelligent message analysis and strategic encryption, by providing a robust approach to safeguard valuable information at the same time as minimizing resource usage, DSEA addresses the need for privacy in a progressive manner. The approach utilizes proximity to structural properties of the message, such as the ratio of alphabetic characters, presence of vowels, and semantic connections, to inform the selection of encryption techniques. DSEA thus allows encryption to be applied at a more granular scale, using its identification and prioritization of sensitive text segments, which results in a significantly lower computation overhead compared to traditional techniques for full-document encryption. Experimental results show that DSEA has a better performance comparing with the existing selective encryption schemes, especially in the encryption time percentage, encryption processing time, and encryption proportion.
Authors - Bendre M. R., Vikhe V.P., Vanve G.B. Abstract - Within the carrier and business industries, there would be an ongoing demand for employees who are promoted to higher positions in the service and corporate sectors. The human resource team faces significant pressure to maintain employee commitment and motivation. Incentives such as promotions, bonuses, and wages are applied to motivate employees to feel closer to their work. The employee promotions are primarily deliberate, expressing gratitude for the employee's dedication to enhancing business standards, ensuring team competency, preventing talent from seeking other opportunities, and upholding the excessive degree of overall performance, all through the assessment year, human resources gather a significant quantity of facts on all elements of worker engagement events and activities. The data collected is continuously expanding in terms of employee service, but it is of little value if it does not provide meaningful insights. As a result, machine learning plays a crucial role in human resource analytics by extracting valuable information from collaborative employee data. The issue lies in the conventional approach to promotion, which is both time- and resource-intensive due to the numerous steps required for segregating and promoting employees. This had a significant impact on the smooth transition of employees into their new positions. Because of this reason, it's miles greater sensible if human assets can predict which workers are more legal and appropriate for advancement or upgrade, earnings increase, and so on. This research aims to propose or expect worker promotion. Utilizing machine learning techniques to forecast which employee might be eligible for a promotion, contingent on the data gathered and their previous achievements. To determine the likelihood of advancement probabilities the classification algorithms together with decision trees (DT), logistic regression (LR), random forests (RF), and k-means clustering are considered broadly utilized within the field. The k-nearest neighbors (K- NN), random forest (RF), and decision tree (DT) classifiers are applied to make the expected forecast.
Authors - Nishant Survase, Chitti Saharsh, Sachin Dhadwe, Krishnadeep Thakare, Yash Ishwarkar, Nilesh Pinjarkar Abstract - Railway stations, key transportation nodes, frequently have complicated layouts that disorient travelers, leading to delays. Conventional signage alone is not enough for effective navigation, particularly with increasing urban populations. Augmented Reality (AR) becomes a solution, superimposing virtual, step-by-step directions onto actual views through smartphones or AR glasses. This paper explores AR's capability to improve navigation in stations by combining GPS, GLONASS, and adaptive machine-learning algorithms. Both marker-based and markerless AR approaches, combined with realtime locationing, also offer custom guidance. Analytics of learning further refine user engagement, with increased feedback mechanisms as well as operational effectiveness. As such, AR can efficiently handle congestion, enhance accessibility for the disabled, and optimize passenger flows. Keywords: Augmented Reality (AR), Railway
Authors - Prafulla Bafna, Punam Nikam Abstract - Mental illness can be the reasons of extreme behavioral, emotional, and physical health issues. Majorly there are 4 mental disorders which are based on disposition, uneasiness, identity and insanity. Most of the times symptoms pertaining to these mental diseases are common. But remedies on each mental disorder is different. Due to the commonly existing symptoms of each disease, identifying the exact type of mental disorder is difficult. To smoothen the process of identifying exact mental disorder we use machine learning algorithms. The algorithms are executed on 1020 patient records containing nine parameters which show mental status such as l consciousness level, general behavior, and so on. To predict the exact type of mental clutter/disorder , KNN and SVM are implemented using 80 :20 ratio of training-to-testing data. SVM proved to be more accurate that is low misclassification error and greater recall. The accuracy of prediction is steady for 300 to 1020 records.
Authors - Supriya, Ananya G Bhat, Chandana B A, Niharika P Abstract - This study seeks to enhance the accuracy of credit card fraud detection by utilizing advanced machine learning techniques, with a specific focus on the XG Boost algorithm. Various ML approaches, including Decision Trees, Logistic Regression, Naive Bayes, Random Forest, and XG Boost, are evaluated for their efficiency in detecting fraudulent transactions using patterns derived from historical data. Recent advancements highlight the integration of diverse authentication methods and randomized training datasets to mitigate vulnerabilities in fraud detection systems.
Authors - Saraswati Patil, Mustafa Limdiyawala, M.S. Dawngliana Fanai, Meghaj Kharwadkar, Shivshankar Mahajan Abstract - Natural disasters such as earthquakes, floods, and tsunamis pose severe threats to human lives, infrastructure, and economies. Effective prediction and response strategies are vital for minimizing their impact. This paper introduces an AI-driven Disaster Prediction and Relief Dashboard, an integrated platform leveraging machine learning and geospatial mapping to forecast natural disasters and optimize relief operations. Using Random Forest and Gradient Boosting algorithms trained on historical data, the system predicts the likelihood, magnitude, and severity of disasters. Geospatial visualization highlights high-risk zones and delivers real-time situational awareness for authorities. Additionally, the platform streamlines relief management by dynamically allocating resources based on predicted disaster severity and location. By integrating predictive analytics with operational planning, the system enhances preparedness and responsiveness, contributing to more resilient disaster management.
Authors - N.N.S.S.S. Adithya, P. Vanishree Sah, B. Jyothirmai, Nidhi Mishra, D. Indira Abstract - Early prediction of lung cancer is crucial for reducing the death rate. Artificial intelligence, particularly deep learning, is employed to analyze CT scan images for more accurate automated prediction of types of lung cancer. This process of prediction is called classification. Lung cancer classification can be done with pre-networks such as VGG16 and ResNet50.But the main drawback of these techniques is that cancer cannot be detected on Histopathology images (i.e. image of tissues). As VGG16 and ResNet50 are designed for more general usage, they are not suitable for analyzing Histopathology image. This study involves the development of a customized neural network model which can solve the problem of analyzing Histopathology images. This CNN model can help us detect lung cancer at a very early state in lung tissues. Detecting lung cancer at a very early stage can help doctors to cure the patient and save the life of a patient.
Authors - Sneha S. Biradar, Suvarna Kanakaraddi, Neha Tarannum Pendari Abstract - This personalized music therapy framework for children with Autism Spectrum Disorder (ASD) involves data collection (AQ scores, age, gender, and demographics), severity identification, and customized music creation. Using a publicly available dataset, children were classified by severity, allowing the design of therapeutic music with varying duration, tempo, and complexity. Compositions were set to 15 minutes for low severity, 30 minutes for moderate, and 50-60 minutes for high severity. Preliminary results suggest this approach boosts engagement and may improve cognitive, emotional, and social outcomes, demonstrating the potential of combining advanced analytics with personalized music therapy for ASD.
Authors - Ruby S Chanda, Vanishree Pabalkar, Priya Pradipkumar Tiwary Abstract - Companies are increasingly using data analytics and AI to personalize interactions and provide tailored recommendations. Additionally, there is a growing focus on emotional intelligence and understanding customers' needs beyond their transactional behavior. Some real-world examples are – Netflix uses AI to analyze viewing history and preferences, recommending personalized content, Spotify leverages data on listening habits to create tailored playlists and discover new music. The need for AI models in NPS and CX enhancement arises from the increasing complexity of customer interactions and the vast amount of data generated. AI can help in predicting consumer’s future behavior by analysing their demographic and purchase data and identifying patterns. This empowers businesses to create more tailored and meaningful customer experiences, resulting in greater satisfaction and loyalty. This project aims to develop a Generative AI-based Net Promoter Score (NPS) predictor to enhance Customer Experience (CX) in the retail industry. By leveraging advanced AI techniques like VAEs and GANs, the model will be able to analyze vast datasets of consumer behavior and demographics, providing more accurate and personalized NPS predictions.
Authors - Shriraj A. Patil, Chudaman D. Sukte, Jayesh R. Patil, Chinmay R. Mhaske, Mandar Dakhorkar, Manohar K. Kodmelwar Abstract - This paper presents a novel IoT-based fruit-picking system that integrates Reinforcement Learning (RL), Transfer Learning (TL), and Neuroevolution to address the inefficiencies of current robotic harvesting methods. As demand for efficient agricultural practices rises, traditional fruit-picking systems face significant challenges, including operational inefficiencies, fruit damage, and limited adaptability to diverse environments. Our proposed solution leverages RL to optimize picking strategies through adaptive learning, enhancing the robotic arm's efficiency over time. TL is employed to improve fruit recognition capabilities, utilizing pre-trained models for accurate ripeness detection, even with limited training data for specific fruit varieties. Additionally, Neuroevolution evolves control strategies for the robotic arm, enabling it to adapt to dynamic harvesting conditions. Comprehensive simulations demonstrate significant improvements in picking accuracy, efficiency, and adaptability compared to existing methods. The findings highlight the potential of integrating these AI models within IoT frameworks to revolutionize fruit harvesting, ultimately contributing to smarter farming practices and enhanced agricultural productivity. This research underscores the interdisciplinary nature of modern agriculture, combining advancements in AI, robotics, and IoT technologies to provide innovative solutions for the challenges facing the agricultural sector today.
Authors - Gatla Vijayendher, K Sai Karthikeya, E Jayanth Madhav, Tadepalli Satya Kiranmai Abstract - The increasing number of tumor cases has caused an alarming situation in the health care space. The tumor detection and diagnosis is a very computationally heavy and requires multiple medical imaging devices such as Magnetic Resonance Imaging (MRI) and Computed Tomography (CT). It is very vital in order for early detection of tumor which can be done by precisely measuring their size which can improve a treatment by a huge factor in the patients. Existing diagnostic approaches often face challenges due to the diverse appearances of tumors and the constraints of current models. This study examines the different cutting-edge deep learning methods, with a focus on utilizing Generative Adversarial Networks (GANs) to enhance tumor identification across various categories, types, and imaging techniques. We also investigate the role of data augmentation strategies in enhancing the model performance. Furthermore, we examine the integration of Convolution Neural Networks (CNNs) to achieve accurate and robust results while preserving the data privacy. The goal of this study is to understand the detailed current scenario of the early detection and ways about the different techniques in various kinds of tumors which are present in the human body.
Authors - Amol More, Sanjeev Kumar, Sandeep Kore Abstract - This study carefully checks how well three different types of heat sinks move heat under controlled laboratory conditions. The main focus is on Copper Pin Fin Heat Sinks, Aluminum Phase Change Material (PCM) Pocketed Heat Sinks, and Aluminum Plate Fin Heat Sinks (PFHS). These were put through a wind tunnel test that simulated forced convection. This gave a thorough comparison of how well they kept heat in. To make the experiments work, heat was applied to the bottom of the heat sinks with 10W, 20W, and 30W of power, to represent various thermal loads. The speed of the air was changed from 1 m/s to 5 m/s to see how speed affected how well heat was removed. It was also improved by making changes like adding a copper plate to the aluminum fins and making holes in both the shield connection and the pin fins which were used in the experiment. By changing the surface area and turbulence, these changes are meant to see if they can improve the rate of heat transfer. The study's results should give us useful information about how to build heat sinks so they work best in a wide range of situations, from home electronics to industrial systems. It is expected that the results will help make thermal management solutions that work better, which will improve the performance and life of heat-sensitive parts.
Wednesday August 26, 2026 9:30am - 11:30am IST Virtual Room BGOA, India
Authors - Ayinampudi Siva Rama Raju, Suneetha Dwarapu, Gaddiboyina Sai Jahnavi, Tummalapalli Sai Sri Varshit, Kalimahanthi Sai Nikhil Kartikeya Abstract - Sentiment analysis, a key task in natural language processing (NLP), identifies emotional tone in text. This study compares two sentiment classification approaches: a lexicon-based method using VADER (Valence Aware Dictionary and sEntiment Reasoner) and a transformer-based deep learning method with RoBERTa (Robustly Optimized BERT Pretraining Approach). Using the Amazon Fine Food Reviews dataset of 568,454 customer reviews, the analysis categorizes sentiments as positive or negative. Preprocessing steps, including text normalization and handling missing values, ensure data reliability. VADER efficiently processes short, informal texts using a predefined lexicon but struggles with complex linguistic structures and contextual subtleties. RoBERTa leverages transformer-based architectures to capture intricate word relationships, enabling superior accuracy and nuanced sentiment detection in contextually rich texts. A comparative evaluation demonstrates that RoBERTa outperforms VADER by a significant margin, underscoring the strengths of deep learning for detailed sentiment analysis. These findings emphasize the trade-offs between speed and contextual depth in sentiment analysis models and provide valuable insights for customer feedback interpretation, opinion mining, and broader NLP research.
Wednesday August 26, 2026 9:30am - 11:30am IST Virtual Room BGOA, India
Authors - Ruby S Chanda, Rahul Dhaigude Abstract - In order to predict stock values, this study investigates the combination of machine learning with sentiment analysis. The quick spread of news and the growth of social media sites like Twitter have made public opinion a bigger factor in financial markets. This study extracts market sentiment from tweets and news stories using Natural Language Processing (NLP) techniques, namely the VADER sentiment analysis tool, which has been tailored with financial lexicons. To forecast stock price fluctuations for firms like Amazon and Tesla, sentiment data is included into a Generative Adversarial Network (GAN) model together with technical indicators like moving averages and Bollinger Bands. The model is evaluated using performance metrics like Root Mean Square Error (RMSE), demonstrating its ability to capture price trends and market sentiment dynamics. While results highlight the potential of GANs for real-world applications in financial trading, the study also acknowledges limitations such as data quality and model uncertainty. Future directions include improving sentiment algorithms and incorporating additional market factors..
Authors - Aanjaneya K, Anjana P, S Sameera, Ajith Sundaram Abstract - The environment-related worries of Generation Z together with sustainability-based activities established them as leaders who champion green consumerism. Digital natives of this generation opt to make buying choices on social media platforms according to their established reputation. The platforms of Instagram together with YouTube and LinkedIn function as essential spaces for spreading sustainability content which affects how people behave regarding their purchasing choices. Social media promotes consumer engagement through direct communication and enables fast information flow about green events so it stands as a key factor in developing positive green purchasing attitudes. Current research analyzes the impact of social media information sharing on Gen Z sustainable buying motivation through an investigation of green-value and subjective-norms as intervening variables. This research depends on the Stimulus-Organism-Response (SOR) model to see how social media leads consumers toward buying green products. This research study addresses the mental factors behind environmentally conscious buying to provide concrete recommendations for business organizations and government institutions. Companies can use the research results as a foundation to create better sustainability-oriented marketing plans that aim at Gen Z consumers.
Wednesday August 26, 2026 9:30am - 11:30am IST Virtual Room BGOA, India
Authors - Navneet S Patil, Shashidhar Kumbar, Sakshi Bhantanur, Arjav Jain, Satish Chikkamath, Sujata Kotabagi Abstract - Human movement prediction is a key machine learning domain whose purpose is to predict future movement from previous motion patterns and context information, with usage in autonomous vehicles, virtual reality, video games, and health care. In this study, the goal is to apply Convolutional Neural Networks (CNNs) for predicting human movement from the UCF50 dataset, whose collection contains action videos with a wide variety of actions. CNNs excel at discovering spatial and temporal patterns from video data and, thus, can be used in understanding motion complexities. In this work, a CNN-based approach is developed using a CNN architecture to assess motion dynamics and make accurate forecasts about future moves. By systematically preprocessing the dataset and optimizing the model’s architecture, the study achieved an accuracy of 99.09demonstrating the reliability and efficiency of CNNs in motion prediction tasks. Furthermore, the paper discusses existing methodologies in human motion prediction, comparing their performance and highlighting the advantages of CNNbased models in processing visual data. The results here bring out the potential of CNNs for real-world applications and set the foundation for future advancements in human activity recognition. The current study adds insight into machine learning methods and how they can be used to enhance motion prediction, with implications toward innovations in those fields that rely on precise modeling of human activities
Authors - V.Sudeep, V.Nishant, MM.Mohamed jasir Faiez, T.Monish, Yuvaraj kumar.GP, Akhil K J, Praveen.K Abstract - Docker containers are central to modern software development and deployment due to their portability, efficiency, and scalability. By isolating applications and dependencies, they provide a lightweight alternative to virtual machines, enabling consistent environments across platforms. However, Docker containers pose security challenges, including shared kernel risks, vulnerabilities in container images, and misconfigurations, which can lead to breaches.This paper examines security concerns in Docker containers and proposes a framework to identify and address vulnerabilities. The framework helps detect issues like outdated components and misconfigurations, offering insights to enhance security. Through practical use cases, it highlights its effectiveness in closing security gaps and equipping developers with tools to protect containers. The study emphasizes the need for proactive security measures and continuous vigilance in securing containerized systems.
Authors - Chaithra S Raju, Nimisha S, Arathi A N Abstract - The Digital payment landscape in India has seen rapid progress, driven by technological advancements, government initiatives and increased smartphone penetration. Digital wallets are becoming a payment method as they are convenient, secure and seamlessly integrate with financial services. Although Generation Z known for a Digital-first approach, inconsistency in the adoption of Digital wallets can be observed among this segment. This research will look at the reasons why Generation Z may adopt or not adopt Digital wallets, namely perceived ease of use, perceived usefulness and perceived security.A cross-sectional survey was undertaken for the 220 Gen Z respondents using a structured questionnaire. The statistical analysis was done by using SPSS analysis of variance to examine the impact of these factors on adoption behaviour. . The findings highlight that while convenience and utility drive adoption, security concerns remain a critical barrier.This study provides valuable insights for fintech companies, policymakers, and firms looking to bolster the digital payment infrastructure and build trust in Digital wallet services. Overcoming security concerns and improving the user experience can accelerate the transition towards a cashless economy.
Authors - S.Asha, Siddharth M Nair Abstract - According to a study, one out of every 20 people above the age of 65 are suffering from Alzheimer's. People with such neurological conditions have poor navigation skills and often wander around without having knowledge of where and what they are doing. In such situations, tracking them down is extremely important as it is life threatening to themselves and the people around them. It is also important to monitor elderly individuals' vitals like heart rate and steps along with detecting an impact (fall) so that necessary actions can be taken. Other than the strong personal motivation the current market needs a product through which people suffering from such neurological conditions can be supported. But not many are present in the current market and the ones that are, require the patient to wear some dedicated device like a neck ring or other uncomfortable devices. Often, people, especially elderly individuals lose their lives because 'it was too late'. There is a major requirement in today's market for a system which would send alerts and concerned individuals in case of any abnormality in detected data so that it would not be 'too late' to act. The sensors that are incorporated within the Apple Watch provide an ocean of valuable data which can be harnessed by caretakers and other concerned individuals. Now-a-days, people are too involved and busy with their work to stay at home and be there for elderly individuals at all times. Through this data, people can take care of their loved ones even when they are not around. By receiving timely notifications in case of any emergencies, the world would become a safer, more reliable place for all elderly individuals, especially those who suffer from Alzheimer’s and other neurological conditions.
