Authors - Reena (Mahapatra) Lenka, Jaya Chitranshi, Vanishree Pabalkar Abstract - Artificial Intelligence (AI) has undoubtedly emerged as an extremely dynamic and powerful tool in the area of IT. It is currently handling complex-work in data-analysis, working through predictive modeling, showcasing high level of capabilities and supporting the function of strategic decision-making. AI is dependent on real-time data to identify patterns of attrition, understand dissatisfaction and predict future exits. An AI system in the area of HRM, would thus help the organization in filtering staff-retentions and impending attrition. Meaningful insights can be developed with the use of AI that will help organizations work on reducing employee-turnover on one hand and engaging with their workforce in the long-run, on the other. The innovation elaborated in the study, connects with forecasting the employee-attrition and retention strategies in human resource management. It makes use of AI (Artificial Intelligence) through which HR data sources are incorporated to analyze performance evaluations, engagement surveys, attendance records and demographics in real time and historical context. The machine learning embedded in the system will detect patterns that show evidence of turnover and associate attrition scores to each employee, then recommend targeted retention strategies through personalized career guidance, modification of the workload, and incentives. The system incorporates attrition feedback loops to heighten prediction accuracy and refine strategies over time, enabling the organization to control and reduce turnover rates, boost workforce stability, and achieve cost savings. It is scalable across verticals like corporate HR, healthcare, education, and retail, where talent retention is imperative.
Authors - Reena (Mahapatra) Lenka, Jaya Chitranshi, Vanishree Pabalkar Abstract - A person's ‘perception’ and ‘consumer behaviour’ regarding mutual funds both are influenced by a variety of factors, including socioeconomic characteristics of a given society, financial awareness, risk tolerance, and prior investing experiences. As a professionally managed investment choice and a convenient investing tool, mutual funds are particularly well-liked among middle-class and urban individuals. In contrast to traditional savings instruments, investors typically connect mutual funds with advantages such as access to liquidity, diversification opportunities, and the potential for higher returns. Additionally, perceived dangers, market volatility, and a lack of thorough understanding of financial instruments all have a significant impact on consumer behaviour. A mutual fund is a professional system that collects funds from different investors for investment and protection. Since shared reserves have no legal meaning, the term applies as ambiguously as possible to aggregate speculation that is controlled, accessible, and open to investors. Mutual funds enjoy strengths and weaknesses instead of putting resources directly into personal protection. Today, they represent a significant portion of household budgets. Therefore, the current review focuses on general asset-related buyer behavior and preeminent mutual fund companies. Information was gathered from important resource sources. Important information was collected through systematic research. An opportunity-sampling method was used to collect responses, and the process was targeted toward major Indian cities. This review provides information on donor mindfulness of communal property, donor knowledge, donor propensity, and communal property sufficiency. Ideas were also developed to enhance mindfulness of joint assets and measures to select appropriate common assets to increase profit.
Authors - Jeevesh Sharma Abstract - Blockchain technology is largely used in banking, but it also has applications in gaming, real estate, supply chain management, and healthcare. By 2023, digital money will be the most widely discussed blockchain application. The potential uses of blockchain technology concerning various facets of any sector, market, agency, or governmental organization have gained attention in recent years. Blockchain scalability analyzes the effects on the security of scaling blockchain networks to support more transactions per second. This innovative distributed peer-to-peer architecture drew the interest of companies and communities outside and inside the financial sector. Furthermore, the system it operates in has been created around numerous scenarios that address the trust issue in open networks without the requirement for a trustworthy third party. Even though its decentralized structure allows for a wide range of potential applications, scalability remains a hurdle. Function extension, excessive delay in confirmation, and performance inefficiency are three important areas where blockchain scalability has been hindered. This research paper provides a thorough summary of previous research on scalable blockchain systems.
Authors - Ajidhashini Thulasidass, M. Suresh Abstract - Climate change is making agriculture more challenging in many parts of the country. This is notably true in developing nations where small farmers depend on systems that obtain water from rain to keep their land open. Numerous studies were conducted from 2015 to 2025 to discover how climate change is transforming farming systems in rural regions, how people are responding, and what policies are in place to make these systems safer and more resilient. It looks at significant challenges, including rising temperatures, unclear rain, soil depletion, and more pests that consume food. There is less food, which contributes to reduced food production and declining market stability. One method to make things better is to employ local expertise. Another is to employ agroforestry. Most individuals have problems agreeing because they don't have enough money, technology, or aid from the government. There are tips for extra reading at the end of the essay. To enhance farming, we may employ both new and ancient equipment and processes, as well as long-term research and approaches that engage both men and women.