Authors - Anitha D, Swetanshu Agrawal, Samudra Banerjee Abstract - Particularly affecting patient response to alkylating treatment, the methylation status of the O6-methylguanine-DNA methyltransferase (MGMT) promoter is a well-established prognostic and predictive biomarker in gliomas. Conventional evaluation techniques are prone to limits including sampling mistakes and intratumoral heterogeneity and call for invasive tissue biopsies. In this work, we present a non-invasive, deep learning-based system for multi-modal magnetic resonance imaging (MRI) based MGMT promoter methylation prediction. The method combines improved preprocessing, automated tumor segmentation, and a customized EfficientNet-based classification architecture with structural MRI sequences including T1-weighted, contrast-enhanced T1-weighted, T2-weighted, and FLAIR imaging. Our model achieves strong performance, high accuracy and generalizability in methylation status prediction. Comparative study including current literature shows either better or equivalent prediction performance, so highlighting the clinical possibilities of this technique. The suggested pipeline advances the function of virtual biopsy in neuro-oncology by providing a scalable, dependable, radiation-free substitute for MGMT methylation testing, therefore enabling individualized therapy planning.
Authors - Richa Goenka, Meenu Chawla, Namita Tiwari Abstract - In recent years, phishing attacks have emerged as a substantial hazard, endangering online businesses and security by exploiting users to divulge sensitive financial information through fraudulent websites. Despite various proposed methods, accurately distinguishing between legitimate and fraudulent sites in real-time remains challenging. This paper provides a new approach to identifying phishing URLs by employing a feature selection approach that integrates Genetic Algorithm and Particle Swarm optimization. This system optimizes feature selection through population initialisation, fitness evaluation, GA operations, and PSO integration, dynamically balancing exploration and exploitation. The objective is to identify significant features for supervised machine learning techniques, enabling precise phishing URL detection. For classification, multiple machine learning classifiers are employed among which XGBoost provided the best results. Experimental results using the hybrid feature selection prove that the machine learning classifier works much better than the prevailing feature selection approaches. This comprehensive approach provides a reliable method for detecting phishing URLs, improving internet security, and reducing the threats associated with phishing attacks.
Authors - Soumitra De, Jaydev Mishra Abstract - In this paper, a new method is focused to handle indeterminacy part of an imprecise data using neutrosophic set to generate proper constructive message. This method is capable to handle imprecise part of a neutrosophic data. Earlier no uncertain data set was handled this indeterminacy part of any uncertain data. We have drawn an output using this new method of any patient related data set that has suffering from disease. Vague logic is unable to process indeterminacy part. So only neutrosophic set is handled indeterminacy part of a imprecise data to outcome.
Authors - D.K. Chaturvedi, Nisha Verma Abstract - The technological intervention in our day-to-day life, impacted our social, physical, psychological and spiritual domains. The shoes are not untouchable from the latest innovations. The footwear is an essential wear in present time. The technology is completely changed the footwear industry and the customer flavour. Now the customer is looking for customized, smart footwear, which is environment friendly. The present footwear is using polymer soles (i.e. PVC, PU, EVA or Rubber), chemical based adhesives and animal leather upper material, which are not eco-friendly. A lot of research is going on to make sustainable and eco-friendly shoes with different biodegradable materials. The footwear industry is embracing both smartness and sustainability, blending technological innovation with eco-conscious practices. The smart footwear uses many types of sensors/IoTs to include different features of smartness. This paper discusses some innovations in footwear technology, important issues, challenges and their remedies related to design and development of smart sustainable footwear.
Authors - Kashinadh.S, Dhanush Devaraj, Yedukrishnan VS Abstract - Food safety and nutritional transparency are essential for public health, particularly as diet-related illnesses like obesity and diabetes rise. The Food Safety and Standards Authority of India (FSSAI) introduced a menu labeling policy in 2020, requiring restaurant chains to display calorie counts and nutritional information.The consumer awareness on the menu labelling is poor, and compliance is still low among restaurants. FSSAI Food Safety Connect app, which is designed to help with grievance redressal was having some negetive shades because of the complaints registered and reviews posted . This study employs stakeholder interviews, compliance audits, and sentiment analysis to evaluate how effective the policy is. The findings indicate that the compliance is lacking because of financial barriers and enforcement is also lacking. This study recommends implementing chatbot-driven grievance resolution, using QR codes for digital menus, and leveraging AI for compliance tracking as strategies to boost adherence. These solutions leads to the Sustainable Development Goals (SDGs 3 and 12) by enabling customers to make informed decisions while also promoting food safety. Menu labeling will become a more effective public health tool if we can improve digital enforcement in food industry.
Authors - Vanishree Pabalkar, Anuja Bokhare, Reena (Mahapatra) Lenka, Jaya Chitranshi Abstract - [1] Crime is one of the most worrying and widespread issues of our society. Criminal deterrence is essential for people safety. The overall crime inference when assessed, does help to keep a record of crime and assist in avoiding adversities. The aim of the study is to examine patterns in data acquired over time. Criminal violations offend humanity, and it should be prosecuted as soon as possible. Criminology is the scientific method of understanding crime and the motives behind the act. Criminology is an interdisciplinary area which gathers data and conducts further study into such offenses. While there is such a large amount of data on criminal activities, identifying and preventing crimes is one of the most difficult tasks. It is imperative to develop approaches and procedures for predicting future crimes and taking appropriate preventative steps. Cluster analysis includes breaking down huge data to minute groups with similar or identical characteristics. We can evaluate and assess methods, structures, layouts and interactions that are present in the data using visualization tools, so as to help uncover interesting areas and acceptable parameters for future analysis.
Authors - Akshar Sodankoor, Avanish Shenoy, B Monish Moger, Mohnish Gowda, Prafullata Kiran Auradkar, Subramaniam Kalambur Abstract - Virtualization is essential for efficient resource utilization in cloud and development environments. With the growing adoption of gRPC as a Remote Procedure Call (RPC) framework, evaluating its performance across different virtualization technologies has become crucial. This work benchmarks the performance of four gRPC call types: unary, client-streaming, server-streaming and bi-streaming, across four lightweight virtualization technologies. Docker, gVisor, Firecracker and nanos unikernel. The analysis examines CPU utilization, memory utilization and network capabilities to provide a comprehensive comparison. The results show that docker delivers the best performance across all metrics. Firecracker shows comparable latency performance to docker, but consumes higher memory. Nanos unikernel exhibits CPU utilization similar to that of docker, but has the highest latencies in all cases except unary gRPC call. gVisor exhibits the lowest CPU utilization under heavier workloads and also has the lowest latencies for client-streaming and server-streaming gRPC calls.
Wednesday August 26, 2026 9:30am - 11:30am IST Virtual Room DGOA, India
Authors - Aryan Goyat, Aditya Maan, Vimmi Malhotra Abstract - Deep learning has transformed artificial intelligence and enabled major breakthroughs in applications like computer vision, natural language processing, medicine, cybersecurity, and robotics. Through the use of deep neural networks, it enables automatic feature learning and exceeds machine learning-based methods in accuracy and flexibility. Challenges including excessive computational expense, uninterpretable nature, and ethics are still major hurdles to its widespread application. This article discusses the development and applications of deep learning and presents new research directions that seek to overcome its limitations. Federated and decentralized learning methods improve security and privacy by enabling collaborative model training without raw data sharing. Explainable AI (XAI) techniques, including SHAP and LIME, enhance the interpretability of deep learning models, making their decision-making more transparent. In addition, energy-efficient deep learning methods, such as model pruning, quantization, and neural architecture search (NAS), are being designed to minimize computational and environmental expenses. The emergence of self supervised learning further minimizes dependence on labeled data, making deep learning more feasible across domains. Future developments will center on the fusion of deep learning with reinforcement learning, symbolic AI, and evolutionary algorithms to build more generalizable and efficient systems. These technologies will power the next wave of intelligent, ethical, and sustainable AI solutions.
Wednesday August 26, 2026 9:30am - 11:30am IST Virtual Room DGOA, India
Authors - Anita Agrawal, Ruhi Panjwani, Aditya Mallik Abstract - This paper explores the design and implementation of a multi-user, multi-access web application tailored specifically for automated weather stations (AWS). By examining real-world scenarios and user interactions, we identify key design considerations, including system performance, security, and scalability. The study aims to provide practical insights for developers to create efficient and user-friendly web applications that effectively handle large-scale weather data and cater to various user access levels.
Authors - Krishna Shirsath, Abdullah Ansari, Riyaz Memon, Phiroj Shaikh Abstract - Real-time collaboration is essential for modern software development, enabling developers to work together seamlessly from different locations. This paper presents DevTogether, a collaborative coding platform that facilitates efficient teamwork through live code editing, customizable collaboration sessions, integrated chat, a collaborative drawing board for design prototyping, and real-time video meetings powered by WebRTC. The platform incorporates an intelligent code assistant using the Gemini API, along with an autosave mechanism to ensure workflow continuity. Built using the MERN stack, DevTogether emphasizes scalability, low latency, and responsive performance while addressing synchronization conflicts and communication challenges. Experimental evaluations and simulated performance metrics underscore its effectiveness compared to similar platforms.
Authors - Arnav Shukla, Subhashree Choudhury, Logeshwaran R. Abstract - One of the key limitations of any crowdfunding platform that uses blockchain technology is the lack of transparency regarding fund usage by campaign creators after receiving donations. The current paper addresses this issue. For this investigation, we conducted a detailed examination of the relevant literature on decentralized applications (DApps) and the use of smart contracts to automate administrative tasks. We identified a significant gap in existing platforms: the ambiguity surrounding post-donation and how fund management can be unfair, which blockchain alone does not address effectively. Our findings highlight the strengths of blockchain security and automation capabilities, which ensure a safe and trustworthy environment for users. Our proposed solution integrates a decentralized voting model and a collusion prevention algorithm called EigenTrust to empower donors with a participatory role in decision-making processes, eliminating the need for centralized authority. In this study, we implement and evaluate a decentralized voting model that uses specific techniques to prevent any attacks or chances of misuse of powers that would then empower donors to participate in critical campaign decisions, enhancing trust and satisfaction by allowing them to verify the responsible use of their contributions. By reducing uncertainty around fund allocation, our model increases a more engaging and end-to-end secured donor experience, encouraging donations and supporting the long-term success of social crowdfunding projects. In all, this paper presents a novel approach to increase crowdfunding platforms using blockchain technology with a voting model paired with collusion prevention to address the issue.
Authors - P S Sree Harsha, Harshitha S, Ayush Sisodiya, M P Deepti, Sarasvathi V Abstract - Concerning the Conventional recruitment methods, which has many challenges such as the poor matching of the candidates to the right requirements, issues of transparency and issues of privacy in dealing with sensitive information. This leads to the hiring of candidates who are not qualified to meet their requirements and compromise the organizational data. To solve these problems, Human Resource Blockchain Intelligence Recommendation System (HRBIRS) is proposed to use Blockchain and Recommendation system technologies for getting an efficient recruitment process. Blockchain is evolving technology providing transparent and secure data. Privacy issues are resolved by decentralized architecture. The Hybrid recommendation system makes recruitment effective by determining the qualifications, skills and experience of the job seekers relevant to certain job recruitments posted by the HR. One of the most attractive features of HRBIRS is the peer approval and endorsement systems within the organization without bias to a particular candidate. This paper provides detailed information on the architectural design, implementation and its performance features and demonstrates how this system could be beneficial for recruitment processes throughout the organization by providing data security and enhancing the effectiveness of selecting the right candidate.
Authors - Arunangshu Giri, Dipanwita Chakrabarty, Manash Routray Abstract - The present study emphasized on how enrollment of loyalty programs at fuel retail outlets get enhanced through digital communication and intelligent promotional strategies. The study has evaluated the three major dimensions like participation intention and promotional efficiency to understand the efficacy of the loyalty programs organized by different fuel retail outlets. 454 Indian customers were interviewed over a three months period using a structured questionnaire. Cross-sectional descriptive research design was followed for the study. Both qualitative and quantitative analysis was done using NVivo and SPSS-28 software. The study revealed that customer participation in loyalty programs was highly influenced by flexible redemption options, digital reward system and exclusive benefits. Again, the study has shown how staff knowledge, promotional policies, customer loyalty and digital engagement influence promotional effectiveness. The study acknowledged the pivotal role of intelligent communication strategies in enrichment of customer adaptability towards digitized loyalty programs. Establishing a seamless communication between the customers and digital platforms along with effective staff training can be prudential for optimal customer engagement, sustainability and loyalty.
Authors - Lalithya Govardhan, Shalini M S, Gagan Deep P S Abstract - This work shows an extensive multimodal system of mood detection and customized playlist recommendation based on EEG, GSR, and face recognition. Brainwave activity for emotional evaluation is sensed by EEG electrodes, while GSR sensors provide skin conductance, heart rate variability, and temperature values for physiological behavior. Facial behavior with emotional facial expressions is determined through facial landmarks. Preprocessing entails Butterworth filters for EEG frequency bands, GSR data normalization, and facial feature extraction from a pretrained model (FER) to track eyebrow position, mouth curvature, and eye openness. EEG features are examined using frequency domain analysis, whereas GSR and facial features are classified using Random Forest. To increase precision, a fusion model aggregates predictions by weighted averaging or majority voting, with EEG assigned the greatest weight due to its high correlation with mood. After determining the emotional state, a suitable playlist is suggested: energetic songs for happiness, relaxing music for stress, comforting songs for sadness, and relaxing music for relaxation. This feature-based recommendation system enhances personalization through the use of features like tempo, genre, and mood to provide a dynamic and interactive listening experience for the user
Authors - A. Revathi, A. Sunidhar Reddy, Geetika Alapati, R. Pranay Abstract - This paper presents the performance of the grocery identification system concerning Telugu grocery items, considering both native and non-native speakers. Speech recognition for Telugu groceries presents a unique challenge due to variations in pronunciation, accent, and noise conditions. This study explores the implementation of a Gaussian mixture model (GMM) classifier in conjunction with rasta-perceptual linear prediction (RASTA-PLP) features to enhance the accuracy of Telugu grocery identification. Rasta-PLP effectively captures robust speech features by suppressing unwanted spectral variations, while GMM provides a probabilistic framework for classification. The proposed system is trained on a dataset comprising commonly used Telugu grocery names and evaluated under diverse acoustic environments. Experimental results demonstrate improved recognition performance, showcasing the effectiveness of RASTA-PLP in feature extraction and GMM in classification. This work contributes to developing efficient speech-based interfaces for regional language applications, facilitating voice-driven grocery identification systems. The recognition accuracy of the proposed system is approximately 99%, ensuring high reliability in real-world applications. This technology benefits society by aiding visually impaired individuals and non-Telugu speakers in grocery identification, enhancing accessibility and convenience. By enabling seamless voice-based interaction, promotes inclusivity and improves social equity through technological advancement.
Authors - A.Revathi, Reethikaa Vallinayagam, S. Sivaranjani, A. Deepthi Abstract - This research work introduces a system for identifying genuine speech and recorded (replay) speech through Mel-Frequency Cepstral Coefficients extraction and uses k-means clustering for classification purposes. Speech features obtained from various speakers undergo normalization procedures before receiving cluster assignment during training sessions. During the testing phase speaker identification depends on measuring the distance between input features against cluster centroids. The confusion matrix indicates system performance by showing correct genuine speech detection through high diagonal values yet exhibiting lower off-diagonal values to indicate possible attacks based on recorded speech. Auto-correlation together with cross-correlation serve to evaluate the similarities between speakers. Strong recognition of the same speaker is indicated by high auto-correlation values but weak cross-correlation values demonstrate effective differentiation between different speakers. The AVSpoof dataset serves as the experimental foundation because it includes ten recording subjects who are distributed between five male and five female speakers. The acceptance and accuracy evaluation for the system happens through testing samples which proves its ability to recognize genuine speech from recordings as well as identify distinct speakers properly.
Authors - K Shailaja, Shirina Samreen Abstract - Diabetes mellitus is becoming an increasingly critical health issue across globe that demands early diagnosis and continuous monitoring. Although traditional machine learning models have been employed to predict diabetes, they frequently lack clarity and have difficulty in identifying uncommon or rare patient cases. This research proposes a novel explainable hybrid framework that combines deep learning-based temporal modeling with classical machine learning and unsupervised anomaly detection. The framework utilizes multimodal data sources, including static clinical features, time-series Electronic Health Records (HER) and wearable sensor data, for robust diabetic risk assessment. Explainability is achieved using SHAP (SHapley Additive exPlanations) and counterfactual reasoning to provide both global and local interpretability. In addition, an autoencoder-based novelty detection module identifies patients whose health patterns deviate significantly from the normal. Experimental results on benchmark datasets demonstrate improved prediction accuracy, better anomaly identification and enhanced interpretability making the model suitable for real-world clinical decision support systems.
Authors - Vishal Ambhore, Ketki Kshirsagar, Parikshit Mahalle Abstract - This survey paper thoroughly examines the landscape of access control within the context of the Internet of Things (IoT). With the rapid expansion of IoT technologies, ensuring secure access to resources and data has become a critical concern. We thus present a very wide-ranging review of existing literature, doing an all-round analysis of current access control mechanisms in order to discuss their strengths, weaknesses, and applicability to IoT environments. The result of our investigation shows few major gaps and challenges, such as issues of scalability, interoperability concerns, and a call for access policies that account for contexts. Taking this into account, we introduce new approaches and improvements to positively address these shortcomings and boost the security and efficiency of access control in IoT systems. By integrating findings from multi study research endeavors, we provide new points of view and approaches toward IoT security improvement. Furthermore, we present potential future research directions and challenges to guide development for more resilient and adaptive access control solutions in IoT ecosystems. The paper is a ready reference for researchers, practitioners, and policymakers who want to bolster the security posture of IoT deployments and mitigate newly emerging cyber threats effectively.
Authors - Garima Ahuja, Heena Hooda, Vimmi Malhotra Abstract - The combination of agriculture and Information and Communication Technologies (ICT), known as e- Agriculture, is changing farming practices, especially in regions facing limited resources and climate challenges. e-Agriculture includes digital tools such as mobile-based advisory platforms, Geographic Information Systems (GIS), Internet of Things (IoT), Artificial Intelligence (AI), and data analytics. These technologies provide timely information, improve the use of inputs, and increase access to markets and financial services [1]. Applications of ICT in farming across different regions show improvements in crop planning, yield monitoring, and risk reduction. Evidence from India, Kenya, and the Netherlands highlights positive results where digital tools are adapted to local needs. Improvements include better harvest management, lower post-harvest losses, and increased climate resilience [3]. Despite these benefits, challenges such as poor rural connectivity, low digital skills, and limited policy support continue to affect the full use of such technologies. Scaling e-Agriculture requires affordable access, local training, and supportive ecosystems. Broader adoption can support sustainable farming systems and contribute to goals such as food security, poverty reduction, and environmental protection.
Authors - Khloe Bhel, Krutthika Hirebasur Krishnappa Abstract - Healthcare data growth at an exponential rate together with rising cyber threats require sophisticated cryptographic systems to protect Electronic Health Records (EHRs) from unauthorized access and data tampering. The research develops a cryptographic system which uses simulated quantum key distribution through the E91 protocol to generate symmetric encryption keys that is encrypted using Advanced Encryption Standard (AES) in Cipher Block Chaining (CBC) mode to protect healthcare data. The system operates by using IBM Qiskit’s AerSimulator backend to create entangled qubit pairs for deriving a quantum-safe key between two communicating parties. The entropy measurement of the generated key approaches the maximum value of randomness which provides effective protection against brute-force attacks. The AES encryption process using achieves rate of approximately 6.58 MB/sec during encryption operations and 9.34 MB/sec during decryption operations. The proposed method demonstrates efficient computation and deployment potential for healthcare applications with limited resources including edge-based IoT medical devices and federated learning systems. Security analyses show that this approach gives a good protection against both classical attackers and near-term quantum attackers. The research shows how post-quantum cryptographic methods can be practically used to protect future healthcare systems.