Tuesday August 25, 2026 9:30am - 11:30am IST Virtual Room CGOA, India
Authors - Reena (Mahapatra) Lenka, Ronak Gupta, Vanishree Pabalkar, JayaChitranshi Abstract - Organizations in the contemporary global economy encounter substantial challenges in managing payroll operations, particularly concerning data security, transparency, and compliance with diverse regulatory standards. Traditional payroll systems rely on centralized infrastructures and are vulnerable to fraud, inefficiencies, and elevated operational costs. These centralized systems pose significant risks, including unauthorized data access and single points of failure, leading to potential data breaches and financial losses. Organizations handling payroll operations in today's international market confront several obstacles, such as protecting data, upholding openness, and adhering to various legal standards. Because they frequently rely on centralized infrastructures, traditional payroll systems are vulnerable to fraud, inefficiency, and excessive operating expenses. In order to solve these problems, this article investigates the integration of blockchain technology into payroll management. We suggest a multi-layered architecture that consists of (1) an off-chain Human Resources (HR) system for payroll and employee management, (2) a distributed storage layer that uses technologies such as the Interplanetary File System (IPFS) for safe data storage, and (3) an on-chain blockchain layer that uses smart contracts to guarantee immutable transaction records and automate payroll processing. This decentralized approach enhances transparency, bolsters security through encryption and consensus mechanisms, and streamlines payroll operations by reducing manual dependencies. Furthermore, the system facilitates real-time cross-border payments and integrates with Decentralized Finance (DeFi) platforms, offering employees innovative financial services. By leveraging blockchain for payroll, organizations can enhance trust, reduce operational costs, and eliminate redundancies, making it a promising use case for HR departments worldwide. This framework demonstrates how blockchain technology can revolutionize payroll management, increasing organizations' efficiency, cost-effectiveness, and trust.
Authors - Shripada Rao, Aadithya Mahesh, Navya Jaideep, Rajeshwari Hegde, Vinay Rao, Saurabh Suman Choudhuri Abstract - This paper introduces a novel approach to integrate LLM capabilities directly on mobile devices to enhance chat applications. By implementing a tonality-driven paraphrasing feature, our system can rephrase poorly written messages into clear, professional text while preserving the intended tone. Unlike conventional server-side AI solutions that raise privacy concerns, our approach processes data locally using fine-tuned models (TinyLlama Instruct 1.1B and Qwen2 0.5B) with parameter-efficient techniques such as LoRA and QLoRA. Experimental evaluations demonstrate competitive paraphrasing quality, improved inference speed, and reduced resource consumption on mobile devices, making this work a promising step toward privacy-preserving on-device conversational assistance.
Authors - Anusha.S.Pujar, Gourishankari.S.P, Saniya.G, Pooja.B.L, Sanchit.H, Amit.N, Suneetha.V.B Abstract - For the purpose of controlling traffic flow, identifying congestion, and averting accidents, contemporary urban traffic monitoring is essential. An enhanced YOLOv5 model is presented in this study for precise vehicle tracking and identification under a variety of circumstances, including day and night. A multi-scale feature detection layer for seeing cars of all sizes in congested regions and an improved pixel-to-real-world distance calibration for accurate speed and distance estimation are two important improvements. Real-time traffic management is improved by integrated collision warning and congestion identification algorithms. Experimental results demonstrate improved detection reliability and mean Average Precision (mAP), making this approach suitable for scalable urban traffic control systems.
Authors - Pratiksha Kulkarni, Rakshita Patil, Veena S Kulkarni, Satish Chikkamath, Suneeta V Budihal, Sujatha Kotabagi Abstract - The major reason for deaths across is due to alarming increase in Heart Disease. There are many factors which elevates the risk of cardiovascular diseases which includes high blood pressure, obesity, cholestrol, smoking habits , lack of physical activities and heavy work pressure. Diagonising and identifying the Heart disease in prior is a challenging task.Which can be overcome by Machine learning methods based on huge dataset of patient traits and medical indicators that help in prediction of heart diseases.
Authors - Dipti Varpe, Gouri Kulkarni, Nishant Thakare, Mitesh More, Suyog Shinde, Navanath Patil Abstract - Traditional farming methods often require extensive manual labour, leading to inefficiencies and increased costs. Recent advancements in agricultural robotics provide innovative solutions to automate essential tasks. The Node-MCU based Solar Powered Multipurpose Farming Robot is an autonomous system designed to enhance farming operations, including ploughing, weeding, and harvesting. Powered by solar energy, it offers a sustainable and energy-efficient alternative to conventional farming practices. The robot operates using a Node-MCU microcontroller, ensuring precise navigation and task execution. Integrated soil moisture and temperature sensors enable real-time environmental monitoring, optimizing farming decisions. Programmed via Node-MCU IDE, the robot is customizable and supports various smart farming features. With its modular design, multiple farming tools can be attached, reducing labour demands and improving efficiency. Additionally, IoT connectivity enables remote monitoring and control through cloud-based platforms. By integrating renewable energy, automation, and IoT-driven sensing technologies, this project enhances agricultural productivity while promoting sustainability. The system lays the groundwork for intelligent robotic solutions in modern farming.