Authors - Suprit V. Hatti, P. G. Sunitha Hiremath, Manohar Madgi, Neha Tarannum Pendari Abstract - This study presents a comprehensive analysis of the factors influencing cognitive decline in Alzheimer’s patients in the USA. The objective is to examine cognitive decline and dementia prevalence across various U.S. regions, focusing on gender, age, and race/ethnicity disparities. Using data from the Behavioral Risk Factor Surveillance System (BRFSS), collected from 2015 to 2021, 59 locations were categorized into Northeast, Midwest, Southeast, Southwest, and West regions. The original dataset contained 31 attributes. The analysis included year-wise trend examination, gender-wise, age-wise, and race-wise distribution assessments, and identification of key risk factors. The Midwest (21.98%) and West (21.83%) showed higher cognitive decline rates, with significant disparities across demographics. The ‘Overall’ age group showed the highest prevalence at 8.54%. Females were most affected, with diabetes, asthma, arthritis, depression, and cardiovascular diseases being the most correlated comorbidities.
Authors - Reshma Y Totare, Anushka Kurandale, Sakshi Kuyte, Kiran Mane, Snehal Nale Abstract - With the increasing reliance on online reviews across various digital platforms, user feedback has become a vital factor in influencing public perception and decision-making. Since users cannot physically verify products or services online, they often depend on reviews to assess quality and credibility. This dependency has led to a rise in deceptive practices, where fake reviews are used to mislead audiences—either by promoting certain offerings or undermining competitors. Detecting such fraudulent content presents a significant challenge in the field of natural language processing (NLP), due to the subtle and human-like nature of these reviews. In this project, we present an approach for fake review detection using a deep learning model that combines Long Short-Term Memory (LSTM) networks with Bidirectional Encoder Representations from Transformers (BERT). Our model utilizes LSTM’s ability to capture long-range dependencies along with BERT’s contextual language understanding to enhance detection accuracy. To improve practicality and trustworthiness, we incorporate several additional features plugin support for easy integration into various review-based platforms, multilingual capability to handle reviews in different languages, and LIME (Local Interpretable Model-agnostic Explanations) to provide word-level interpretability of predictions. We evaluate our model on publicly available datasets containing both real and fake reviews, and the results demonstrate that our LSTM-BERT approach significantly outperforms traditional machine learning techniques. This work contributes to the growing efforts in combating misinformation and enhancing the credibility of online content across diverse platforms.
Wednesday August 26, 2026 9:30am - 11:30am IST Virtual Room EGOA, India
Authors - Ajay Talele, Siddhi Shingate, Om Gaikwad, Mahek Sayyad, Sai More, Pooja Nanaware, Ashwini Borole, Pratiksha Bile, Bharat Bangar Abstract - Crime prevention is being transformed by the Internet of Things (IoT) through proactive, smart solutions. Abstract This paper presents some techniques at the time of crime prevention by the use of IoT. Through the use of IoT driven surveillance systems, integrated smart sensors, real-time data aggregation and analytics, and automated alerts to emergency services and law enforcement, crime prevention, identification, and response are improved significantly. AI in Smart City: The study also calls out real-time examples such as smart cities establishing AI powered CCTV, IoT-enabled access control, predictive policing[1] The contemporary world faces the overwhelming challenge of crime. Criminal activity exists in almost every country and the statistics for certain nations are quite alarming. The advancement of technology has been one of the most significant factors in the development of crime control and prevention measures, such as drone surveillance, GPS tracking and tagging, closed circuit television cameras, etc. Technological advances such as the IoT (Internet of Things), Machine Learning, and Edge Computing call for the attention of the scientific world in regards to the question: how can they utilize these innovations for the purpose of minimizing criminal activities globally? In conclusion, the research results reflect on the employability of AI and IoT technologies in reinforcing security and obstructing offences against state and equally, challenges and ethical issues that accompany their employments [2]
Authors - Sambhram Pattanayak, Archana Paswan Abstract - The evolving landscape of dialogue replacement in film and television production is examined, focusing on advancements and challenges in on-location Automated Dialogue Replacement (ADR) recording and the emergence of real-time dialogue replacement technologies. Advancements in portable recording equipment, microphone technology optimized for field use, and specialized software solutions have significantly enhanced the feasibility and quality of ADR conducted outside traditional studio settings. However, on-location ADR presents unique challenges related to acoustic control, environmental noise, actor availability, logistical complexities, and technical limitations. The paper also examines the current state and potential impact of real-time dialogue replacement technologies, which offer immediate solutions and increased flexibility on set and in post-production. Case studies illustrate the practical applications of on-location ADR, while the discussion of future trends highlights the anticipated integration of artificial intelligence, advancements in hardware and software, the influence of remote collaboration tools, and the potential for incorporating virtual and augmented reality into dialogue replacement workflows. This analysis underscores the ongoing evolution of audio post-production techniques aimed at enhancing efficiency and creative possibilities in modern filmmaking. The research findings presented in this study underscore the importance of ADR in film production and introduce novel approaches and technologies. These advancements provide filmmakers with state-of-the-art tools and techniques, revolutionizing their creative processes and enhancing the quality of their productions.
Authors - Mario Castro Romero, Carlos Ernesto Carrillo Arellano, Leonardo Daniel Sánchez Martínez Abstract - Software-defined wireless sensor networks are emerging as a transformative technology for industrial, research, and IoT applications leveraging their control-data plane separation. This paper analyzes routing protocols, emphasizing their advantages, limitations, and QoS-impacting factors (e.g., latency, bandwidth, among others). Through a systematic review, we identify innovative solutions to enhance QoS in SDWSNs, addressing challenges like energy efficiency and dynamic adaptability. Previous results demonstrate that centralized routing and machine learning-based algorithms significantly improve reliability in critical applications.
Authors - Manpreet Kour, Naman Jain, Neeraj Gupta, Geetanjali Bhola Abstract - Waste recycling is important both in the global economy and in the global climate as a whole. As a result, classification of recyclable waste has become a critical goal for humanity, and deep learning models have important potential to fulfill this task. In this study, six advanced folding network models of neural networks - EfficientNetV2L, EfficientNetB1, EfficientNetB0, MobileNetV2, ResNet50, and VGG16, were compared for the effects of the garbage classification task. The results show that EfficienctNetB0 gave better performance than the other models. Furthermore, data augmentation techniques were used to improve classification accuracy, as data records contained a limited number of samples. Notably, MobileNetV2 not only achieved competitive accuracy, but also became a green choice for its low carbon emissions.
Authors - Sanchit Prashant Joshi, Parth Atul Gargate, Yash Prabhakar Apotikar, Rupesh C Jaiswal, Mousami V. Munot Abstract - Social media platforms operate at top speeds when transferring image-based data. The shared and posted images and videos on WhatsApp and Instagram consume the majority of network resources. Lossless compression techniques were applied to images while maintaining image quality throughout data storage and transmission processes because this fundamental method produces perfect information reconstruction after decompression. The research evaluates Predictive Coding (DPCM) and Context-Based Coding and Arithmetic Coding and Dictionary-Based Techniques (LZW) and Block-Based Compression through analyses of their efficiency metrics and computational complexity and practical usage. New developments in JPEG2000 and LZW compression have led to increase speed and efficiency through Parallel Symbol Encoding in Arithmetic Coding and Compression Ratio Prediction. The Optimized Run-Length Encoding (ORLE) system uses dynamic compression approach adaptation according to image orientation to enhance its flexibility. The speed of real-time applications increases remarkably when using FPGA implementations. This survey examines trade-offs among compression ratio together with computational expense and suitable data sets to perform an evaluation between classical and modern methods. Future development in lossless data and image compression relies on emerging trends such as AI-driven compression models as well as hardware-accelerated algorithms and hybrid frameworksDeflate is the fastest compression technique taking about 0.043 seconds, with a Maximum compression ratio of 26.66 given by WebP.
Authors - Anuj Sudhir Kulkarni, Rama Gaikwad, Prathamesh Zad, Sai Lahane, Shivam Shelke, Saurav Jadhav Abstract - The rapid development of artificial intelligence (AI) is changing the financial landscape. It offers innovative solutions to optimize personal financial management and advisory services. This research focuses on developing an AI-based platform to improve financial decision-making by analyzing users' investments to provide insights into financial health. Key features include Portfolio Visualizer, Risk Radar, Fundamental Analyst, Price Forecaster and Financial advisory services ensure a comprehensive view of financial planning, emphasizing AI frameworks and interpretable applications. To build user trust and transparency, challenges such as mitigating bias are explored. Real-time problem solving and fine-grained scalability with a commitment to accessibility and precision This research highlights the ability of AI to democratize financial advisory services. and overcome limitations in the current system. Future directions include real-time risk assessment. Advanced portfolio management and innovative AI integration for dynamic market simulation.
Wednesday August 26, 2026 12:30pm - 2:30pm IST Virtual Room AGOA, India
Authors - Mohan S G, Abhilash K Raj, Nayana S A, Pradhaan S, Rajendra Bhat Abstract - This paper explores the application of the You Only Look Once (YOLO) v11 model for real-time object detection in Indian road conditions, addressing challenges posed by unconventional objects like animals, autorickshaws, carts, and tractors. A dataset from dashcam and mobile footage was annotated using the Computer Vision Annotation Tool (CVAT) tool and combined with COCO to train YOLO v11. The model significantly improved detection accuracy, increasing classes from 30 to 108. Its high accuracy and real-time performance make it suitable for autonomous vehicles and traffic monitoring in India.
Authors - Vani E S, Gourav Subnani, Prajwal Gupta, Shivee Jaiswal, Mihir Sahu Abstract - Plant species classification accuracy is crucial for biodiversity conservation and ecosystem monitoring. Traditional taxonomy-based methods, which rely heavily on expert analysis, can be inefficient and prone to errors, particularly when processing large datasets. This study leverages deep learning and machine learning techniques to automate plant species identification, with a strong focus on leaf vein morphology analysis. The proposed approach begins with preprocessing leaf images by converting them to grayscale, extracting significant structural features, and skeletonizing vein patterns. Key morphological characteristics, including vein distributions, textures, and geometric attributes, are then used as input for classification models. They use both sophisticated deep learning models like Convolutional Neural Networks (CNN) and more traditional machine learning approaches like Random Forest (RF), k-Nearest Neighbours (kNN), and Support Vector Machines (SVM). The Xception architecture, known for its depth wise separable convolutions, is particularly effective in capturing intricate vein structures, enhancing classification accuracy. This automated system reduces the dependency on manual identification efforts, making it scalable for large-scale biodiversity research. By integrating deep learning-driven analysis, the proposed framework provides a robust and efficient solution for plant species classification, aiding conservation initiatives and ecological studies.
Authors - Sangita Lade, Muhammad Parkar, Shreyas Nagarkar, Om Shintre, Shivam Padalkar Abstract - The rapid evolution of machine learning (ML) has transformed industries by enabling automation, prediction, and optimization for complex real-world problems. However, developing ML pipelines involves repetitive tasks such as data preparation, model building, and evaluation, which are time-consuming and prone to errors. This paper introduces an automated system for generating ML code using Jinja2 templating and supervised MLbased feature prediction. The system analyzes 5000 ML code templates to extract parameters like data type, preprocessing techniques, model architecture, and hyperparameters. A supervised ML model predicts missing parameters based on partial user input, enabling dynamic code generation. The framework supports diverse data formats (tabular, image, text) and ML tasks (classification, regression). Experimental results demonstrate high accuracy in parameter prediction and significant time savings (70-80% reduction in setup time). The system simplifies ML development, reduces errors, and accelerates experimentation, making it accessible to researchers, developers, and students.
Authors - Rohini T.V, Srikrishna Adiga G, Tejas C, Sunil Mashyale, Sunil Kumar C Abstract - DevLaunch is a cloud-native deployment platform purpose-built for MERN stack apps, using AWS Amplify to make hosting and configuration easy. The platform provides real-time monitoring, auto-resource provisioning, and a CDN-tuned Next.js frontend to abstract away deployment nuances. In addition, DevLaunch increases developer efficiency by reducing the need for manual setup and cutting deployment and debug times by 40% and 30%, respectively. The auto-scaling architecture and simplicity of the system make it a secure and highly scalable way to deploy contemporary web applications.
Authors - Namrata Jangam, Nipun Jadhav, Riya Chavan, Priya Chavan, Rutuja Surve Abstract - Time-honoured treatment has long relied on pharmaceutical plants as genuine remedies due to their bioactive compounds. With increasing demand for natural products and sustainable healthcare, accurately identifying and classifying these plants is crucial. However, distinguishing species is challenging due to similar physical traits and varying environmental conditions. Machine learning (ML) and deep learning (DL) have shown substantial ability in medicinal plant detection and classification by analysing large datasets and extracting subtle features. Image recognition techniques, particularly convolutional neural networks (CNNs), can identify morphological traits like leaf size, shape, and texture for classification. Studies have demonstrated that CNN models can achieve up to 90% accuracy in medicinal plant identification, enhancing the process for novel drug discovery and therapeutic applications.
Authors - Ananya Kini, Saranya Rubini Abstract - In recent times, there have been several advancements in computer vision and image processing, and when combined with machine learning models, is very helpful in posture recognition applications. Posture detection is a useful tool in sports and fitness, as it helps people avoid injuries caused by poor alignment and achieve optimal posture in order to stay healthy. This paper reports on ”PoseNet : A Novel YOLODriven Framework for Badminton Posture Detection and Correction”, which is a Python-based application that utilizes Roboflow for dataset construction, annotation and augmentation, YOLOv5 for custom training the model on the dataset, and MediaPipe for giving corrective suggestions to the user. The novelty of this framework lies in its dual-stage architecture, combining YOLOv5 for classification and MediaPipe for real-time correction, specifically tailored for badminton. Additionally, it leverages a badminton-specific dataset, ensuring domain relevance and precise analysis. It predicts the stance that the player is planning to achieve and then tells whether the stance is correct based on their key points. The model obtained a high classification accuracy, with mAP50 value of 96.2% and mAP50-95 value of 81.1%.
Wednesday August 26, 2026 12:30pm - 2:30pm IST Virtual Room AGOA, India
Authors - K. V. Deshpande, Sanskruti Parkhe, Varad Pawar, Vaishnavi Thorat, Rutuja Bagad, Priti R. Kale Abstract - In today's digital world, cyberattacks targeting critical infrastructure pose a significant threat to government agencies and organizations. These attacks can disrupt essential services and compromise national security, making it crucial to identify and respond to them quickly. This survey paper discusses the challenges faced in monitoring cyber threats and presents a proposed solution: a real-time cyberattack monitoring tool. This tool uses machine learning and web scraping to gather data from various online sources, storing it in a structured format for easy access. By visualizing the collected data through an interactive dashboard, cybersecurity teams can quickly identify and understand the nature of ongoing attacks. Additionally, the system includes an alert mechanism that notifies teams of high-frequency attack patterns, enabling prompt action. Overall, this solution aims to enhance the ability of organizations to protect their critical infrastructure by providing timely insights and effective incident response strategies.
Authors - Garima Ratra, Akriti Kumari, Vimmi Malhotra Abstract - Blockchain has been widely adopted across numerous industries and applications to improve privacy and security factors. However, with the rapid expansion of this technology, its significant energy consumption has become a growing concern, particularly in mobile cryptographic applications. Traditional consensus mechanisms, such as Proof-of-Work (PoW), require substantial computational power, making them unsuitable for mobile environments. This paper reviews innovative blockchain protocols that prioritize energy efficiency while ensuring security and decentralization. By analyzing alternative consensus mechanisms including Proof-of-Stake (PoS), Delegated Proof-of-Stake (DPoS) and energy optimization strategies, we assess their effectiveness in lowering power consumption. The study highlights the role of sustainable blockchain approaches in enhancing mobile application efficiency with minimized environmental impact. This will help in increasing the energy efficiency and understanding the impact and applicability of blockchain by switching to greener systems.
Authors - Jishnu Prakash k, Rithish S V, Liz Maria Liyons, Vijval Srinivasan, Deepthi L R Abstract - The rapid spread of misinformation and rumors on social media platforms, particularly Twitter, poses significant risks to public perception and decision-making. This study presents a comprehensive approach to analyzing and mitigating rumor propagation by identifying key influencers and optimizing propagation time within online communities. Our dataset consists of over 800,000 nodes with interactions categorized as retweets, mentions, and replies, each assigned an influence score to quantify user impact. Using the Infomap algorithm, we initially detected 13,500 communities and filtered them to retain 67 influential clusters with a higher number of nodes and stronger influence scores. To analyze the spread of rumors, we developed an algorithm that tracks propagation within these communities, leveraging top influencers as initial spreaders and computing the average propagation time. Furthermore, we introduced a node deletion strategy to iteratively remove high-impact influencers, reducing the overall propagation time and limiting misinformation spread. Finally, we adjusted the propagation times by normalizing them with the earliest influencer timestamps to ensure precise measurement. Our findings highlight that targeted removal of key spreaders significantly disrupts rumor diffusion, providing insights into optimizing influence-based network interventions for misinformation control.
Authors - Yukta, Nishant Yadav, Sukrati Chaturvedi Abstract - The rapid expansion of social media platforms and online news consumption has led to an increased spread of fake news, posing significant challenges to society by influencing public decision- making and causing severe consequences in domains such as politics and healthcare. Traditional methods for identifying fake news are often too slow to combat its swift dissemination. Therefore, identifying and addressing misinformation is crucial for maintaining the accuracy and trustworthiness of information disseminated on social media. Natural language processing (NLP) and artificial intelligence (AI) play a vital role in this endeavor by facilitating the effective examination of extensive datasets to uncover patterns that are suggestive of misinformation. Machine learning models, particularly transformer-based architectures, enhance fake news detection by improving accuracy and interpretability. The integration of Explainable AI (XAI) methods further enhances trasparency and trust in these models
Authors - Praful Sambhare, Nitin Choudhary, Abhay Rahangdale, Atharva Rane, Sahil Raina Abstract - The integration of artificial intelligence (AI) within the fashion industry is becoming increasingly prevalent, with the aim of delivering a shopping experience that is personalized, seamless, and engaging. As an example of this trend, GlamBot is an AI- driven fashion assistant, a sophisticated fashion assistant that employs a range of advanced AI methodologies. These methodologies include Natural Language Processing (NLP), image- based similarity searches, and voice recognition technologies, all of which together transform the interactions that users have with fashion platforms. The functionality of GlamBot significantly improves the user experience by providing tailored fashion recommendations. This is achieved through the analysis of text inputs, the execution of visual similarity searches, and the processing of voice commands. Consequently, fashion discovery is rendered more intuitive and accessible for users, thereby facilitating a more engaging interaction with the fashion domain. . By analyzing and understanding user preferences, GlamBot builds personalized profiles that evolve over time to deliver increasingly accurate recommendations . GlamBot’s image-based search feature allows users to upload pictures of fashion items they like. Using advanced ResNet50 image recognition models, GlamBot analyzes these images and provides visually similar product recommendations, bridging the gap between users’ visual preferences and available fashion products .