Authors - Harie Sanker V, Gayathri E , Akshay R, Vandana Madhavan Abstract - The human resource management practices have developed alongside technological advancements. AI-based coaching and engagement chatbots are new tools with a great potential to improve employee engagement. Conventional coaching processes have certainly proven their effectiveness but could never attain the scalability or cost-efficiency needed for widespread implementation. Machine learning- and natural language processing-adapted AI solutions can assist by providing real-time feedback, setting goals, and automating HR services based on specific individual need assessments. This research is aimed at understanding the influence of the AI coaching and chatbots on motivation, active engagement, and general employee satisfaction within technological organizations. This study employed primary data collected from questionnaires, finding that AI coaching promotes motivation, satisfaction, and expansion of remote working opportunities. However, lack of trust in AI and perceived ethical absence of transparency on the part of recommendations made by the AI can be seen as significant drawbacks. To overcome these challenges, a hybrid form of human-AI coaching is suggested where AI maximum benefits are retained without missing out on the human touch and consideration that defines coaching. This may facilitate the understanding of the substantial influence that AI has on employee engagement and ultimately on the future of human resource management. The study also suggests how organizations can optimally leverage AI coaching and engagement chatbots while minimizing associated risks.
Authors - Vanishree Pabalkar, Anuja Bokhare Abstract - Sentiment analysis in multilingual social media data is a challenging and critical task due to the diversity of languages and sentiments expressed by users worldwide. In this study, we focus on conducting sentiment analysis on the Twitter US Airline Sentiment dataset, which includes tweets in English from users expressing their opinions about various US airlines. We address the research gap of multilingual sentiment analysis by leveraging advanced NLP techniques and machine learning algorithms. Count Vectorization and TF-IDF Vectorization is used during the study to extract features after cleaning up the data and processing the text. To categorize tweets as having positive or negative sentiment, we assess the effectiveness of three classifiers: Multinomial Naive Bayes, Bernoulli Naive Bayes, and Logistic Regression. We examine these classifiers' accuracy on a different collection of unlabeled tweets without ratings in more detail. The work provides valuable insights into the opinions posted by individuals on Twitter about US airlines and intends to develop multilingual sentiment analysis, especially for social media data. The findings serve as a basis for creating sentiment analysis algorithms that are more precise and reliable and that can be used to various language groups on social media platforms.
Authors - Tanuja R. Patil, Samiksha Dandgall, Vishwanath P. Baligar Abstract - Early diagnosis of brain diseases is very important and challenging nowadays. Detecting neurological disorders such as brain tumors using magnetic resonance imaging (MRI) has become an important research topic. Recently many machine learning and deep learning models have been proposed to detect and classify the brain abnormalities. Many of these models have high time complexity and still efficient models are required. The proposed model makes use of a Novel and Low Complexity Approach to solve the problem of classification of brain images. This approach is a less complexity deep learning model which uses novel methods for Denoising, Segmentation, Feature extraction and Classification of brain tumors. Here, it makes use of the advantages of bit plane approach and a unique feature extraction method. The proposed model makes use of the data set from Kaggle in which, size of the training data set is 2870 with four classes namely No Tumor, Glioma Tumor, Meningioma Tumor and Pituitary Tumor. The size of the testing data set is 394. A feature vector which matches most with the feature vector of the input image is considered as the class of the input image. The proposed method makes use of advantages of time domain and able to give good results. The overall performance of the proposed algorithm considering both training and testing data set is 97.34%. The proposed idea is comparable with the many existing models and the results are compared with three models CNN, VGG19 and Inception-V3 models and found to be promising.
Authors - Arnav Rahul Jade, Jatin Santosh Jaiswal, Nishad Sachin Kamat, Vedant Mahesh Kandarkar, Amruta Pabarekar Abstract - The Cloud-Based Plant Health Monitoring System is designed to help farmers and agricultural experts precisely identify plant diseases using artificial intelligence and cloud technology. Traditional plant health assessments rely on manual inspection, which can be time-consuming and prone to errors. This project automates the process by allowing users to upload images of plant leaves, analyzed by a machine learning model hosted on a cloud platform. The system identifies whether the plant is healthy or has a disease, providing instant results through a simple mobile or web application. To achieve this, the system uses a Convolutional Neural Network (CNN) trained on a dataset of plant leaf images, covering both healthy and diseased conditions. The application is designed to be user-friendly, allowing even non experts to access plant health information easily. This approach reduces the need for excessive pesticide use, saves time, and supports sustainable farming practices by helping users respond to plant health issues promptly. Keywords-plant health monitoring, artificial intelligence, convolutional neural network (CNN), plant disease detection, cloud computing, mobile application and machine learning.
Tuesday August 25, 2026 12:30pm - 2:30pm IST Virtual Room CGOA, India
Authors - Arunabh Barooah, S. Saranya Rubini Abstract - In the context of the digital music industry, accurate classification of music into genres and the ability to recommend appropriate genres for a user greatly improves usability of various streaming platforms. Several traditional machine learning approaches that rely on metadata face significant limitations, such as inconsistencies in the data, the growing diversity of musical styles, and a lack of focus on the actual musical content of songs. These shortcomings often result in suboptimal performance, particularly in recommendation systems. In response to these issues, this work proposes Deep Tune Network (DTN), a deep learning system for automated genre analysis and discovery of music based on similar acoustic patterns. This system uses Convolutional Neural Networks (CNNs) and Mel-frequency cepstral coefficients (MFCCs) to identify the repetitive patterns inside audio signal needed to classify music into different genre. The model achieves a maximum test accuracy of 93.01%, demonstrating its reliability in real-world applications. Additionally, a cosine similarity-based recommendation system is implemented to suggest acoustically similar songs, bridging accurate genre classification with personalized playlist curation.