Authors - Merlin Priya Jacob, Sukhada Aloni, Hetal Rawat, Shrishti Sakore, Shreya Naik, Shubham Pardhi Abstract - Using Ethereum, IPFS, and the MERN stack, this project offers a transparent and safe blockchain-based examination system. The solution guarantees tamper-proof question paper management by utilising IPFS (via Pinata) for decentralised storage and Ethereum smart contracts. Instructors submit tests to IPFS, ensuring data integrity by storing their cryptographic hashes on the Ethereum blockchain. While MetaMask allows for safe user interaction and authentication, Hardhat makes it easier to design and deploy smart contracts on a testnet. A strong online application is powered by the MERN stack, with Node.js managing database functions and instructor authentication. Exam papers are safely retrieved by authorised superintendents, guaranteeing regulated access. Academic assessment integrity is improved by this decentralised method, which reduces the possibility of paper leaks and unauthorised changes while offering a scalable and effective substitute for conventional test systems.
Authors - Sarbjit Kaur, Jasmeen Gill Abstract - Wireless sensor networks (WSNs) play a vital role in sensing environmental conditions in far-flung areas. However, their energy consumption remains a critical issue, affecting the network's lifetime and coverage area. Clustering has emerged as an efficient strategy to prolong sensor network lifespan, and the Fruit Fly Algorithm (FFA) and Ant Colony Optimization (ACO) are promising techniques for cluster formation and efficient path establishment, respectively. In this study, we propose an innovative approach that combines FFA for cluster formation and ACO for path establishment. This novel algorithm is implemented in MATLAB and evaluated in both homogeneous and heterogeneous environments. We compare our proposed algorithm with the Biogeography-Based Optimization Algorithm (BOA) and the Low Energy Adaptive Clustering Hierarchy (LEACH) algorithm. Our results indicate that the proposed algorithm significantly outperforms both BOA and LEACH in terms of network lifetime and coverage area, particularly in heterogeneous environments.
Authors - M Jayaram, Kodari Madhavi, Amboth Anil Kumar, Pachipala Naveen, Gajula Rithvik Abstract - Human-Computer Interaction prioritizes universally accessible systems, crucial for individuals with physical disabilities. This study introduces an Eyes as Interfaces: A Novel Eye-Tracking Mouse Cursor System, a hands-free solution enabling seamless digital environment interaction. For people those with paralysis, muscular dystrophy, or spinal injuries, this technology provides independent computing access, eliminating dependency on external assistance. A Convolutional Neural Network (CNN) is the system's backbone that provides real-time pupil detection, mapping eye gaze to exact cursor movement. Advanced image processing like this guarantees smooth operation regardless of changing lighting and user conditions. By accurately mapping eye movements to cursor actions, users can navigate and communicate with computer interfaces, opening avenues for information access, communication, and work participation. This encourages independence and enables users to access the web on their own, performing tasks like document creation and web navigation. Beyond personal benefits, this technology promotes inclusivity by bridging the digital divide, allowing for real-time, unrestricted participation in learning, employment, and social activities. Its smooth integration in widespread digital platforms ensures that it carries the highest level of potential in changing lives of people with mobility disabilities Worldwide.
Authors - Kailash Agarwal, Parikshit N. Mahalle, Bhagwan D. Thorat Abstract - Heart failure is a serious medical condition that affects millions worldwide, and early prediction is essential for timely intervention and better patient outcomes. While machine learning models have demonstrated strong predictive capabilities in healthcare, many high-performing models, such as Support Vector Machines (SVM), function as black boxes, making them difficult to interpret in clinical settings. This study examines how Explainable AI (XAI) techniques can enhance transparency in heart failure prediction.Using a publicly available dataset from Kaggle, we preprocess the data with label encoding, feature scaling and Hyperparameter tuning before training various machine learning models for binary classification. Our results indicate that the Support Vector Classifier (SVC) with a Linear kernel achieves the highest predictive accuracy. However, to improve interpretability, we compare its performance with explainable models like Decision Trees and apply post-hoc explanation techniques such as SHAP (SHapley Additive Explanations) and Permutation Importance.Through this comparative analysis, we highlight the trade-off between model accuracy and interpretability, offering insights into the feasibility of XAI-driven models in real-world clinical decision-making. Our findings reinforce the importance of developing AI systems that not only perform well but also provide understandable and trustworthy insights for medical professionals.
Authors - Bhagwan Thorat, Omkar More, Prathamesh Medage, Aditi Mali, Janhavi Maske, Sanika Maind Abstract - Traditional healthcare systems have primarily focused on physical health, often overlooking mental well-being. With the rapid advancement of technology, integrating AI-driven emotion detection into mental health management can offer valuable insights. This paper presents a comprehensive mental health management system that utilizes Haar Cascade classifiers and a Keras deep learning model for real-time emotion recognition via OpenCV. A Flask-based web interface, built using HTML, CSS, and Python, enables users to monitor their emotional states and facilitates therapist booking and automated receipt generation. By leveraging facial expression analysis, the system provides a data-driven approach to mental health assessment, enabling early intervention. The platform also ensures accessibility and efficiency, reducing the burden on healthcare providers. Experimental evaluations demonstrate the system’s effectiveness in accurately detecting emotions and its potential in AI-assisted psychological support. Future enhancements will focus on multi-modal emotion detection, incorporating natural language processing (NLP) and IoT-based physiological monitoring for a more holistic approach to mental health assessment. This research contributes to the growing field of AI-powered mental health solutions, bridging the gap between technology and psychological well-being while promoting early detection and accessible care.
Authors - Sonali Patil, Siddhesh Arun Patil, Ayush Patil, Piyush Pawar, Siddhesh Sandeep Patil Abstract - Around 970 million people face mental health disorders across the world and depression affects 75% of these people because of their sleep disturbances. The connection between persistent sleep problems and depression emerges when affected individuals become twice as likely to develop depression thus establishing sleep as a major sign for mental health forecasting. Current approaches to this problem deal with three key issues which are dataset biases, small available sample sizes along with the reliance on self-reported symptoms instead of actual physiological signals. Our deep learning solution relies on multi-channel Convolutional Neural Networks (CNNs) to analyze wearable sensor data because it tackles existing analysis limitations. DreamT-150 contains heart rate (HR), blood volume pulse (BVP) and electrothermal activity (EDA) measurements from 150 sleep patients. Three models including MultiChannelCNN and MultiChannelEfficientNet and MultiChannelResNet analyzed the signals which appeared as time-series graphs. The best model proved to be EfficientNet-B0 because it demonstrated superior generalization. The pre-trained layers from EfficientNet adjusted the vulnerability of training loss which led to stable model performance. The research demonstrates sleep-derived physiological signals' usefulness for non-invasive mental health predictions which can lead to real-time monitoring systems. The upcoming research aims to boost both dataset range and better models for clinical adoption requirements.
Wednesday August 26, 2026 12:30pm - 2:30pm IST Virtual Room BGOA, India
Authors - Bhoomi C. Parikh, Zankhana Shah Abstract - Stress is any type of mental imbalance that can lead to mental disorders ranging from low to high severities which can be classified as acute and chronic stress conditions. Chronic stress leads to hyperactivation of the sympathetic nervous system, resulting in physical, psychological, and behavioural problems. Currently, there is no recognised standard for stress assessment. Thus depressive disorders leading to stress are a flight or fight response to the stimulus generated by human nervous system caused due to any unacceptable behaviour or circumstance. Throughout this response adrenaline hormones are secreted that leads to increased respiration and heart rates, along with increased muscle activity. Such type of biological alterations prime the organism for a physical response that affects human body mechanisms in terms of sleep abnormalities, digestive disorders or work imbalances in routine lives. Thus WESAD is a multimodal wearable dataset which combines both affective states(baseline, depression and happy) and other sensor modalities such as blood pressure, ECG, skin conductivity , EMG, breathing, and three-axis acceleration. There are also other classification parameters based on physiological changes which are also found in WESAD dataset and by using different types of Machine Learning Classifiers analysis is done . The algorithms with the highest accuracy can be used for developing a novel and a hybrid model which can categorise stress based on Heartrate Variability and stating HRV as a biomarker for stress detection. Both characteristics related to time and frequency of heart rate are categorized in the research study. In the context of the three-class classification based on three affect states , baseline, stress, and amusement result up to 99% was obtained. Use of two affective states like stress and amusement gave an accuracy up to 84% using DT classifier.
Authors - Ajay J, Kavitha R, V S Ashwin, Dhanya M Abstract - The exchange-traded funds have seen greater condition as an entertainment choice in modern financial markets as they were designed for providing diversified exposure in equities as well as bonds and commodities types of asset classes. Despite the steadily increasing attractiveness and usage of such products, there still exists a gaping research gap in predicting their performance relative to sectors, especially in a comparatively emerging market such as India. This work intends to fill that gap by comparing Machine Learning models such as Random Forest and SVM with traditional models for sector-wise performance forecasting like ARIMA and Holt-Winters. Based on data available in investing.com, the work analyzes daily ETF prices across seven key sectors—Pharmaceuticals, FMCG, Banking, IT, Infrastructure, Consumption, and Healthcare—from 2021 to 2024. Performance of the model is evaluated by overall fit criterion: R² (coefficient of determination), Mean Absolute Error (MAE), Difference in Square Errors (DSE). Machine learning techniques have been found to considerably out-perform classical statical models in capturing complicated market activities, especially in volatile sectors like Infrastructure and Banking. Hill-Winters and ARIMA models reliably forecast stable sectors, such as Pharmaceuticals and Healthcare, while their kings fade away in overly dynamic markets. These research observations offer information to assist investors, portfolio managers, and policymakers as an illustration of the possibilities that exist for machine learning applications in financial forecasting. The integration of machine learning approaches should thus be magnified to improve on ETF price forecasting and investment strategies.
Authors - S.Prince Samuel, R.kiruba, P.Kingston Stanley, R.Karthick Abstract - The Internet of Things (IoT) has seen an increase in cyber attacks, especially botnet attacks, mainly brought on by weak security on networks. As a result of the rise in IoT users, the requirement for electronic data interchange, and the desire for virtual services, the frequency of cyberattacks to gain access to private data has increased in recent years. As a result, industry and researchers have given the security of IoT applications and particular data attention. A botnet is a formally organized group of infected, internet-connected devices managed by cybercriminals. Attacks from botnets, which spread spam and viruses and are no longer under the control of authorized users, can damage IoT devices. To effectively detect botnet attacks, proposed a botnet attacks detection system based on Transfer Learning (TL). The transfer learning (TL) model is built upon convolutional neural networks (CNNs), which are widely used for their effectiveness in feature extraction and pattern recognition in complex datasets. For the existing model achieved 91.93%, the proposed botnet attacks detection model performed with over 99.54% accuracy on two well-known public benchmark IoT security datasets: CICIDS2017 and UNSW-NB 15. This shows the proposed model’s effectiveness in predicting botnet attacks in an IoT environment.
Authors - Aaron Mendonca, Arya Gawde, Nikki Mehta, Nilay Koul, Rohit Parmar, Nikita Raichada Abstract - Communication barriers have a major influence on the deaf and mute society in India, resulting in social isolation and restricted access to education, employment, and everyday interactions. Indian Sign Language (ISL) is the primary mode of communication, but its lack of widespread understanding restricts integration with the larger society. This study presents a real-time ISL recognition and translation system that integrates deep learning, spatio-temporal analysis, and natural language processing (NLP) to overcome this communication barrier. This study proposes a real-time ISL gesture recognition and translation system utilizing Long Short-Term Memory (LSTM) networks, which are ideal for sequential gesture recognition so that accurate mapping of static and dynamic ISL gestures into text and speech can be done. A spatiotemporal feature extraction pipeline is incorporated using MediaPipe-based skeletal keypoint detection to guarantee strong recognition through capturing hand, facial, and body landmarks. The dataset, created with deaf and mute people’s inputs, provides regional gesture diversity and sign diversity. It has been engineered to operate effectively in real-world environments, with adaptations to lighting changes, background noise, and the complexity of gestures. This work contributes to assistive technology, accessibility, and human computer interaction, fostering social inclusion through facilitating effective communication between the hearing and non-hearing populations. This paper is a step towards a more inclusive digital communication environment, empowering the deaf community in various aspects of life.
Wednesday August 26, 2026 12:30pm - 2:30pm IST Virtual Room CGOA, India
Authors - Mohit Matte, Sandeep M.Chaware, Pratik Dahagaonkar, Anurag Deotale, Laukik Pagar, Jayesh Sarwade Abstract - Agriculture, in particular, has drawn a lot of attention lately due to the introduction of innovations like machine learning and smart computing. It is becoming increasingly challenging for farmers to effectively manage land and optimize profit in a particular terrain due to the changing economics of agri-produce. Crop yield forecast is heavily reliant on environmental parameters such soil composition, rainfall, humidity, and cultivable area, among other crucial indicators. Because they don't adequately account for a variety of environmental factors, traditional Crop Yield Prediction approaches like historical averages frequently don't yield reliable results. Furthermore, farmers find it challenging to choose crops and cultivate them effectively due to shifting market patterns in supply and demand. While a shortage of a certain crop could result in lost profit chances, a surplus production could result in reduced market pricing. Thus, combining yield prediction models with demand and supply research can assist farmers in improving crop planning for increased profitability. These challenges are addressed and accurate forecasts are generated using a machine learning-based approach. Crop prediction is done with classification models, whereas yield prediction is done with regression models trained on both historical and present data. To identify best course actions, these models examine a number of performance indicators. For practical use, the top-performing model is integrated into the backend. With a MAE of .64 , an R-squared mark of .96, Random Forest Regression outperforms the other models employed for yield prediction. At 99.39%, the Naïve Bayes classifier has the best accuracy for crop prediction. Predictions are further improved by adding market data to these models, such as price swings, customer demand, and past sales patterns. Farmers can improve profitability and minimize waste by matching their agricultural techniques with market demands through the integration of demand and supply analytics. This study demonstrates how machine learning may transform crop management by assisting farmers in making data-driven decisions to match their output with supply and demand in the market, as well as by optimizing resource allocation and raising total yield.
Authors - Vanishree Pabalkar, Reena Lenka, Jaya Chitranshi, Kalpesh Bhave Abstract - Barrier coating refers to a type of coating applied to the surface of a material, such as paper, cardboard, or plastic, to create a protective layer that prevents the penetration of liquids, gases, oils, or other substances. The primary purpose of barrier coatings is to enhance the material's resistance to moisture, oxygen, grease, and other environmental factors, thereby improving its functionality and extending its durability. In the context of the paper industry, barrier coatings are often used to make paper and paperboard suitable for packaging applications, particularly for food products, where protection from moisture and grease is essential. These coatings can be made from a variety of materials, including polymers, waxes, biopolymers, and even certain types of natural and sustainable compounds, depending on the desired properties and environmental considerations. Barrier coatings are crucial in the development of sustainable packaging solutions, as they allow paper-based materials to replace plastics and other non-renewable materials in various packaging applications. End Use of barrier chemical coated paper: Pizza Boxes, Pet food Bags/Boxes, Ice cream Frozen food, Fish Trays, Meat Packaging, Paper Cups & Plates, Cakes / Cookies, Wet Vegetables.
Authors - Archita Bhattacharyya, Ayan Bhaumik, Mrinal Kanti Deb Barma Abstract - The rapid expansion of the Internet of Medical Things (IoMT), a healthcare-driven subset of the Internet of Things (IoT), has introduced significant cybersecurity threats, underscoring the need for effective and privacy-preserving anomaly detection systems. In this study, we present an anomaly detection framework for IoMT data using autoencoder-based reconstruction loss analysis and feature space visualization. The reconstruction loss distribution enables the identification of anomalous samples using a predefined threshold. In addition, anomaly scores plotted against sample indices help visualize deviations in model behavior, distinguishing normal from suspicious activities. To better understand the latent feature space, the t-SNE visualization provides clear clustering of encoded representations, highlighting the separation between normal and anomalous patterns. This integrated approach offers an interpretable and effective means of detecting anomalies in IoMT environments.
Wednesday August 26, 2026 12:30pm - 2:30pm IST Virtual Room CGOA, India
Authors - Naman Yadav, Preety Sharma, Ayush Singh, Atharva Deshmukh, Aditya Thakur, Akshat Gora Abstract - This paper studies handwritten digit recognition methods with Convolutional Neural Networks (CNN) while performing a performance comparison with EfficientNetV2. The investigators applied the EMNIST dataset for model education and performance testing before using it to examine the model generalization characteristics through HASYv2 dataset analyses. The research examines key obstacles in handwritten digit recognition through multiple aspects such as different writing styles and diverse dataset characteristics as well as inefficient computing capabilities. The research evaluates enhanced accuracy through preprocessing methods along with model optimization methods. The research data reveals CNN provides excellent performance on EMNIST although it falls short on HASYv2 whereas EfficientNetV2 extracts superior features yet requires more computation power. The evaluation reveals the effective features and challenging aspects of both models so researchers can focus on developing hybrid structures and growing datasets for actual handwriting recognition systems in OCR applications and banking and automated document processing fields.
Wednesday August 26, 2026 12:30pm - 2:30pm IST Virtual Room CGOA, India
Authors - Anupama K, Kalyani Suresh Abstract - As digital tools become integral to parenting, understanding the psychological and practical factors influencing app adoption is crucial. Parenting apps are leaning progressively more towards integrating AI for personalization and societal benefits, which is an emerging area of study in the Indian context. While research points towards parental attitudes being significantly affected by AI mediated technologies, AI research culture is poised to draw on the experience and theory related to parenting. Drawing from Human-AI interaction theories, this study explores the hedonic and utilitarian motivations driving the use of AI-powered parenting apps among young Indian parents. The study uses a quantitative approach, to assess the extent to which young parents are motivated to use the AI-driven apps within the different levels of family support scenarios. Cluster analysis revealed the presence of four clusters based on their levels of hedonic or utilitarian motivations. Findings suggest that young Indian parents who use AI powered parenting apps are mostly Beta users – moderately engaging with selective feature usage - with both hedonic and utilitarian motivations playing crucial roles. Family support is found to improve hedonic and utilitarian motivations to use AI driven parenting apps. This study provides initial insights into the complex interplay between pleasure and practicality in technology adoption, setting the stage for larger-scale research on the impact of AI in parenting practices in India.
Wednesday August 26, 2026 12:30pm - 2:30pm IST Virtual Room CGOA, India
Authors - Prabira Kumar Sethy, Sachin Sharma, Ajit Behera, Satyaprakash Barik, Amresh Bhuyan Abstract - Signature verification constitutes a fundamental component of biometric authentication methods used in financial and legal identity verification systems. The research presents an offline signature verification method that examines geometric and morphological region-based features to authenticate test signatures. The methodology analyzes binarized signature images to extract important attributes such as area, perimeter, centroid, eccentricity, solidity, extent, major and minor axis lengths, orientation, convex area, Euler number, and equivalent diameter. After analyzing the reference signature collection, the most prominent image region gets processed for feature extraction. The test signature is evaluated through feature-wise similarity calculations while undergoing pre-processing identical to reference images. The normalization process for each feature difference allows comparison against specific thresholds to determine cumulative similarity scores. Authentication confirmation for a signature occurs when its score level exceeds the 95% predetermined acceptance benchmark. Experimental results demonstrate that our method achieves optimal computational efficiency while providing high verification accuracy and clear distinction between real signatures and forgeries. The framework merges reliable performance with simple operation and quick processing abilities making it ideal for lightweight biometric systems.
Authors - Atul Kumar, Devendra Kumar, Niranjan Kumar Abstract - The Internet of Things (IoT) has transformed the digital environment, but its fast expansion raises substantial cybersecurity concerns. IoT devices are naturally vulnerable to a variety of assaults, and the data they manage can be used by malevolent or unauthorized service providers. The introduction of IoT into cloud-based systems creates new security vulnerabilities. Cloud-based IoT solutions provide flexibility and scalability, but they also increase security vulnerabilities. The complicated interconnections between these traditional devices and systems demand strong measures to ensure privacy and integrity. This article tackles important security problems in IoT adoption by strategies to suggest in bridging present gaps and prepare for future difficulties. Its goal is to improve service security systems and device and strengthen IoT ecosystems through proactive approaches.
Wednesday August 26, 2026 12:30pm - 2:30pm IST Virtual Room CGOA, India
Authors - Moushmee Milind Kuri, Ganesh Pathak Abstract - Cyberbullying is a growing concern across social media platforms, necessitating advanced detection mechanisms to mitigate its impact. Traditional machine learning models often struggle with understanding contextual dependencies and ensuring model interpretability. This paper proposes a hybrid deep learning approach that combines BERT and RoBERTA for feature extraction and Bidirectional Long Short-Term Memory (BiLSTM) networks for sequential dependency learning. To enhance interpretability, attention mechanisms such as Self-Attention and Bahdanau Attention are integrated, allowing the model to focus on crucial words contributing to classification. The proposed system aims to improve accuracy, scalability, and explainability while addressing key challenges in cyberbullying detection. This research lays the groundwork for developing more transparent and effective AI-driven moderation systems for online safety.
Authors - Devi V S, Durgalashmi C V Abstract - This study assesses the effectiveness of income tax deductions and exemptions in promoting savings and investments in India. The Indian government has implemented various tax incentives to encourage individuals to save and invest, including provisions under sections 80C, 80D, and others. These deductions and exemptions are designed to stimulate economic growth by fostering long-term financial planning among individuals. The research examines the impact of these provisions on individual taxpayers' behavior and their overall influence on savings and investment patterns. Through a comprehensive analysis of available data, the study identifies the key tax incentives that have led to increased savings in instruments such as Provident Funds, National Savings Certificates, and insurance products. Additionally, the research evaluates the extent to which these tax benefits contribute to fostering a culture of investment and financial security. The study concludes that while tax deductions and exemptions have provided some incentives for savings, their effectiveness is often limited by lack of awareness and financial literacy. To further promote savings and investments, the study recommends improvements in policy communication, accessibility, and the alignment of tax incentives with broader economic goals.
Authors - Akash Tibeli, Saroja V Siddamal, Suneeta V Budihal Abstract - The AHB to APB Bridge is crucial component in System-on-Chip (SoC) designs, Achieving efficient communication between the pipelined AHB bus and the non-pipelined APB bus. In the proposed work a AHB to APB bridge is built using a bridge architecture which enables to translate pipelined, burst-oriented, high speed AHB transactions into sequential, low-power APB transactions by maintaining synchronization and data integrity. It was developed with a FSM to manage transactions and pipelining to maintain efficiency. Verification was performed using a Universal Verification Methodology testbench environment through direct and random testcases of burst, single, sequential, non-sequential transactions. 80 testcases were tested to obtain a functional coverage of 88%.
Wednesday August 26, 2026 12:30pm - 2:30pm IST Virtual Room DGOA, India
Authors - Judy K George, Elizabeth Sherly Abstract - Convolutional Neural Networks are extensively employed in critical domains such as computer vision, medical imaging, and autonomous systems. Enhancing model interpretability by providing users with concise and context-relevant explanations of CNN decision making such as visualizing feature maps or saliency regions, enables a deeper understanding of the model’s internal representations and inference process. The proposed work presents a deep learning framework integrating a ResNet-based U-Net architecture with a Fixation Point Generator (FPG) to perform classification and saliency aware reconstruction on the hand-written dataset. The model leverages transfer learning by employing a pre-trained ResNet-18 as the encoder backbone, enabling robust feature extraction. A custom decoder reconstructs input images while a classification head predicts digit labels. To enhance model interpretability, a Fixation Point Generator predicts spatial attention maps (saliency maps) from high-level global features, highlighting regions of interest that influence model decisions. This implementation aims to bridge the gap between classification performance and model explainability, offering insights into the model’s focus areas through learned attention. The model got an accuracy of 98.44 on the Malayalam handwritten dataset, 97.81 on English handwritten dataset, and 99.56 on the MNIST dataset.
Authors - Meghali Kalyankar, Om Pratap Gajra, Prathamesh Vilas Sagvekar, Mehul lalit Sharma, Zoheir Shahid Shaikh Abstract - Deepfakes are an emerging threat to digital authenticity and security, hence a proper detection technique needs to be created in order to establish public study confidence. A thorough roadmap to the development of deepfake detection software has been provided in this paper, reviewing the state-of-the-art algorithms, such as XceptionNet, EfficientNet, and hybrid models integrating spatial and temporal analysis. It provides methodologies for implementation, data preprocessing, and software pipeline development, serving as a practical guide to researchers and developers. Theoretical study to application-oriented practice closes the gap in terms of bottom line development and adaptive detection systems addressing the growing menace of deepfake media.
Authors - PRANAY SAMAL, K R LOKESH KUMAR, CHAKILELA SAIRAJ, G VIDYA SRI, SUSHAMA RANI DUTTA, SUKLA SATAPATHY Abstract - Music plays a significant role in our day-to-day life, and selecting appropriate songs can enhance the experience. This paper describes an intelligent music recommendation system that applies machine learning to recognize what users prefer and recommend music that they will like. It incorporates various approaches, including considering user decisions and music attributes, to enhance suggestions. The system adapts based on user actions and refines recommendations over time. Findings indicate that this method provides easier and more precise music discovery. This paper emphasizes how technology can assist in providing a higher quality and better personalized music experience.
Authors - Arya Tripathi, Akash Mecwan Abstract - In recent years, the RISC-V architecture has emerged as a promising platform for embedded systems, offering flexibility and open-source accessibility. Consequently, the demand for secure communication in embedded devices, particularly within the Internet of Things (IoT) ecosystem, has driven the adoption of cryptographic algorithms. Integrating cryptographic functionalities into RISC-V architecture presents unique challenges, requiring innovative solutions to optimize performance and security. In response, the proposed design introduces an approach to address these challenges by incorporating a dedicated cryptoprocessor module into the RISC- V architecture specifically designed to handle encryption and decryption tasks efficiently. The cryptoprocessor module employs the Blowfish-64 algorithm to ensure robust security while lowering the computational overhead. Blowfish is a well-established symmetric-key block cipher known for its simplicity and efficiency. The compact design of the cryptoprocessor module significantly reduces resource utilization and execution time compared to existing implementations while preserving security and functionality. The design emphasizes low resource utilization, achieving a utilization rate of 34% (11,340 out of 33,216 available units) with an execution time of 160 ns. The implementation is carried out using Verilog HDL for the Cyclone II EP2C35F672C6 based FPGA.
Wednesday August 26, 2026 12:30pm - 2:30pm IST Virtual Room DGOA, India
Authors - Kirti Karande, Sujata Kadu, Deven Shah Abstract - Breadth-First Search (BFS) is a foundational graph traversal algorithm, it’s systematic layer-by-layer exploration of nodes makes it invaluable for a variety of domains, including transportation networks, social network analysis, and artificial intelligence. However, traditional BFS implementations face challenges when dealing with large-scale graphs due to memory limitations and inefficiencies in handling massive datasets. This project addresses these challenges by integrating BFS with a CSV-based data storage system, enabling efficient traversal of large graphs without relying on a traditional SQL database or requiring the entire graph to be loaded into memory. The graph data, comprising nodes and edges, is stored in CSV files, which act as lightweight and accessible storage. The implementation is memory-efficient due to the use of Pandas DataFrames for handling CSV data and NetworkX graphs for traversal. Additionally, the integration of a machine learning model from Scikit-learn, a memory-efficient library, ensures effective prioritization of edges without excessive computational overhead. In this project, we address a key limitation of the traditional Breadth-First Search (BFS) algorithm: its inability to consider edge weights during traversal. It is unsuitable for scenarios where varying edge weights significantly impact the traversal outcome, such as in shortest-path calculations for weighted graphs. To overcome this drawback, our project integrates a machine learning (ML) model to analyze and prioritize edges based on their weights, effectively augmenting BFS for weighted graphs. By leveraging CSV-based storage and combining it with an ML-driven edge prioritization mechanism. this project offers a scalable solution for managing and analyzing large, weighted graphs. This combination ensures that the navigation system not only computes the shortest path but also suggests the most practical and efficient routes tailored to user preferences or constraints.
Authors - Dipak Ligade, Chiranjit Das, Rupali Parte, Masira Kulkarni, Shivraj Jadhav, Abhishek Mohite Abstract - —This project aims to automate tree counting and forest land diversion assessment through satellite image combined with advanced computing techniques. The treatment of forest re sources needs accurate monitoring because growing environmental challenges such as deforestation, biodiversity loss, and climate change require it for sustainable land management. The research uses satellite imagery along with machine learning and deep learning tools, specifically convolutional neural networks (CNNs), to precisely detect and count trees across expansive territories. The study demonstrates how satellite analytics technologies will enhance forestry applications with their capabilities for better tree enumeration at higher efficiency and greater accuracy.
Authors - Suchanta Ravan, Prashant Dhotre Abstract - Conventional identification techniques that depend on privacy concerns and credentials are becoming more vulnerable to web-based risks like hacking and data breaches. The need for sophisticated authentication techniques has grown dramatically because of identity theft, cyberthreats, and illegal access. Traditional security methods, such as PINs and passwords, are insufficient for high-security applications since they are vulnerable to phishing, brute-force assaults, and credential breaches. To improve safety and tackle problems like privacy threats, spoofing, and accessibility problems, this study suggests a strong adaptive authentication mechanism that integrates biometric along with behavioral assessment. The multimodal authentication framework guarantees a smooth and easy verification process while also enhancing security. This structure guarantees a smooth and safe authenticating process by utilizing cutting-edge security methods like encryption, machine learning, and multifaceted biometrics in conjunction with a user-centric architecture. By combining behavioral biometrics with conventional authentication techniques, total authentication reliability is increased, and cyber risk is mitigated. The effectiveness of the suggested approach in lowering susceptibility to cyberattacks while preserving superior usability and consumer satisfaction is demonstrated by experimental findings, Highlighting the importance of two-way authentication.
Authors - Manasi Golesar, Priti Jagtap, Kamlesh Khatod, Kshitij Malode, Vaishali Pawar Abstract - This research presents a novel approach to bot detection in web applications using behavioral biometrics and machine learning. Our system leverages a Flask based web application with a registration form as a testbed to distinguish between human and automated users. The implementation collects multidimensional behavioral data including mouse movements, typing patterns, form fill speed, and browser fingerprinting to build a comprehensive user profile.Two machine learning models, Random Forest and XGBoost, are dynamically compared for performance, with the superior model being automatically selected for deployment. The system incorporates a honeypot field as a simple yet effective first pass filter and implements progressive model learning through a database backed training pipeline that continually improves detection accuracy.Key innovations include the real time behavioral analysis during form completion, automated weekly model retraining, and an administrative interface that allows for manual labeling of edge cases to enhance the training dataset. Our approach achieves high detection accuracy while maintaining a low false positive rate, effectively balancing security with user experience.This research demonstrates that integrating behavioral biometrics with adaptive machine learning provides a robust defense against increasingly sophisticated bot attacks without requiring traditional CAPTCHA challenges that often degrade user experience.
Authors - Devarsh Damodaran, Krishna Bharathi V, Dhanya M Abstract - This study explores user sentiments towards AI-powered fitness applications by analyzing user reviews from platforms like Google Play Store. With the increasing adoption of digital health solutions, understanding user satisfaction, trust, and key concerns is crucial. Using Natural Language Processing (NLP) techniques, sentiment analysis was conducted to classify user feedback into positive and negative sentiments. Machine learning algorithms like Logistic Regression and Support Vector Machine (SVM) were utilized for classification. Findings are prominent drivers of satisfaction, where usability, effectiveness, and personalization are essential drivers, while cost, technology glitches, and unrealized expectations drive dissatisfaction. These findings give interesting insights for fitness-tech business companies and app developers to drive engagement and better experience for their users.
Authors - Silpa Raj R, Durgalashmi C V Abstract - The widespread adoption of social media enables female entrepreneurs to leverage innovative tools and strategies, fostering the development of sustainable business practices. The present study analyses how the female entrepreneurs in Kerala utilize social media (SM) in promoting sustainable innovations in their business activities. Research investigates how social media affects sustainable business practices among women entrepreneurs in Kerala, with the focus of four key variables: idea generation, customer connectivity, collaboration, and sustainable outcomes. This study aims to fills the gap by exploring how women use social media for entrepreneurial practices and adoption of sustainable outcomes. This study used a structured questionnaire to collects data from women entrepreneurs in Kerala. The variables including frequency of idea generation through social media, customer or stakeholders’ collaborations and the adoption of sustainable practices influenced by digital platforms are observed. To ensure the participation of entrepreneurs actively using social media for innovation, purposive sampling techniques were employed. The hypotheses were tested using the statistical tools like chi-square, correlation and regression analysis, providing empirical evidence on impact of SM usage among Kerala’s women entrepreneurs. The study emphasizes the significance of social networking platforms in encouraging innovative methods that contribute to sustainability via three major variables. The findings suggest practical implications for policyholders, entrepreneurs and researchers. The research adds existing corpuses of research on digital entrepreneurship and sustainability focusing on how social media improve sustainable practices.
Authors - Vineet Wagh, Srushti Chopade, Sneha Patil, Vighnesh Padwal, Sarika Kuhikar Abstract - In Institutions and schools, attendance management is a crucial task for faculty to monitor class strength. Traditional methods such as manual entry, biometrics, and RFID-based systems are commonly used, but they are time-consuming and, in the case of biometrics, potentially unhygienic. This paper presents an automated face recognition-based attendance system that utilizes preinstalled CCTV cameras to monitor student presence in real-time. The system employs RetinaFace for face detection and the face_recognition library for face encoding and matching. Known face images are preprocessed to generate face encodings, which are then compared with detected faces in each frame to determine attendance. The proposed system offers accuracy, efficiency, automation, and contactless operation while seamlessly integrating with existing infrastructure. A web interface allows users to start and stop attendance tracking, remove duplicate records, and download attendance logs in CSV format. The system demonstrates its applicability in educational environments by providing a scalable, non-intrusive, and secure solution for automated attendance management.
Authors - Madhumati Shinde, Premanand Ghadekar Abstract - Cloud computing is a dynamic part of today's high-tech framework, given that frequent welfares such as cost-effectiveness, scalability, convenience, novelty, and safety. Its impact is multifaceted, transforming competition and corporate operations in the digital age. To improve speed, optimize resource usage, and support sophisticated applications, cloud computing makes use of a variety of learning strategies. A learning technique's effectiveness in the field of cloud security depends on its ability to recognize, stop, and handle security threats. In order to identify and reduce security threats, machine learning particularly anomaly detection using supervised and unsupervised learning is crucial with advancement of federated learning. Deep learning models like RNNs and CNNs process extensive datasets to uncover intricate attack patterns, while federated learning improves privacy by training models on decentralized data sources. Reinforcement learning facilitates adaptive security strategies, continually enhancing threat responses. Security is paramount in cloud computing as it safeguards sensitive data, applications, and services hosted on cloud platforms from unauthorized access, breaches, and cyber threats.This paper highlights the security concerns in cloud environment with framework to improve the performance matrix to recognize federated cloud trust.
Authors - Ajuram. P, E. Grace Mary Kanaga Abstract - Large Language Models have demonstrated great effectiveness in generating text and images. However they can become even more efficient by perfecting the prompt given to them. This paper proposes a multi LLM framework that dynamically orchestrizes several specialized LLM models in accordance with complex user prompts. First, a primary LLM analyzes the user prompt and breaks it down into multiple sub tasks. Then, for each identified sub task with respect to its type (text to text, text to image, or image to text), a suitable LLM is assigned. The context, instructions, and the output format is also provided by the primary LLM for each chosen model. The sub tasks are executed either in parallel or in sequential order. This approach automates the workflow, optimizes model utilization, and improves response relevance, making it suitable for applications requiring multi modal collaboration and processing.
Wednesday August 26, 2026 12:30pm - 2:30pm IST Virtual Room EGOA, India
Authors - Devika Vijapur, Nidhi Desai, Aishwarya Naik, Smita Ganur, Supriya Katwe Abstract - Road accidents are one of the global safety concerns leading to loss of millions of lives every year. One of the factor leading to this is over-steering. Oversteering is phenomenon that occurs when the rear wheels of the vehicle lose grip which causes the vehicle to turn more than expected. The detection of oversteering in real-time is crucial for the improvement of vehicle safety to prevent accidents as well as for advanced driving assistance systems(ADAS). This paper presents a holistic approach to over-steering detection using a decision tree algorithm. The proposed system analyzes various vehicle dynamics parameters such as lateral acceleration, yaw rate and steering angle to identify the patterns that cause over-steering. The system incorporates collection of real-time data from Inertial Measurement Unit (IMU) sensors that enhances reliability of oversteering detection under various conditions. The model is trained from the data obtained, using decision tree algorithm and obtained accuracy of 96.08%. The hardware implementation is done by placing ESP-32 integrated with MPU 6050 and Arduino Nano 33 BLE sense accordingly in the vehicle. Based on thresholds of the parameters mentioned in the paper, oversteering is detected.
Wednesday August 26, 2026 12:30pm - 2:30pm IST Virtual Room EGOA, India
Authors - Sneha S. Temgire, Y.S. Angal, Ashwini V. Waghmare, Chetana Sharma, Ashwini Gajre Abstract - Agriculture is essential for food security and economic growth, but traditional farming faces challenges such as plant diseases, inefficient irrigation, and labour-intensive monitoring. This project focuses on automated and manual irrigation in addition with plant disease detection and growth monitoring using image processing on a Raspberry Pi 3B+. By leveraging TensorFlow Lite and OpenCV, the system can analyze plant health and trigger appropriate irrigation actions. The aim is to design accurate agriculture system by reducing water wastage and improving crop monitoring. A key feature of this system is web-based monitoring, where the Raspberry Pi transmits real-time plant health data and sensor readings to an HTML-based webpage. Users can remotely access this data via a web interface, enabling continuous monitoring of plant conditions, disease status, and irrigation control from any location. By combining machine learning, image processing, IoT automation, and real-time web-based monitoring, this system reduces manual labour, optimizes water usage, and ensures early disease detection. The web interface enhances accessibility, allowing farmers and researchers to track plant health remotely and make informed decisions.