Tuesday August 25, 2026 12:30pm - 2:30pm IST Virtual Room CGOA, India
Authors - Om Gadhvi, Srushti Pawar, Shravan Gadhvi, Mansi Jadhav, Manya Gidwani Abstract - Farmers in Maharashtra, India face substantial costs and low utilization rates on equipment, leading to significant financial strain. We suggest implementing an AI-powered Dynamic Machine Price Prediction Model that we can use for a digital platform that enables renting out equipment. This model uses Linear Regression, Random Forest, and Gradient Boosting to output what prices equipment should sell for, given the age, how often farmers use it, when they use it, and market demand. Gradient Boosting turned out to be the most accurate model in our exams, giving us a 94% R² score so that our model predictions are trustworthy. The website is built using the MERN stack, and it employs secure transactions through PayPal and a feature that enables farmers to search for equipment based on their location. By adjusting equipment prices on the go, we provide insights to farmers into how much they should charge for renting out their equipment in all conditions The proposed system enhances resource utilization, sustainability, and economic resilience in the agricultural sector.
Authors - Elakkiya R, Sagunthala G, Tanisha Sinha, Gugapriya G Abstract - This project presents an IoT-based predictive maintenance system based on machine learning algorithms—Random Forest, Logistic Regression, SVM, and LSTM—to identify motor faults precisely. Realtime data such as sound, vibration, and RPM are recorded through hardware prototyping, while Simulink simulates speed and torque. Data is transmitted to Firebase for real-time monitoring, prompting automated fault notifications. This system improves industrial efficiency by minimizing sudden failures, reducing maintenance expenses, and increasing machinery lifespan through prompt, data-driven interventions.
Authors - Atharva. Makode, Yash. Kasar, Yatish. Gharat, Yuvraj. Gage, Mandar Ganjapurkar, Kiran Deshpande Abstract - This paper proposes MediLink, a blockchain-based platform to store, share, and manage secure medical records. Patients and care providers in healthcare systems today typically face problems of data privacy compromises, system-to-system noninteroperability, and bureaucratic administration. These present risks to compromised patient care as well as expose data security gaps. MediLink addresses these challenges head-on by developing a decentralized, transparent system that places patients in control, with full control over their medical information while ensuring the integrity and confidentiality of the same. The platform employs smart contracts to carry out important functions such as processing insurance claims, handling patient consent, and keeping track of audit trails. Not only does this reduce human errors between humans but also saves administrative burdens and costs as well. Utilizing the strength of blockchain technology, MediLink not only secures data—yet also streamlines the easy exchange of patient records between healthcare providers. The end result? Seamless care coordination, enhanced patient outcomes, and a smoother experience for all
Authors - D.V.N. Bharathi, K.P.K. Lalitha Vitala, K. Bala Sindhuri, Sai Nikilesh Kudapa, M.L. Hiranya, Lingam Swamy Surendra, Manoj Kumar Juttuka, Kishore Varma Manthena Abstract - This paper explores the design and efficiency of optimized Wallace multipliers integrated with approximate adders to enhance energy efficiency, reduce delay, and minimize hardware complexity in the first address. Five distinct departments are introduced: architectures (AA1–AA5), each offering unique trade-offs regarding power consumption, processing speed, and circuit area. These adders are incorporated into Wallace multipliers, which improve computational speed while lowering energy requirements and design complexity. The proposed designs are evaluated using 90 nm technology to determine their applicability in resource-constrained and error-tolerant domains, such as image processing, machine learning, and IoT applications. The findings demonstrate a versatile balance between accuracy and resource efficiency, making these designs well-suited for real-time systems with different performance demands.
Authors - Deepa Unni, Murale Venugopalan, Sanju Kaladharan Abstract - Community Health Workers (CHWs) does an important role in delivering health to the public. CHW programmes often fail due to the impracticable expectations, lack of proper planning and also the efforts required to implement these activities are often underestimated. Prior to the COVID-19 pandemic, many of the developing countries used digital health technologies to address a range of health issues. Limited research has been held so far to discover the role of e-governance in addressing sustainable community health workforce. This paper bridges this gap by proposing an E-Governance Enabled Sustainability Model for CHWs.
Authors - Aparna Joshi, Moreshwar A. Mahale Abstract - Crop disease diagnosis forms one of the most significant facets of precision farming, which owes a great boost to the power of machine learning. This work presents a new method through which the use of the Tomato Leaf Disease Dataset to classify the tomato leaf's disease is achieved. The database contains 1609 images for ten disease classes: Bacterial Spot, Early Blight, Late Blight, Leaf Mold, Septoria Leaf Spot, Spider Mites, Target Spot, Tomato Yellow Leaf Curl Virus, Tomato Mosaic Virus, and Healthy Leaves. The features are extracted through the combination of Thepade's Sorted Block Truncation Coding (TSBTC) and Haralick Moments (GLCM) to improve texture and intensity description. Extracted features are categorized into multi-level classification (2-ary, 3-ary, 4-ary, 5-ary). The results achieved are stored in an Excel file and re-run using the support of Weka tool where classifiers like Naïve Bayes, Logistic Regression, Sequential Minimal Optimization (SMO), J48, Random Forest, and Random Tree. are employed. The study identifies the optimal model so that accurate and automated diagnosis of crop disease can be performed.
Authors - Prema Sahane, Shreyas Borse, Kartik Narkhede, Manasi Choudhari, Pradnya GaikWad, Ashwini Bhosale, Rutuja Khedkar Abstract - An innovative strategy to address one of the main causes of blindness in diabetic patients is presented in AI-Vision: Forecasting Diabetic Retinopathy for Preventive Care. Through sophisticated predictive modelling, this study uses artificial intelligence (AI) to transform the diagnosis and treatment of diabetic retinopathy (DR). Our approach improves the accuracy of DR diagnosis and makes it easier to identify risk factors that contribute to the progression of the disease by combining cutting-edge Convolutional Neural Networks (CNNs) with extensive medical datasets. Healthcare practitioners may now use individualized preventative tactics based on patient profiles thanks to our cutting-edge model, which uses real-time data analytics to deliver actionable insights. By empowering doctors to intervene promptly, this proactive approach not only seeks to identify DR in its early stages but also lowers the likelihood of serious sequelae .Our research also shows how AI can be used to streamline automated screening processes. AI-Vision hopes to establish a new benchmark in preventative healthcare by bridging the gap between ophthalmology and AI, with the ultimate goal of eradicating avoidable blindness in diabetic populations worldwide.
Authors - Jalindar Gandal, Nilesh Gorade, Kunal Sonawane, Khushal Patil Abstract - Cattle health and productivity are fundamental to the livelihoods of agriculturalists and the overall agricultural economy. Traditional cattle health monitoring methods are often time-consuming and lack the capacity for real-time assessment of cattle health, resulting in reduced milk productivity and economic losses. To address these challenges, we propose the implementation of an Internet of Things (IoT) technology-based, low-cost, real-time cattle health monitoring system. The system comprises wearable sensors for continuous monitoring of vital parameters such as body temperature, heart rate, and activity level. These sensor values are relayed wirelessly to a cloud server, where data is processed and analyzed to identify anomalies indicative of potential health-related problems. This information is presented to the farmer through a user-friendly mobile application, which displays real-time alerts and suggests preventive or remedial actions. The system facilitates early disease detection, leading to improved cattle health, enhanced milk production, and enhanced farm profitability. The system emphasizes cost-effectiveness by utilizing readily available hardware, thereby increasing accessibility for smallholder farmers. The system offers a long-term solution for cattle health management. This paper aims to demonstrate the transformative potential of integrating low-cost IoT technologies with livestock farming to establish precision agriculture and enhance the prosperity of rural farming communities.
Authors - Samadhan Pujari, Hetvi Saroliya, Vedika Gawde, Ekansh Manral, Jalpa Mehta, Deepika Patil, Rashmi Malvankar Abstract - Mortality prediction is crucial for public health, aiding resource allocation, policy-making, and preventive strategies. This study applies ARIMA and SARIMA models to analyze mortality trends in India (1990–2022) using data on infectious diseases such as Malaria, HIV/AIDS, Tuberculosis, as well as non-communicable diseases like Nutritional deficiencies and neonatal disorders. ARIMA [1, 3, 4] captures non-seasonal trends, while SARIMA, incorporating seasonality, proves more accurate. Implemented using Python libraries like pandas, stats models, and scikit-learn, their accuracy is assessed using Mean Squared Error (MSE) and Mean Absolute Percentage Error (MAPE). Findings indicate that SARIMA outperforms ARIMA, emphasizing the role of seasonality in mortality patterns.[6,7,16] A significant decline in deaths from infectious diseases like Malaria and Measles is observed, attributed to public health initiatives, immunization programs, and improved healthcare facilities while neonatal and non-communicable diseases remain pressing concerns. Accurate data collection is essential for improving predictive modeling, and ARIMA/SARIMA provide critical insights for public health planning. Future research could integrate factors such as climate change, economic conditions, and machine learning techniques to refine forecasting models further. This study reinforces the significance of time series forecasting in public health decision-making and strategic healthcare interventions.
Authors - Vaishali S. Katti, Kailash J. Karande Abstract - This paper reviews significant advancements in sentiment analysis, emphasizing various innovative applications in fields such as poetry analysis, human resources, customer feedback, and mental health monitoring. The study systematically examines methodologies adopted in recent research, elucidating their contributions to the field while presenting diagrams to enhance understanding. This overview not only highlights the evolution of sentiment analysis but also explores its implications across diverse sectors.