Authors - Sauvik Bose, Rina Bhattacharya, Rajeshwari Roy Abstract - Avian monitoring is a crucial component of biodiversity conservation, providing insights into population trends, habitat changes, and environmental stressors. The fast growth of mobile telephony has raised issues regarding its potential upon the avian population, their behaviors and breeding, predominantly due to electromagnetic radiation exposure. This study investigates the feasibility of using drones for avian monitoring near mobile towers in Arambagh Municipality (22.8838° N, 87.7819° E), Hooghly, West Bengal, India, which is a semi-urban landscape with rich avian diversity and has undergone a significant growth in mobile tower installation over the last few decades. Drones offer a non-invasive, scalable, and high-resolution method for ecological monitoring, surpassing traditional survey techniques in terms of not only efficiency and data accuracy but also consuming less time and effort. A drone (model: DJI MAVIC MINI) equipped with a high-resolution camera is deployed at selected base station sites within the study area. The study pattern included regulated flight patterns, periodic monitoring. Findings disclosed noticeable behavioral variations in birds near mobile base stations. The repulsion of smaller birds to the high EMR zone has been distinctly observed along with anomalies in roosting and breeding habits. A correlation was observed between radiation levels and avian health oddities, underscoring the need for further research. In the future, research ought to be performed on in-depth monitoring efforts in urban and semi-urban areas along the different geographical landscapes. Improving drone technology for ecological studies and exploring alternative communication infrastructures with reduced environmental impact is much needed.
Authors - Dhanyashree S, Keshav S, Deepak Gupta, Shobhana Palat Madhavan Abstract - This study investigates a chain reaction triggered by cliffhangers in media consumption, focusing on their role in driving binge-watching, self-regulatory depletion, binge-eating, and reduced mental well-being. Grounded in Self-Regulatory Depletion Theory, a sequential mediation model is proposed and analyzed through serial mediation regression. Data from 170 Indian respondents revealed that cliffhangers significantly predicted binge-watching, which in turn increased self-regulatory depletion. Depletion heightened binge-eating tendencies, and binge-eating negatively impacted mental well-being. Bootstrapped mediation confirmed an indirect pathway from cliffhangers to reduced mental well-being via binge-watching and self-regulatory depletion. These findings underscore the ethical responsibility of streaming platforms to mitigate compulsive viewing and highlight interventions for mindful consumption.
Authors - Kavya Soni, Sujal Rajput, Babita Tiwari, Chirag Joshi, Gaurav Kumawat Abstract - Surface water quality is essential for ecological stability and mortal health, but it faces growing pitfalls from urbanization, industrialization, and husbandry. Traditional in-situ monitoring styles are essential yet limited in their spatial and temporal compass. This paper aims to provide a comparative analysis of different techniques available for surface water quality analysis. We have analysed studies grounded on freely available satellite data from Landsat, Sentinel- 2, and MERIS to determine crucial water quality parameters similar to chlorophyll- at attention, turbidity, and dangerous algal blooms. The review demonstrates the effectiveness of various methods to use spectral imaging to predict parameters such as BOD, chlorophyll content in water. Further to this multi-sensor data integration within the pall calculating platform Google Earth Engine aids in dynamic water quality assessments. Results indicate these technologies indeed give scalable low-cost observers of submarine ecosystems and implicit means of filling gaps between in- situ measures and comprehensive water resource operation. The study identifies implicit in the integration of a Civilians approach grounded on remote seeing in climate modelling, monitoring of ecosystem health, and sustainable water governance.
Authors - G. Indhumathi, G. Saranya, S. Riju Sundar, S. Paul Joseph Abstract - Sericulture or silkworm breeding for silk is faced with the challenges of maintaining the ideal environmental conditions, feeding patterns, and disease recognition. Manual and improper monitoring lead to compromised production and quality. This project introduces the implementation of an Augmented Reality (AR)-based real-time system for sericulture management using the intersection of IoT and AI. The system keeps tracks of temperature, humidity, and feeding patterns and presents real-time visualization of data in an interactive AR platform. An AI subsystem identifies diseased silkworms via image processing, annotates them in AR, and recommends treatment. Predictive analysis also maximizes environmental conditions and feeding patterns for maximum production effectiveness. The uniqueness of the system is its interconnection of AR, AI, and IoT that provides easy monitoring, automatic detection of diseases, and data- in-formed decision-making. The utilization of the system enhances the quantity of silk yield, product quality, and saves labor, and its disruptive contribution to sericulture management is evident through innovative technologies.
Authors - Geetanjali Popat Rokade, Sonali Patil Abstract - The pressing need for secure, private, decentralized frameworks for machine learning in healthcare has been fueled by the increasingly popularization of Federated Learning (FL). In conventional FL, the aggregation is centralized, allowing potential data leakages or model-poisoning attacks against a central point of failure. A possible solution to these aforementioned impediments is Blockchain-Based Federated Learning (BDFL), as such a setup can utilize the immutability, transparency, and distributed consensus of the blockchain to enhance security and achieve better performance. Nevertheless, the existing review articles have not offered a thorough investigation of BDFL consensus algorithms, their specific applications to the healthcare sector, and an iteratively empirical performance evaluation of their efficiency, scalability, and robustness. This paper provides a systematic and empirical review of state-of-the-art BDFL consensus programs in their application to health care; it analyzes these programs' performances based on consensus efficiency, incentive mechanisms, privacy-preserving capabilities, and computational scalability. Key approaches examined in this study include Proof-of-Contribution (PoC) [2,3], Byzantine Fault Tolerance (BFT) [5], DAG-based Blockchain FL [4,13], Multi-center Federated Learning (MCFL) [24], and Proof-of-Accuracy (PoAcc) [20]. The reason for this focus is that these methods best integrate security, efficiency, and fairness in the context of decentralized health data cooperation. The results indicate that MCFL models would optimize institution-wise healthcare cooperation, PoAcc would optimize the accuracy of medical diagnosis, and the DAG-based blockchain would guarantee high throughput scalability for FL. This review sets out an extensive framework for selecting the best models in BDFL, which will encourage developments in AI-nurtured healthcare data analysis, clinical decision support, and secure EHR management. This study's findings will propel future advancement in federated learning security, quantum-safe consensus mechanisms, and hierarchical blockchain architectures for global health applications.
Authors - Parvathi NB Panicker, Bhadra R, PR Mahadevan, Vandana Madhavan Abstract - Diversity, Equity, and Inclusion have integrated into organizations through incorporations in their Strategic Plans. The presence of a globally dispersed workforce in the IT/ITES sector implies that these strategies are particularly vital in those organizations. Many organizations made pronouncements of publicly declaring their DEI initiatives; however, usually a difference exists between such declarations and the experiences of the employees. This study investigates the given DEI initiatives in IT/ITES organizations through two lenses: namely, by organizational disclosures as well as employee perception. The qualitative research methods involved the gathering of data with corporate DEI reports, sustainability statements, and employee-generated reviews through semi-structured interviews with employees. Thematic analysis reveals leading gaps of representation of leadership, equity in progression of careers, and inclusion incidences in the workplace. Diversity is preached at entry-level but drops off in representation at leadership levels. Promotion and pay equity remain as sticking issues: underrepresented groups tend to progress in their careers at slower rates. Employees considered organizational DEI commitments as more aspirational than actual, with workplace inclusion and psychological safety differing in various organizations. Employees expressed skepticism because many DEI efforts do not set measurable success metrics. The study also underscores that organizations should go beyond performative DEI efforts by incorporating employee feedback, installing structured mentorship programs, and adopting outcome-based DEI evaluation systems
Authors - Sohana R, Niharika R, Khushi Shah, Tanya Singh, M Shahina Parveen Abstract - The project majorly includes a methodology to create an AI - driven career counselling platform that can be used to recommend various career options for students (focusing on starting to give them more exposure from a younger age. So that they can incorporate the necessary skills required or in general know what is in it for them in every career option available) based on every individual's profile and varied interests. We utilize artificial intelligence to make sure we can provide personalization of suggestions. The platform takes factors like the interests of students, their strengths and what kind of work environments they would want to work in, and then evaluates a list of suitable options. There are also prevailing recent studies that indicate that such systems powered by AI have enhanced the accuracy and reliability of career counselling services by a great extent especially by analyzing extensive behavioral and educational data. Upon this our platform utilizes augmented reality for simulating real- world career environments, making sure that students get a chance to explore their potential career paths by interactively taking part in the simulations. There has also been extensive research that has demonstrated that Augmented reality-based tools on the whole improve and provide enhancement in immersion, hands on experiences and helps with better exploration for various professions. Therefore, we want to merge AI and AR to arrive at best of both worlds and hence approach this problem by providing with an innovative platform that fosters informed decision making and comprehensive career exploration among students. Ultimately our platform's mission is to spread awareness and to align the aspirations that students have with their career paths and to lead to the overall improved educational and career outcomes and job satisfaction.
Authors - Janwale Asaram Pandurang, Minal Dutta, Savita Mohurle, Vaduguru Venkata Ramya Abstract - This study investigates the classification of images of spine X-ray into three groups: Normal, Scoliosis, and Spondylolisthesis, deep learning models improves with attention mechanisms. A labelled dataset of X-ray images was working, addressed with imbalances class through oversampling techniques. Pretrained convolutional neural network (CNN) models, including Xception, InceptionV3, and DenseNet, were fine-tuned for this categorised task. The combination of attention mechanisms enhanced interpretability of model and precision score. Working with the models, InceptionV3 achieved perfect accuracy, outperforming Xception and DenseNet. The findings insides the efficacy of attention-based deep learning approaches with potential applications in clinical diagnostics, in medical image classification, for spinal conditions.
Authors - Yogesh K. Sable, Rajesh Kumar Kashyap, Sagar Satpute Abstract - Microgrids have emerged as cutting-edge and game-changing energy solutions, providing a plethora of benefits in the search for a robust and sustainable energy future. In-depth examination of the many facets of microgrids is provided in this review, with specific consideration paid to their capability in the mix of sustainable power sources, support for charge and e-portability, and contribution in a debacle readiness and flexibility. The topic of conversation is the arrangement of limited energy frameworks by means of microgrids, which might work both autonomously and related to the essential electrical network. They successfully consolidate environmentally friendly power assets, like sunlight powered chargers and wind turbines, and advance the development of electric vehicles through wise accusing and connection of the framework. Additionally, because of their intrinsic resilience, they may keep operating in the face of grid failures and natural disasters, supplying crucial backup power to crucial facilities. Case studies highlight the real-world uses of microgrids in various contexts and highlight their potential effects on environmental sustainability, cost savings, and energy efficiency. The improvement of microgrids is expected to assume a significant part in making versatile and maintainable energy framework as the globe faces rising environment related concerns.
Authors - Jaden Ekbote, Sheshank K Patil, Ramakrishna S, Nalini C Iyer Abstract - In the era of 5G, the dual imperatives of high performance and energy efficiency have led to the development of sophisticated network management techniques. This paper introduces an innovative slice-aware energy optimization framework that leverages simplicial homology to model and analyze network coverage. By representing base stations as vertices in a simplicial complex and encoding overlapping coverage as higher-dimensional simplices, the approach captures connectivity and potential coverage gaps through homological invariants. An optimization algorithm is then formulated to minimize overall power consumption while fulfilling stringent slice-specific quality-of-service (QoS) constraints for enhanced Mobile Broadband (eMBB), Ultra-Reliable Low-Latency Communications (URLLC), and massive Machine-Type Communications (mMTC). Extensive simulations in MATLAB demonstrate the viability of the proposed method, showing significant power reductions over baseline uniform allocation schemes without compromising slice performance. This work underscores the potential of topological methods in addressing the energy challenges inherent in next-generation network deployments.
Authors - Aditya Poddar, Soham Sarkar, Ananya Hegde, Shravya Reddy, Animesh Giri Abstract - As climate change accelerates, there is an urgent need for solutions that balance ecological responsibility with economic incentives. While capping carbon emissions is widely recognized as essential, it remains a challenging task to quantify carbon sequestration correctly and ensure complete transparency in carbon credit markets. The increasing demand for effective carbon sequestration measurement and transparent carbon credit trading demands an innovative approach using advanced technologies. This research focuses on applying big data using Kafka for parallel data streaming in a distributed environment, together with machine learning models to optimize the prediction of carbon capture, integrating blockchain technology which provides security and transparency in transactions involving the carbon credit market. Through our research, we aim to provide an interdisciplinary framework that will improve the accuracy and scalability of carbon sequestration predictions, building trust and accountability in carbon trading to support a more sustainable and economically viable future.
Wednesday August 26, 2026 3:30pm - 5:30pm IST Virtual Room AGOA, India
Authors - S. T. Patil, Gaurav Sulsule, Urmila Kakarwal, Sanika Kolawale, Prathmesh Deshmukh Abstract - This paper suggests a deep learning-based solution for real-time detection of drowning and slipping accidents through computer vision. The system, which is grounded on the YOLOv8 (You Only Look Once) model, offers effective and efficient detection by analyzing video streams in real-time to detect dangerous incidents in settings such as swimming pools, building sites, and home homes. The system has a web-based user interface, real-time alerting capabilities, and SQLite database for storing data. The model was trained and tested with a large set of labeled images with an emphasis on balancing detection performance on frequent and infrequent incident classes. The results include robust detection performance with few false negatives and positives, fast response times, and effective processing of multiple video feeds. Despite problems with dataset imbalance and integration complexities, the system offers a cost-effective solution for enhancing safety, minimizing human error, and enhancing real-time monitoring capability. The research suggests the viability of AI-based solutions for safety-critical domains, with advantages of automated incident detection over conventional surveillance techniques.
Authors - Payel Das, Siri Kethineedi Abstract - This study explores the elements of consumer resistance to sustainable marketing with an integrated theoretical approach adopting cognitive dissonance theory, institutional theory, and theory of planned behaviour. Although awareness of sustainability is increasing, consumers frequently do not accept sustainable products because of psychological discomfort, institutional barriers, and perceived behavioural limits. Using interpretive structural modelling (ISM), this study elucidates the hierarchy relationships among the main barriers, such as greenwashing, lack of transparency, price sensitivity, norm conformity, and instantaneous gratification. The most impactful of these drivers were identified as greenwashing and transparency deficits, both of which contribute to distrust and ultimately erode consumer confidence. Weak regulations and social norms that perpetuate these problems are demonstrated by Institutional Theory, while price premiums and limited access reduce perceived behavioural control and are described in the theory of planned behaviour. This study proposes a multi-tiered effort for policymakers and businesses to address resistance. Transparency will be enforced through independent certifications and stringent sustainability standards regulated by the regulatory frameworks. To regain consumer trust, companies must embrace true sustainability and communicate honestly. Price premiums can also be lowered through innovation, subsidies, and supply chain efficiencies to help play a role in them become more affordable. Understanding these barriers allows businesses to understand how they can build consumer trust, policymakers to enact effective regulations, and society as a whole to begin moving toward more sustainable consumption habits.
Authors - Payel Das, Sonali Bolisetty, Digumarthi Iswarya Abstract - This study seeks to identify the success enablers of omnichannel retailing by using Interpretive Structural Modeling (ISM) for building a hierarchical framework. From findings in a consumer survey with 108 consumers and expert evaluations, the research uncovers user enablers such as technological infrastructure, data analytics capability, personalization, mobile optimization, and seamless integration. The study is based on Service-Dominant Logic (SDL) and the Technology Acceptance Model (TAM) to investigate theory around foundational, operational, and experiential issues that result in customer engagement and brand loyalty. The findings underscore the critical importance of strong technological infrastructure and the use of real-time data in helping with friction reduction between digital and physical touchpoints. Using AI-powered analytics, it can improve personalization, which affects perceived system usefulness and thus, customer satisfaction. In addition, the study emphasizes the need for brand consistency and proper employee training to provide trouble-free service experiences. We also explore privacy and security concerns and their impact on consumer trust and omnichannel adoption. To policymakers, this research calls for prescriptive regulations that will protect data privacy and grow the space of technological innovation. For practitioners, it provides actionable insights to maximize omnichannel universality, improve customer pursuits, and develop sustainable total brand loyalty. In doing so, with the introduction of SDL and TAM, the study contributes to theoretical knowledge and proposes a comprehensive framework for the businesses who are dealing with the complexities of omnichannel retailing.
Authors - Ketki Kshrisagar, Chinmay Kalbhor, Sudarshan Chitte, Atharva Chivate, Pragati Chopade, Sanika Chougule Abstract - This paper describes the design and implementation of an automated control system for grain storage temperature and humidity. Operations begin using a microcontroller, Arduino Uno, and DHT11 or thermocouple sensors, for real-time environmental conditions, whereas the temperature and humidity are controlled through a Peltier module and a USB spray humidifier, to give the ideal storage conditions. Another complementing feature is an I2C LCD, which visualizes real-time parameters for the users locally to monitor environmental conditions. Besides, the system also includes a Blynk app, which allows the users to monitor and control it through a phone interface from a remote location. The main function of this is to act as a standalone and inexpensive system, which is primarily aimed at reducing grain spoilage and ensuring quality. Test results have confirmed that it was able to provide applicable environmental control for various storage scenarios
Wednesday August 26, 2026 3:30pm - 5:30pm IST Virtual Room BGOA, India
Authors - Fathima Mariya A K, Sarath S, Jyothisha J Nair, Sunitha E V Abstract - Thermal images often hide mixed signals, making accurate analysis challenging. However, segmentation and analysis are significantly compromised with the task of mixed pixels (a pixel containing the signals from several endmember sources). This study proposes a hybrid approach combining gradient-based thresholding (80 percentile and 85 percentile) and different clustering techniques (K-Means, Variational Bayesian GMM, Dirichlet Process GMM and Constrained GMM) to boost precision in mixed pixel identification. Results show that the gradient threshold has a positive effect on detection error (20.77 percentile), closely matching the values of K-Means (20.82 percentile) and Constrained GMM (20.69 percentile). The deviation from those methods to VBGMM and DP GMM is more moderate by 13.60 percentile. This study confirms the usefulness of an integrated approach for a more accurate interpretation of thermal images. Deep learning and multi-spectral will be researched to boost segmentation accuracy in the future.
Authors - Rashmi S. Bhumbare, Pallavi S. Gaikwad, Anjali M. Gutte, Araju M. Shaikh, Gayatri K. Chaudhari Abstract - In this paper, ensuring secure, transparent, and tamper-proof elections is a critical challenge in modern democratic processes. Traditional voting systems, including paper ballots and electronic voting machines (EVMs), suffer from issues such as fraud, lack of transparency, and centralized control. This project presents a Blockchain-Based Voting System, implemented as an Android application using Java/XML, with SHA- 256 encryption ensuring vote security and Firebase Realtime Database handling user authentication and data management. The system leverages blockchain technology to record votes in an immutable and decentralized ledger, preventing manipulation and unauthorized access. The implementation includes secure voter authentication, encrypted vote submission, blockchain-based integrity verification, and real-time result compilation. This approach eliminates traditional vulnerabilities such as vote tampering, duplicate voting, and unauthorized system access. Furthermore, the decentralized nature of blockchain ensures transparency, allowing voters to independently verify their votes while maintaining anonymity.
Authors - Beena B.M, Devika Madhusoodanan, Nithin Sagar, Vismaya R, Hridyalakshmi Santhosh Abstract - Energy conservation in cloud data centers remains one of the biggest research challenges today. Energy efficiency has become an important concern in the management of contemporary data centers due to the rapidly growing computational needs and the environmental impact of power consumption. This study examines various power management techniques, including Dynamic Voltage and Frequency Scaling (DVFS), Dynamic Power Management (DPM), and Adaptive Voltage Scaling (AVS), to optimize CPU power consumption. Using frequency data from historical and current CPU usage, these algorithms control CPU frequency settings and assess their impact on energy consumption and performance. The results indicate that DVFS reduces power consumption by 25-30%, DPM achieves energy savings of 28-35%, and AVS provides savings of 35-40% by dynamically adjusting both voltage and frequency. A performance matrix evaluates the power savings and utilization efficiency of these strategies to determine the most suitable approach. AVS was found to be 5-10% more energy efficient than DVFS and DPM, demonstrating its advantage in real-world applications. Furthermore, AVS exhibited the highest precision (96%) to adapt to workload fluctuations, compared to 95% for DVFS and 92% for DPM. This study focuses on adaptive power management and provides key findings on algorithmic solutions for energy efficiency in software-defined cloud infrastructures. The findings contribute to reducing data center energy consumption while maintaining performance, aligning with the UN Sustainable Development Goals by promoting sustainable and eco-friendly cloud operations.
Authors - Ajay Menon, Anjali Sivan, Navya S, Sandhya G, Astha Santhosh T Abstract - The micro, small and medium enterprise sector plays a crucial role in Kerala’s rural economy and makes a substantial contribution to socio-economic development and job creation. This study explores business continuity intentions among women-led micro enterprises in rural Kerala, using thematic analysis of in-depth interviews with six units from agro-processing, dairy and fisheries sectors. Drawing insights from qualitative data, this study uses the Theory of Planned Behaviour (TPB) to show that continuity intentions are strongly influenced by perceived behavioural control, strong family support, and positive attitudes. However, institutional inefficiencies and financial limitations create significant obstacles. This study also introduces ‘Team-led resilience’ and ‘Gendered leadership dynamics’ as critical factors, highlighting collaborative support and autonomous female leadership. These findings highlight the importance of financial literacy, access to credit, and supportive government policies, which will also help to expand the traditional TPB framework, emphasizing the importance of social and financial resilience.
Wednesday August 26, 2026 3:30pm - 5:30pm IST Virtual Room BGOA, India
Authors - M. Suresh, T. A. Alka, Aswathy Sreenivasan Abstract - The main purpose of this study is to theoretically explore the evolution of trends in teaching entrepreneurship through a Bibliometric analysis. The final number of documents selected is 1375, which are analysed through the Biblioshiny package under R programming. The results show that there are technology-related and non-technology-related trends that have evolved in teaching entrepreneurship. Major trends are happening in teaching methods, learning, courses, global reach, teamwork, and the emergence of technology trends such as artificial intelligence, virtual reality, etc. Bibliometric results draw that the major themes evolved in this domain are related to innovation trends in teaching entrepreneurship for shaping entrepreneurs for tomorrow, transformation, learning culture, technology trends, academic entrepreneurship in the covid-19 pandemic, learning types, concepts, skills required, sustainability and teaching entrepreneurship, entrepreneurialism and thinking in teaching entrepreneurship. The major future research avenues are; entrepreneurial intention; effectuation; entrepreneurship, business model innovation; innovation; digital transformation, and entrepreneurial university; academic entrepreneurship; innovation. The limitations of the research are; the Scopus database is only used for the search. Only the documents in the English language and final publication stage papers were selected. The inherent drawbacks of the bibliometric methodology may influence the results. The study offers theoretical implications for future research work including Sci-Val future research topics and practical implications by offering insights to entrepreneurs, investors, researchers, academicians, policymakers, etc. The novelty and the originality of the study are underlying in the in-depth theoretical exploration through a comprehensive literature review of the past thirty years.
Authors - Ganga S, Nitharshana P, Varun Madhusoodan, Rojalin Patri Abstract - Advancement of financial technology has resulted in the emergence of automated investment solutions, such as robo-advisors. While Gen-Z investors are typically receptive to digital innovations, their adoption of robo-advisory services remains an underexplored area. This study investigates the primary factors affecting Gen-Z's inclination to use robo-advisors, applying the Technology Acceptance Model (TAM). A quantitative methodology was utilized, with data collected from 161 respondents and analyzed through multiple regression techniques. Findings indicate that trust and attitude have a significant impact on the adoption intent of robo-advisory services in investment decisions made by Gen-Z investors. The results suggest that fostering trust and shaping positive perceptions of robo-advisors are more crucial for adoption than enhancing usability. This study contributes to fintech literature and offers insights for financial institutions and policymakers aiming to increase robo-advisory adoption among young investors.
Authors - Apolinar P. Datu, Annaliza C. Sinfuego, Garry C. Bayran, Dominic T. Urgelles, Julius R. Beltran, Rossana B. Liray, Janina Odette S. Vidallon, Erwin Joel B. Layug Abstract - The rapid transition to online learning, catalyzed by the global pandemic, has necessitated a critical examination of its implications within the context of general education. This study investigates the multifaceted factors influencing the implementation, delivery, and reception of online classes in general education programs across selected higher education institutions. Employing a mixed-methods research design, quantitative data were gathered through structured surveys while qualitative insights were obtained via in-depth interviews with students and faculty members. Results indicate that technological accessibility, digital competency, instructional quality, learner motivation, and institutional support are central determinants of effective online learning. The research highlights disparities in students’ digital readiness and access to conducive learning environments, which significantly affect their academic engagement and performance. Moreover, pedagogical adaptability and the integration of interactive tools were found to be critical in maintaining student interest and participation in virtual settings. The findings underscore the necessity for higher education institutions to invest in sustainable digital infrastructures, provide continuous faculty development programs, and adopt inclusive, student-centered online learning strategies. This study contributes to the growing body of literature on e-learning by offering empirical evidence on the challenges and enablers of online education in general education curricula. It also presents actionable recommendations aimed at enhancing the quality and equity of online instruction. In doing so, the research supports the advancement of resilient and adaptive educational systems capable of meeting the evolving demands of 21st-century learners.
Wednesday August 26, 2026 3:30pm - 5:30pm IST Virtual Room BGOA, India
Authors - Nisha Fulzele, Chetan Parlikar Abstract - The development of financial instruments has greatly changed consumer expenditure patterns, and credit cards have been central in contemporary economies This paper analyzes the expenditure behavior of credit card customers in Wardha City, with reference to priority drivers of expenditure patterns. Employing a descriptive research method, primary data were gathered from 140 participants using a systematic questionnaire. Analysis proves that young professional salaried individuals constitute the maximum segment of credit card customers, who prefer online payment and high-end transactions. Whereas convenience and payment flexibility come with credit cards, their use in everyday consumption is still limited. Correlation analysis indicates that rewards, cashback, impulse buying, and financial security drive spending most, compared to peer influence and promotional offers, which have lesser impacts. The research indicates that credit card use in Wardha City is increasing, driven mostly by electronic payment behavior and financial stability. By comprehending these behavior patterns, financial institutions can make strategies to encourage prudent use of credit and financial literacy among consumers.
Authors - Radhika V. Kulkarni, Vaibhav Aher, Harsh Ukey, Sujal Dubey, Aarya Labhshetwar, Manjiri Kulkarni Abstract - The majority of community in globe use sign language as the most basic way of interaction with Deaf and speech-impaired people. In most instances, a person finds it difficult to learn sign language for communicating with deaf and dump people, which leads to isolation among those individuals. Most people are unaware of the interpretations made in sign language. Hence, this paper presents an intelligent sign recognition system for translation of dynamic sign language for easy communication among people with hearing and speech impairments. The intelligent system takes advantage of advanced computer vision and deep learning techniques to identify dynamic hand signs accurately. This approach includes video data capture, preprocessing, feature extraction, and real-time gesture recognition. Hand movements are captured from webcam video streams, and the MediaPipe library is used to capture key points over the hand. A sequential model based on deep learning maps the relationships in hand gestures, which ensures high recognition accuracy. Extensive testing on different hand gesture recognition datasets shows that they perform efficiently and reliably in real-world situations. This technology facilitates greater accessibility through the ability to quickly and accurately translate sign language, thereby helping create inclusive communication technologies.
Authors - Molly Goel, Prince Kumar Sharma, Nainshi Singh, Madhvi Gaur Abstract - A secure and dignified electronic voting system is needed to provide the security and decency of a traditional one. While still allowing for flexibility and accuracy, this system has been tested for a long time. The use of blockchain technology can be utilized to actualize distributed voting structures. Despite the technological advancements that have occurred in the past few years, the traditional balloting system still remains unsuited for the modern era. There are numerous issues that prevent the integrity of the elections, such as the lack of transparency and the use of bribes. Besides these, the time it takes to check the vote's integrity is also very long. Current technology has to be used to improve the voting system. One of the most important factors that needs to be considered is the development of blockchain technology. This type of innovation eliminates the character flaw in the voting process and ensures that the correct votes are sent out. The development of blockchain technology is carried out through a stable set of rules that are designed to solve the problems related to the voting process. This type of innovation will help to ensure that the public can easily remember the individuals who participated in the process. The development of a voting poll programming application can help the political selection executives and citizens get the most out of it. However, it can also expose them to various risks. For instance, e-voting can lead to political race safety issues and fraud. Despite the advantages of this type of innovation, it is still not ideal for the people who are interested in maintaining a transparent and honest political selection process.
Authors - Manisha Mane, Saurav Bedse, Vikrant Patil, Pruthviraj Dhande, Om Darekar Abstract - The fast advances in deep learning and computer vision have dramatically improved the ability to detect objects, with applications in surveillance, driverless cars, and smart traffic management. The current paper describes an implementation of the YOLOv8 model for real-time object detection on different categories such as persons, cars, and bicycles. We trained the model on a customized dataset of annotated images, fine-tuning it through extensive hyperparameter tuning and multiple training epochs. Our training setup consisted of 75 epochs, utilizing a Tesla T4 GPU for computation. The model recorded a mean Average Precision (mAP@50) of 76.5% over all classes, with class performance highlighting high precision and recall rates for classes like cars (98.2%) and bicycles (87.8%). To further improve accuracy, we utilized data augmentation methods, batch normalization, and optimizer tuning. After training, the model was subjected to extensive validation, with an inference speed of 8.5ms per image, making it viable for real-time performance. We also incorporated the model into a realistic deployment pipeline, showcasing its efficacy in real-world applications. This paper presents a thorough analysis of the trained model, such as performance metrics, comparison with other versions of YOLO, and discussion of future improvements. Our results emphasize the model’s ability to achieve speed and accuracy balance, rendering it an appropriate choice for object detection in real-time applications. Future research will investigate additional optimizations such as light-weight model variants and domain-specific dataset adaptation.
Authors - Abhay Shinde, Ketal Patil, Nirmitee Chaudhari, Samrudhi Bachhav, Kavita Moholkar Abstract - The energy that never goes out of style is solar energy that is readily available and produces no pollution; its use has increased over the years. It is an endless supply of energy. Optimising solar radiation absorption for power generation is still a major challenge. A solar panel's best position for collecting sunlight is orthogonal to the trajectory of the sun's rays, but throughout time, the sun's rays direction varies. Even though a solar tracking system does a good job of recording the sun's motion during the day, it suffers when adverse weather conditions cause the sun's intensity to decrease. A passive tracking system, which can handle such circumstances and yield better results, can therefore be employed to overcome them. The design and functionality of a solar tracking system are the topics of this research. By aligning the solar panel with the sun's position, which is grounded on a fluid medium, the suggested outcome offers the best possible conversion of solar energy into electrical power.
Wednesday August 26, 2026 3:30pm - 5:30pm IST Virtual Room CGOA, India
Authors - Trupti Chetan Kherde, Dhiraj Jitendra Marathe, Prathamesh Shivaji Kadam, Sanskar Dipak Shinde, Chetan Balaji Phulmante Abstract - Agriculture is one of the fundamental pillars of human civilization. In addition to providing food, it boosts the economy. Crops and plant leaves are susceptible to several diseases during agricultural production. Diseases prevent each species from growing. Early and accurate plant leaves disease diagnosis helps to minimize major damages to plants. Plant leaves disease classification and detection has grown to be major issues. Failure to promptly identify and categorize plant diseases could lead to agricultural plant loss and a sharp decrease in product. Utilizing digital image processing techniques in their fields can help farmers enhance output and decrease losses. Various techniques have been developed and implemented to identify and classify plant diseases. Over the years, considerable advancements have been made in finding different disease by exploring and applying different methodologies. However, because of new developments, and conversations, improvements are needed. Globally, crop production can be greatly increased with the application of technology. Conventional techniques, such as laboratory-based diagnostics and manual inspection, are still dependable but time-consuming and labor-intensive. Emerging technologies, such as Machine learning (ML) and deep learning (DL) techniques have revolutionized automated disease detection, offering robust solutions for analyzing complex patterns in plant images. This survey highlights recent advancements in these areas.
Authors - Jayasmita Kuanr, Deepanjali Mishra Abstract - Dalkhai is conventionally a female-centric folk tradition; nonetheless, patriarchal frameworks have frequently influenced its performance and distribution. Grounded in Judith Butler's theory of gender performativity, which analyses how Dalkhai's lyrical narratives and physical expressions formulate, contest, and navigate gender identities. The emergence of digital media has allowed Dalkhai to explore new avenues of representation, enhancing reinterpretations of old themes and promoting wider interaction. Digital media and technology-enhanced performances have elevated female voices, but they may also commodify or alter traditional expressions to conform to modern cultural norms. This study contends that although digital technology provides opportunities for transformation and inclusivity, it also requires critical awareness about the recontextualization of traditional folk narratives in virtual environments. The study indicates that the convergence of gender performativity and digital media is transforming Dalkhai’s cultural relevance, establishing a dynamic arena for both continuity and transformation. The technology integration and folk traditions such as Dalkhai can transform while preserving their artistic integrity, providing novel opportunities for female representation in the digital era. Therefore, the study examines the changing performance of Odisha’s Dalkhai folk music via the perspectives of gender performativity and digital transformation. It proposes a critical textual and performative examination of Dalkhai's lyrics, gestures, and vocal expressions to elucidate how the folk tradition both reinforces and subverts gender stereotypes.
Authors - G.B.Sambare, Prajwal Solase, Raj Lokhande, Chaitanya Shinde, Sujit Aher Abstract - Heterogeneous wireless networks face challenges in ensuring smooth mobility between WLAN and LTE, as traditional handover decisions based on signal strength often degrade service quality. A more advanced approach incorporates multiple network parameters like signal power, link speed, system delay, and user mobility for optimized vertical handover. Real-time throughput calculations and dynamic network ranking enhance selection, while MCDA techniques improve transfer continuity, reduce delays, and minimize packet loss. Simulation results confirm that this strategy outperforms conventional methods by reducing handover failures and improving network selection. Additionally, advanced techniques like FSHO and SSHO are explored for seamless multimedia services in 5G networks.
Authors - Khushi Ingalalli, Vanshika Kavi, Sainath Walthati, Satish Chikkamath, Suneeta Budihal, Sujata Kotabagi Abstract - With the millions of tweets per day, Twitter is a rich and large database of information on public sentiment on a wide range of issues, including events, products, politics, and social issues. The purpose of this research is to create an automated system that can analyze tweet sentiments to determine attitudes as positive or negative. Through Natural Language Processing (NLP) methods and machine learning algorithms, the system efficiently handles high quantities of unstructured data, making sentiment classification possible in real time. The model begins the analysis by gathering various tweets from various sources, such as hashtags, user mentions, and trends. The tweets are then subjected to preprocessing techniques like removing stop words and treating misspellings, emojis, and special characters. Various classification models, like Naive Bayes, Support Vector Machines (SVM), Logistic Regression (LR) were experimented with to see which was most efficient in sentiment classification. Of these, Logistic Regression (LR) showed the best performance with an F1 score of 0.833 and accuracy of 83%. The efficiency of various feature extraction methods, such as Term Frequency- Inverse Document Frequency (TF-IDF) and word embeddings, was also examined to try and improve model performance. This work emphasizes the increasing importance of Twitter Sentiment Analysis across different fields, such as market research, event tracking, and social research. Sentiment analysis is employed by companies to know customer views and enhance services, whereas policymakers utilize it for measuring public reaction. By combining NLP and machine learning, the suggested system provides better and scalable method for sentiment analysis[1].
Authors - Gopal D. Upadhye, Deepak T. Mane, Devang Gentyal, Chetan Channa, Shubham Landge, Radhika Gadewar Abstract - Toxic comment identification in Hinglish (a combination of Hindi and English) is a difficult task because of code-switching, transliteration, and class imbalance. This paper suggests a machine learning based method for identifying toxic Hinglish comments based on TF-IDF feature extraction along with an ensemble model. In order to mitigate class imbalance, Random Oversampling was utilized, and model interpretability was facilitated using SHAP (Shapley Additive Explanations). The suggested model was trained on publicly released datasets, with 90.0% accuracy compared to individual classifiers. This work contributes to content moderation system for code-mixed languages and offer an extensible solution for social media toxicity detection.
Authors - S M Boomika, C M Tulasi, Sharvani V Nagur, Bhagyashri Badakali, Nalini C Iyer, Preeti Pillai, Ujwala Patil Abstract - LiDAR and cameras play a vital role in autonomous vehicles by providing complementary data for object detection and environmental perception. However, achieving seamless data integration from these sensors depends on partial and temporal synchronization. Unlike conventional methods that depend on pre-calibrated datasets, our methodology utilizes a custom-acquired multimodal dataset comprising both image and video data from a monocular camera and point cloud data from a VLP-16 Velodyne LiDAR sensor. In this paper, we proposed a comprehensive framework for LiDAR and camera calibration and temporal synchronization of real time data, synthesized and validated in a controlled lab environment. Calibration of the raw data was performed using a checkerboard as the target to ensure accurate spatial alignment between heterogeneous sensor systems.The collected corpus is further timestamped, synchronized, and validated.The accuracy of the proposed methodology is evaluated by projecting LiDAR points onto image frames, enabling qualitative verification of spatial and temporal consistency. The proposed method integrates target-based calibration with software-level timestamp synchronization to create a reproducible and scalable calibration pipeline. Results demonstrate accurate alignment across modalities, validating the effectiveness of our approach. This 1 work provides a practical contribution to multi-sensor fusion research, especially for applications requiring custom datasets or operating in constrained environments.
Authors - Wendrila Biswas, Arunangshu Giri, Dipanwita Chakrabarty, Dibyendu Rath Abstract - The study has examined the effect of user engagement (UE), perceived benefit (PB), and perceived risk (PR) of wearable sensor-based healthcare devices adoption. User empowerment (UEM) in IOT-enabled healthcare has been explored on the basis of two established theories, Technology Acceptance Model (TAM) and Behavioral Reasoning Theory (BRT). A cross-sectional online survey was conducted from November 2024 to January 2025 involving 361 valid Indian respondents and the collected responses were analyzed through NVivo software for qualitative analysis. SEM (structural equation modeling) was done for quantitative analysis and hypothesis testing. The findings have shown a positive association between UE and PB and between UE and PR. Again, the study has revealed that PB and PR positively influenced UEM. The study contributes both to existing literatures and making managerial decisions by establishing how benefits from wearable sensor-based healthcare devices can be explored by avoiding the perceived risk of the consumers and how they can get empowered with the same.