Authors - Manjunath Inamati, Goutami Naragund, Chetan Paranatti, Saroja Siddamal Abstract - Electrocardiogram (ECG) signals are vital for diagnosing cardiovascular conditions. However, they are often contaminated by noise, hindering accurate analysis. This paper presents the FPGA implementation of Infinite Impulse Response (IIR) elliptic filters for denoising ECG signals. Elliptic filters were chosen for their sharp roll-off and computational efficiency, while an FPGA platform was utilized for real-time, low-latency processing. The design leveraged MATLAB as the primary tool for filter parameterization and hardware-oriented signal processing due to its comprehensive functionality, ease of use, and seamless integration with HDL Coder for Verilog code generation. Synthesis and hardware deployment were performed using Xilinx Vivado. The system was validated using both synthetic and real ECG signals, demonstrating effective noise suppression while preserving diagnostic features. Results indicate the potential of FPGA-based digital filters, designed with MATLAB, for portable and efficient biomedical applications.
Authors - Aswin Sreerag, Meenakshy P, Akhil A, Anargha Ranjit, Anoop S. Babu Abstract - The field of text-to-video generation is going through a transformation, by the advancements in Artificial Intelligence (AI) and deep learning. AI-powered video generation models helps us in the conversion of textual descriptions into visual content, unlocking new possibilities in the fields of education, entertainment, and multimedia content creation. But the existing systems face challenges such as maintaining temporal coherence, such as ensuring smooth transitions, and correctly aligning video sequences with advanced textual inputs. This research combines Large Language Models (LLMs) and Generative Adversarial Networks (GANs) to develop a high-quality, temporally consistent text-to-video generation framework. The system uses the Gemini model to change textual prompts into detailed and structured descriptions. These descriptions are initially given to the Stable Diffusion model to get the corresponding text-image before being input into a fine-tuned MoCoGAN model for video synthesis. The VATEX dataset, which has extensive video and text pairs, is the primary training resource, this ensures meaningful alignment between textual descriptions and generated visuals. Apart from that, the research also explores the DAMO ViLab which is a diffusion model, that operates without additional training, so that it provides a comparative analysis of different generative approaches. The results shows enhanced video smoothness, improved scene consistency, and stronger semantic alignment with the textual prompts. This work advances text-to-video generation by addressing key limitations, using AI driven storytelling and visualization applications.
Tuesday August 25, 2026 3:30pm - 5:30pm IST Virtual Room CGOA, India
Authors - Jyoti Patil Devaji, P. C. Nissimagoudar, Prerana Savant, V.S.Nandana, Vaishnavi Harlapur, Vidhi Agarwal Abstract - The project aims to focus on improving object detection and safety measures for autonomous vehicles by combining LiDAR and camera data. The main goal is to increase object detection accuracy and resilience by combining LiDAR data with the advanced object detection model YOLOv5. The system detects objects, recognizes and tracks cars, and uses a simple depth estimation technique based on bounding box width to determine how far away vehicles stand. To provide more accurate object localization in 3D space, bounding boxes are constructed around identified objects, and the related depth information is computed using the LiDAR data. A safety function that improves the situational awareness of the autonomous vehicle by generating an audio alert when a vehicle is spotted too close is also included in the system.
Authors - Vismay Tank, Khush Sanghavi, Vyom Gandhi, Pramila Shinde, Jalpa Mehta, Vaishali Korade Abstract - Everyone in the current world hopes that his/her personal information won’t be revealed in any way. Security insurance is essential for protecting personal information from prying eyes. The information may be extensive, and it is important to minimize risk and ensure sensitive information is protected. This analysis addresses the drawbacks of previous customized security and other anonymization techniques by implementing a progressive modified protection saving technique. The core of the suggested technique is composed of two main components. Two additional states that are used in the report table but are hidden in the primary segment are sensitive data and fragile weight. The Fragile Data (DI) of the record holder determines whether the mystery should be retained or, conversely, whether it should be disclosed. Sensitive weight (DW) illustrates how brittle a characteristic's value is in comparison to the others. The following section discusses the Recurrence Circulation Block (FDB) and Semi Identifier Dispersion Block (QIDB), two other portrayals used for anonymization. Exploratory findings show that the suggested framework performs faster and loses less information than existing approaches.
Authors - Manas Tiwari, Rohit Sharma, Nyasa Singh, Sandhya Avasthi Abstract - Although packaged foods are generally considered safe and hygienic, many consumers are unaware of the additives they contain, which can pose serious health risks. This lack of awareness has contributed to a rise in health issues such as allergies, asthma, diabetes, and other chronic illnesses. To address this problem, a mobile application is proposed that helps users make healthier food choices by analyzing packaged food ingredients. The application enables users to scan or upload images of ingredient labels, utilizing image preprocessing techniques (grayscale conversion, Gaussian blur) and Optical Character Recognition (OCR) to extract the ingredients. Users can also input personal health conditions like diabetes, asthma, or allergies, allowing the system to tailor its analysis. By referencing a comprehensive database such as OpenFoodFacts, the application provides immediate health-related insights on the detected ingredients. For ease of understanding, ingredients are classified using a color-coded system: red for highly harmful substances, orange for moderately harmful ones to be consumed in moderation, and green for safe ingredients. This approach empowers consumers to make informed, health-conscious decisions regarding packaged food consumption.
Authors - Nivedya Krishnan M T, Mariya P Jose, Amrita V S Abstract - The Financial Independence, Retire Early (FIRE) movement is primarily focused on giving people financial security through saving and investing in such a way that they become less dependent on regular jobs. This empowers them to stop working earlier than regular retirement ages. The study explored major motivations for adopting FIRE through work-life attitudes, desire for freedom, financial well-being, frugality and minimalism, social influence, and spousal/family support. A survey was administered to collect primary data from students and professionals from rural, urban, and semi-urban areas in India, to reach a sample size of 400 respondents. The structured questionnaire includes a 5-point Likert scale, binary, and frequency-based questions. Data were coded and analyzed using SPSS, with regression analysis employed to test the impact of independent variables on motivation to FIRE. Regression results showed that by far the best predictors of motivation for FIRE were frugality and minimalism, desire for freedom, and spousal support. The model was statistically very highly significant (p
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 - 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 - 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 - Aoudumber Londhe, Ravindra Apare, Parikshit Mahalle, Bhagwati Galande Abstract - Wireless communication technologies, particularly those based on IEEE 802.11, have significantly improved connectivity but remain highly vulnerable to Denial-of-Service (DoS) attacks. These attacks, which exploit protocol weaknesses and resource limitations, can severely disrupt network availability, particularly in mission-critical applications such as healthcare, financial services, and industrial control systems. In this research, we investigate various DoS attack techniques targeting IEEE 802.11 networks, including deauthentication flooding, disassociation attacks, authentication request flooding (AuthRF), association request flooding (AssRF), and cascading DoS attacks.To mitigate these threats, we analyze IEEE 802.11w, which provides management frame protection (MFP), and evaluate its effectiveness under different attack scenarios. The model integrates supervised learning for attack classification, unsupervised learning for detecting novel threats, and reinforcement learning for adaptive mitigation strategies. Additionally, the system incorporates IEEE 802.11w security enhancements and anomaly-based behavior analysis to strengthen network resilience. This study provides a comprehensive review of existing DoS attack mechanisms, explores recent mitigation techniques, and introduces an advanced IDS framework to safeguard IEEE 802.11 networks against sophisticated cyber threats. Finally, the analysis is organized through a survey that evaluates the articles based on publication year, research techniques, performance metrics, toolset and utilized database.
Thursday August 27, 2026 9:30am - 11:30am IST Virtual Room CGOA, India
Authors - Vedant Patel, Vidisha Pradhan, Akshita Kadam Abstract - Blockchain technology is transforming the education sector by offering decentralized, secure, and tamper-proof solutions to many of the inefficiencies in traditional educational systems. This paper explores the role of blockchain in lifelong learning, focusing on how it addresses key challenges such as learner autonomy, credential verification, and the creation of secure, decentralized education ecosystems. Through an examination of current developments and case studies—including Blockcerts, Sony Global Education, and Woolf University—the paper highlights blockchain’s applications in academic data storage, personalized learning pathways, and digital credentialing. Additionally, this study discusses the opportunities blockchain provides for improving transparency and trust in the verification of academic credentials across borders. While the technology presents promising solutions, significant challenges remain, including issues of interoperability, privacy, scalability, and legal frameworks. The paper concludes by outlining unanswered questions and future directions for research, emphasizing the need for standardization, privacy-preserving technologies, and scalable implementations to fully harness the potential of blockchain in lifelong learning.
Authors - Umesh Kumar Pandey, Mamta Santosh Nair, Shikha Gupta Abstract - Start-ups are buzzing word around the world. These start-ups need funding in their early stage with a high risk of failure and violation of the innovator's intellectual property. Any system's prime responsibility is to ensure the fund availability to the start-up and save the innovator's intellectual property since blockchain has become the chief technology in digital crypto-currencies. Bitcoin. Blockchain has become popular in finance, the health sector, social services and many more areas where transactions are recorded among the parties, known or unknown—the critical features of block Chainz. Decentralisation, distribution, immutability, transparency and audit-ability enrich the usability of this technology and increase the trust to use it. Therefore, a system is proposed here to manage start-ups utilising blockchain features. The proposed system ensures that parties to the contract have more confidence and feel safe to grow start-ups in the community and prevent unnecessary conflicts.