Authors - Vaishali Langote, Siddhesh Kulkarni, Aaditya Ghorpade, Aditya Songirkar, Aditya Chincholkar Abstract - Identifying customer retention is essential for decreasing lost revenues as well as maintaining an established base of loyal customers. By reviewing historical data that includes customer demographics, purchasing habits and behaviours, businesses will be able to determine which customers are going to discontinue using their services or products. In generating models that can identify customers at risk, this process includes machine learning models such as decision trees, logistic regression and neural networks. It is important that predictive retention can work provided the right algorithms are selected, and reliable data is sourced. Continual updates and improved models will enhance accuracy, giving firms the opportunity to keep up with changes in how consumers behave. The models will also give businesses the ability to produce more targeted retention marketing plans since they will not only identify at-risk customers but also give clear data on what they are doing to create customer churn.
Authors - Sharon Koshy, Padmadas Sundaram Abstract - The intensifying depletion of natural resources, fueled by world population growth and unsustainable consumption, poses severe threats to global sustainability. Specifically, the ICT and smart infrastructure industries make substantial contributions to resource inefficiencies through growing e-waste, inefficient material recovery, and unsustainable construction methods. Forecasts suggest that by 2050, with a projected 9.8 billion world population, resource use will surpass planetary limits, urging rapid interventions in resource management and the transition to circular economies. Despite growing recognition, inefficiencies in recycling infrastructure, defective waste-to-energy technologies, and inadequate water management persist to drive global resource insecurity and further environmental degradation. Solutions must be backed by evidence-based policy design, technological development, and systemic change. In this context, the combination of Artificial Intelligence and biomimicry offers a new way to increase sustainability and resilience in systems. AI-based models improve resource efficiency, reduce environmental footprint, optimize waste management, facilitate predictive maintenance, and enhance material recovery, while biomimicry offers nature-inspired solutions for sustainable design, energy efficiency, and waste reduction. These technologies not only foster resource recovery but also set the stage for the creation of wiser, more sustainable industries and cities. In conclusion, this study high- lights the revolutionary power of ICT that AI and biomimicry make possible to create closed-loop, self-sustaining models that boost urban resilience, sustainability, and efficiency, maximize recovery of resources, minimize waste, and maximize value for a truly circular future.
Authors - Priya Surana, Sushma Vispute, Madhura Kalbhor, Shubhangi Vairagar, Pragati Ugale, Imtiyaz Syeda, Mahek Yakumsha, Ashish Suryawanshi Abstract - This research presents a YouTube Comments Analyzer that leverages machine learning and deep learning algorithms to examine and classify user comments. A large volume of comments is processed by the system, enabling it to detect key patterns, including sentiment classification and emotion detection. Using natural language processing and machine learning techniques, the tool provides meaningful insights to content creators for understanding their audience and to moderators for identifying problematic content. Researchers can also benefit by studying online commentary at scale. Our team collected video comments from various genres to train and develop the models, followed by evaluation using multiple performance metrics. The analysis tool achieves 96% accuracy in sentiment detection and 90% accuracy in emotion detection, successfully identifying complex patterns that manual evaluation often misses. To demonstrate the practical applicability of our models, we further developed a web-based application that integrates the analysis pipeline, providing an accessible platform for real-time comment analysis. This research highlights the effectiveness of automated text analysis in social media environments and demonstrates real-world applications for YouTube content management and audience engagement strategies.
Authors - Omkar Kalantre, Jyoti Joglekar Abstract - Optimized Bounding box fitting around an object is necessary for accurate localization of the Region of Interest (ROI), so that features extracted from the ROI are useful for many computer vision applications. Current methods tend to be inefficient, imprecise, and with high computational complexity. In this work a novel algorithm is presented that is designed for fitting a bounding box around an object that covers maximum part of the object as ROI,. The improvement in inserting bounding box enhances the process of recognizing, tracking, and classifying objects, which is highly valuable for applications such as surveillance, autonomous driving, and security. In this work we are proposing a novel algorithm for fitting a bounding box around an object to maximize the object area covering and for minimizing the background clutter as a part of ROI.
Authors - Mohan Sellappa Gounder, Rohan Mahantesh Kamatgi, Sharath Prabhu T M, Sanya Gupta, Seema Abstract - This research investigates the application of the DINO (Distillation with No Labels) framework, a self-supervised learning approach, for efficient road and pothole segmentation. By integrating a DINO-enhanced ResNet-50 backbone with a U-Net model, this study addresses segmentation challenges in dynamic environments. The framework employs momentum encoders, multi-crop training, and stability mechanisms to facilitate robust feature extraction without requiring labeled datasets. Through strategic fine-tuning, the model achieves precise segmentation of road surfaces and potholes, making it a promising approach for real-world applications in autonomous systems and infrastructure assessment. This study further discusses model evaluation, comparison with state-of-the-art approaches, and its implications for transportation infrastructure.
Authors - Gopal D. Upadhye, Ranjana Jadhav, Aryan Pungale, Ashish Shadija, Nikita Rajput, Pranav Pendse Abstract - A data-informed system is described for generating crop recommendations and crop yield forecast based on a variety of data sources of farmer-level soil characteristics, historical crop yield records, and meteorological variable data. In the proposed system, crop recommendations based on a classification algorithm and crop yield estimates based on a regression algorithm are provided to farmers. The data-driven crop recommendations and crop yield forecasts will improve decision-making by providing the farmer with data-based recommendations providing the productivity isolation. The data-informed system will utilize machine learning algorithms to process the data and analyze the complex interaction of the various farming agri-parameters in the farm operation. Composition of soil nutrient values, weather patterns, and historical productivity variable data will be a key ingredient in the model to provide farmers with singularly specific crop selections. Ability to yield prediction gives farmers anticipate yield of the crops, improve resource planning. The validation tests demonstrate better accuracy than traditional heuristics, improving farmer overall risk reliability and increasing efficiency, sustainability. The results shows us that the transformative role of machine learning in agriculture and the associated movement toward precision farming practices
Authors - Vanshika R Kavi, Sujata Kotabagi Abstract - Semiconductor production demands high-quality control to detect faulty wafers early on in the production process. Manual inspection and rule-based systems are conventional methods that are time consuming and error-prone. This research investigates machine learning (ML) based wafer detection on a dataset of 590 sensor readings per wafer, with wafers being labeled as good (+1) or faulty (-1). Several traditional ML models, such as Logistic Regression (LR), Support Vector Machines (SVM), K-Nearest Neighbors (KNN), and Random Forest, are tested for defect classification effectiveness. The processing of data includes handling missing values by dropping features with high missing data and using median imputation. Feature selection is done through SHAP (Shapely Additive Explanations) analysis and correlation filtering to select only the most important sensor readings. Feature scaling is done to maintain consistency in data distribution. For handling the class imbalance in the dataset, SMOTE (Synthetic Minority Over-sampling Technique) is employed to create synthetic samples for the minority class to enhance model learning. Once trained, the models are evaluated on the basis of accuracy, precision, recall, F1-score, confusion matrix, and SHAP-based explainability analysis. SVM and Random Forest perform better compared to other models with 97-99% accuracy, and KNN does not perform well because of high dimensionality. The research showcases how ML is able to automate defect detection, increase production efficiency, and minimize human inspection errors. Work for the future encompasses ensemble learning optimization, real-time deployment, and semi-supervised learning optimization for enhanced defect classification in the semiconductor industry.
Authors - Piyusha S. Shetgar, Asha V. Thalange, Rohini R. Mergu, Aishwarya Khobare Abstract - Throughout the world, the number of educational institutions has significantly increased in recent decades. But the majority of recently established universities continue to manage their resources, including their hostels, using traditional methods. These conventional methods are frequently hindered by innate restrictions that negatively impact the organization's overall effectiveness. This study suggests an automated hostel lodging management system that is made with Microsoft Access as the underlying database and Visual Basic as the programming language to handle these issues. To stop unwanted access, the system has an integrated authentication algorithm. The system that has been built leverages face recognition technology to address the shortcomings of conventional approaches. It provides a graphical user interface, dependability, efficiency, and improved security by implementing access control mechanisms.
Wednesday August 26, 2026 3:30pm - 5:30pm IST Virtual Room DGOA, India
Authors - Manasa S, Rupam Bhaduri, Pramod Kumar Naik, Gangadhar T G, Bharath Kumar S Abstract - This paper introduces an adaptive control strategy for a three-phase Dual Active Bridge (DAB) converter, designed to facilitate efficient bidirectional power flow in electric vehicle (EV) fast-charging stations. The proposed control method effectively manages real-time fluctuations in grid conditions and the state-of-charge (SOC) of batteries, ensuring stable operation in both Vehicle-to-Grid (V2G) and Grid-to-Vehicle (G2V) modes. Utilizing a dq-reference frame-based decoupled controller with SOC feedback, the solution is rigorously validated through MATLAB/Simulink simulations. The design encompasses LCL filter modeling, DAB phase shift modulation, and battery interfacing under diverse loading scenarios. Simulation results reveal significant improvements in performance, highlighting the system's ability to maintain high efficiency during both charging and discharging phases. By enhancing the responsive-ness and stability of power exchange between EVs and the grid, this research aims to contribute to the development of advanced fast-charging infrastructure capable of supporting increasing EV adoption while optimizing overall electric grid performance. The findings underscore the potential of adaptive control strategies in ensuring reliable and efficient energy management within smart grid environments.
Authors - Saraswati Patil, Kalyani Rathod, Adarsh Jayfale, Wasim Pathan, Adesh Bhore Abstract - This paper looks at how object detection technology can help blind and visually impaired people. Visually challenged individuals struggle to comprehend their surroundings, especially in outdoor settings where objects constantly shift and move. Object detection solutions can help visually impaired individuals overcome difficulties in daily life. The object detecting system aims to provide a simple, user-friendly, convenient, and cost-effective solution for visually impaired individuals. This yolov11 model has the frame process rate 45 FPS on CPU and 100 – 150 FPS on GPU .The system was tested with different objects and in various environments to see how well it works. Key factors like how accurate it was, how quickly it responded, and how satisfied users were measured. The results showed that the system was good at detecting objects and giving clear instructions to the user in real time.
Authors - Aranya G, Durgalashmi C V, Nidheesh Melethadathil Abstract - This study examines the confidence levels of healthcare and IT professionals regarding the execution of artificial intelligence (AI) in the healthcare sector. Focus on understanding the perceived impacts of AI on patient safety, quality of care, and the ethical and legal implications involved, the research employed an analysis through a detailed questionnaire, gathering responses from 50 healthcare professionals and 50 IT professionals in Kerala using judgmental sampling. Survey model and percentage analysis were used in this study. The findings indicate a mixed sentiment: a substantial proportion of respondents acknowledge the potential of AI to enhance healthcare delivery and patient outcomes, yet there remains significant apprehension concerning data privacy, potential biases, and the need for human oversight. While IT professionals generally display greater confidence and familiarity in AI technologies, healthcare professionals are more cautious, emphasizing the importance of ethical considerations and human involvement in clinical decision-making. The study suggests that bridging the gap between these professional groups through targeted education, hands-on experience, and robust governance frameworks can enhance confidence and facilitate the effective integration of AI in healthcare. Recommendations include ongoing training and clear communication about AI's capabilities and limitations to ensure both ethical application and improved healthcare outcomes.
Wednesday August 26, 2026 3:30pm - 5:30pm IST Virtual Room EGOA, India
Authors - Ritu Ramesh Vernekar, Vijeta D Chitragar, Laxmi Koutanali, Prajwal Sangalad, Hemantaraj M Kelagadi, Suhas B Shirol Abstract - The ESP32 microcontroller and the Blynk IoT application are integrated in a novel system for automatic irrigation and tank water level management. Sensors for water levels, rainfall, temperature, and soil moisture track real-time environmental parameters. Temperature readings ranged from 25°C to 31°C over the 8-day research, but soil moisture was continuously kept within ideal ranges. Water waste was reduced and timely refills were ensured by the water tank level sensor mechanism, which successfully maintained a threshold of 15 cm. Based on sensor data, intelligent algorithms control irrigation, minimize waterlogging, and maximize water usage. Convenience and operational efficiency are increased via remote management via the Blynk app. This intelligent irrigation system provides a sustainable and effective answer to contemporary agriculture by preserving water, improving crop health, and facilitating data- driven farming methods.
Authors - Yash Prajapati, Ketul Patel, Nidhi Acharya, Nidhi Dubey, Nisarg Patel Abstract - The Internet of Things (IoT) is progressively changing and offers IoT ecosystems integrated network security challenges that require sophisticated security solutions. In this paper, we discuss the hybrid model that combines Federated Learning (FL) with Random Forest (RF) algorithms along with the validation of Blockchain to provide adaptive network security within IoT frameworks. The proposed architecture merges Blockchain’s protection against unauthorized access with the automatic updates and data processing of FL, decentralizing the security measures within the IoT ecosystems while increasing detection accuracy and safeguarding sensitive infor-mation. This framework overcomes the constraints imposed by centralized machine learning intrusion detection techniques, providing solutions to real world IoT security issues.
Authors - Harsh Raj, Kanishk Tewatia, Sumeet Gupta Abstract - Channel encoding plays a vital role in modern communication systems by maintaining data integrity and reducing the impact of noise. In this paper, we propose a hybrid model that combines Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks to classify various channel encoders. This approach aims to improve feature extraction and classification performance compared to traditional CNN architectures. In typical scenarios, receivers are aware of the encoder’s type and configuration. However, in non-cooperative environments such as military communications, surveillance, and cognitive radio systems, this information is often limited or unavailable. To address this, we explore a deep learning-based method to identify four types of encoders: block, convolutional, Bose–Chaudhuri–Hocquenghem (BCH), and polar encoders. By integrating CNN and LSTM layers, our proposed model achieves up to 98% classification accuracy and demonstrates strong generalization. Comparative analysis reveals that the hybrid model outperforms conventional CNN-based methods in terms of accuracy and robustness.
Authors - Arunkumar V N, Agna.S. Nath, Aswathi.K. B Abstract - This study examines the disconnect between India’s cybersecurity policies and their real-world implementation, revealing systemic barriers to digital empowerment. Through qualitative analysis, the research identifies four critical challenges: inadequate awareness programs, urban-rural security divides, gender-based vulnerabilities, and educational gaps in cyber-literacy. Findings show urban users exhibit risky digital behaviours despite high connectivity, while rural populations avoid online services due to security fears. Women face compounded risks, with many dependent on male relatives for digital access. The education system largely fails to equip students with basic cybersecurity knowledge. However, community-led initiatives demonstrate promising alternatives. Localized, vernacular training programs have successfully enhanced digital safety awareness and reduced fraud incidents. These models highlight the importance of contextual, participatory approaches to cybersecurity education. The study argues for rethinking cybersecurity as an essential dimension of human development rather than just technical infrastructure. It proposes shifting from compliance-focused governance to capability-building frameworks that prioritize protective freedoms for all citizens. Key recommendations include integrating cybersecurity into school curricula, developing gender-responsive digital safety programs, and creating community-based "digital mitra" networks. By bridging policy intentions with ground realities, this research offers pathways to make India's digital growth truly inclusive and secure.
Wednesday August 26, 2026 3:30pm - 5:30pm IST Virtual Room EGOA, India
Authors - Pratyush Jaishankar, Ayman Aftab, Divyanshu Vyas, Dhanashree G Bhate Abstract - The research proposes a distinctive method to identify unauthorized people who enter restricted areas through a combination of KLD7 millimeter wave radar systems and deep learning algorithms. Gait patterns obtained from Doppler and micro-Doppler signals are analyzed by the system which offers both privacy preservation and non intrusiveness as opposed to conventional methods like CCTV surveillance. The Random Forest Classifier shows excellence by accurately identifying authorized or unauthorized individuals at a rate of 82% while maintaining its capabilities during various challenging environmental situations. The solution provides high practicality when used for real-time monitoring deployments. Future development efforts will direct their attention to growing the dataset while making the solution work efficiently on edge computing devices.
Wednesday August 26, 2026 3:30pm - 5:30pm IST Virtual Room EGOA, India
Authors - Shreya Kapadia, Payal D Joshi Abstract - In the era of IR, event detection has moved beyond simple keyword searches to utilize advanced techniques to extract relevant events from massive news article datasets. The rapid growth of news highlights the need for efficient information retrieval techniques to capture the most relevant events. Traditional lexical-based retrieval methods, such as Whoosh and BM25, are effective in keyword matching; however, they have some limitations in understanding the semantic events from the indexed text. To enhance this limitation, this study introduces a Transformer-based deep learning model for Natural Language Processing (NLP), such as BERT, capable of capturing contextual relationships and improving the relevance of data. This research also explores an optimized approach that seamlessly integrates Whoosh for efficient indexing, BM25 for probabilistic ranking, and BERT for neural re-ranking, designed to improve event detection performance. Additionally, Named Entity Recognition (NER) significantly enhances event extraction by accurately identifying real-world entities like individuals, locations, organizations.The results of this research indicate that the integration of lexical models(Whoosh and BM25) with neural ranking models(BERT) significantly enhances precision, recall, and relevance, thereby exceeding the performance of traditional retrieval techniques. In our experiments BERT achieved a relevance score of 62% ,outperforming BM25 , which scored 55%. This demonstrates superior ability to capture contextual and semantic relationship in text. In conclusion, this study articulates prospective directions for future research within the realm of event detection, improving the efficacy of information retrieval in rapidly evolving news environments.
Authors - Sarika Kuhikar, Kashish Mishra, Tejas Dabholkar, Tejal Narvekar, Siddharth Suyal Abstract - This paper presents the design of a control circuit for a single-phase inverter capable of generating a pure sine wave output that is accurately aligned with the desired voltage amplitude and frequency. With the global shift toward renewable energy sources, the need for efficient and reliable power conversion systems has become more critical than ever. The proposed design utilizes advanced microcontroller technology along with modulation techniques such as Sinusoidal Pulse Width Modulation (SPWM) and Selective Harmonic Elimination (SHE). These techniques help achieve higher efficiency, significantly reduce harmonic distortion, and enhance the overall reliability of the inverter. This innovative approach contributes to improved energy efficiency and supports the development of smarter, more environmentally friendly power systems. The inverter is highly suitable for integration into solar energy systems, offering a stable and clean AC power supply for both residential and commercial applications. Its modular architecture also allows easy scalability to meet varying load demands and future upgrades.
Authors - Aditya Waradkar, Anagha Galagali, Niha Solkar, Shloka Suvarna, Aparna Bannore Abstract - Urbanization has resulted in a high rise in the use of vehicles, thus increasing parking problems like extended search times, fuel consumption, traffic congestion, and user frustration. To counter these problems, this paper introduces ParkSense, an IoT-based smart parking system that combines hardware and software elements for real-time parking space monitoring and management. It uses NodeMCU microcontrollers and IR sensors for car presence detection and an LCD display for real-time on-site updates. It has connectivity with ThingSpeak cloud to provide remote data access and visualization. The frontend is built with the MERN stack (MongoDB, Express, React, Node.js), and the Tailwind CSS provides a user-friendly and responsive interface on devices. ParkSense functionalities include real-time slot monitoring, access to historical data, administrative dashboards, and secure online payments. The system has proven to be highly efficient, reliable, scalable, and easy to use during testing and implementation. It saves considerable parking search time and fuel consumption, thus helping to create a more sustainable city environment. Future developments involve AI-based predictive analytics, dynamic pricing, personalized recommendations, and integration with EV charging stations.
Wednesday August 26, 2026 3:30pm - 5:30pm IST Virtual Room EGOA, India