Authors - Smrity Dwivedi Abstract - This manuscript has oriented to new generation and new technology used for required resources and terms, which gives wide bandwidth for each and everyone’s perception. For this reason, microwave frequency area eight to 10 GHz has been explored for 5G and also 7 to 20 GHz is being explored for beyond 5G. This is why both possibilities were taken right here. First assessment among hexagonal and triangular structure complete floor were designed with CST microwave studio. Results obtained from those designs are -23.35dB for 7.77dBi benefit and -29.103dB for 7.85dBi gain for hexagonal and triangular systems respectively. For enhancing the advantage, a partial ground has been used for triangular structure and 10.5dBi has been completed at -38.65dB S11 and for 9.0988 GHz frequency. Bandwidth is increased from 0.29 GHz to 0.32 GHz. Everything is simulated and analysed by simulation software. Novelty is the simple structure gives beyond 5G applications.
Authors - A.Punidha, E.Arul, E.Yuvarani, S.Rajasakaran Abstract - Understanding how dark patterns influence user sentiment is crucial for developing ethical and user-friendly digital experiences. This study evaluates the performance of XGBoost and Random Forest in predicting sentiment (negative, neutral, or positive) based on user interactions. The models were assessed using accuracy, precision, recall, and F1-score, with results indicating that XGBoost outperforms Random Forest, achieving an accuracy of 88.4% compared to 85.9%. To enhance interpretability, SHAP (Shapley Additive Explanations) was used to break down model predictions and identify the most in-fluential features. The analysis revealed that "Number of Clicks" and "Time Spent on Page" were the strongest indicators of user sentiment, particularly in detecting frustration associated with dark patterns. The results provide valuable insights into how machine learning models interpret user engagement and emphasize the importance of transparent AI-driven sentiment analysis. By leveraging explainable AI techniques like SHAP, this research contributes to improving trust in sentiment classification models and guiding the development of more user-centric digital interfaces..
Thursday August 27, 2026 9:30am - 11:30am IST Virtual Room CGOA, India
Authors - J. Jeslin Shanthamalar, Prateesh Kumar S, Abishin J, Sindhu Chandra Sekharan, Malar Selvi G Abstract - There is a growing necessity for noninvasive and sophisticated diagnostic capabilities with the ability to very early prediction of skin conditions from the patient. Timely diagnosis is a powerful influence on patient care out comes, access to dermatologists is not, especially in rural. We propose an AI and deep learning model for improving classification of skin diseases in a highly accurate advantageous manner. This model, which follows the architecture of Convolutional Neural Network, trained on a multi-class skin disease images dataset where every image has a label per lesion. Through hyperparameter fine- tuning, the model is optimized to achieve performance from metrics that include accuracy and trade-off accuracy vs. precision/recall. With user-friendly access in mind, the model runs into the app (web or mobile) that supports a friendly diagnostic user interface. Advanced security floor work is taken within designed to reduce the effect of adversarial attacks. Multimodal processing (text, image and speech inputs) improves classification substantially resulting in accurate and robust diagnosis. The platform has been built based on healthcare professionals and patients' input to provide a easy-to-use diagnostic tool. Research to edit the ai applications in dermatology through fewer dataset bias, more human like NLP explainable models, as well as ongoing work for improved security. In the end, this system is what makes skin disease detection accessible and fast via AI- determined aids an inclusivity in healthcare.
Authors - Aravind A R, Archa A S, Gouri S Krishna Abstract - With India rapidly embracing digital transformation, initiatives like UMANG by the government are the means to achieve online public services. Although UMANG offers over 1,750 services of many departments, it has some critical accessibility concerns, particularly for differently-abled citizens. In this study, we evaluate the UMANG website for WCAG 2.2 compliance using a two-stage method: automated checking using AccessibilityChecker.org and user review analysis using Appbot. Findings identify prominent issues of poor ARIA labeling, flawed heading order, inadequate color contrast, and improper focus order as hindrances for assistive technology users. Sentiment analysis also indicates frustration with usability, login failure, and performance. For the improvement of accessibility, the present study has some suggestions that make UMANG equivalent to international standards. By making inclusive design central to e-governance, India can enjoy equal access to fundamental digital services, creating an inclusive digital space.
Authors - Thisura S. Wijesekera, Dinuka R. Wijendra Abstract - Kubernetes has become the leading container orchestration platform due to its powerful scalability features, enabling dynamic resource management and efficient workload handling in cloud-native environments. This review examines Kubernetes scaling mechanisms at both the application and cluster levels, focusing on Horizontal Pod Autoscaler (HPA), Vertical Pod Autoscaler (VPA), and event-driven scaling with KEDA for adaptive application scaling. At the cluster level, Cluster Autoscaler (CA), Karpenter, Cluster Proportional Autoscaler (CPA), and Cluster Proportional Vertical Autoscaler (CPVA) optimize node provisioning and resource allocation. Despite these advancements, challenges persist, including reactive scaling delays, resource fragmentation, inconsistent scaling decisions across multiple autoscalers, and security vulnerabilities like Economic Denial of Sustainability (EDoS) attacks. To address these issues, emerging trends in AI-driven observability, predictive analytics, and unified autoscaling frameworks offer proactive scaling, anomaly detection, and self-healing capabilities. This review synthesizes academic research and industry practices to highlight the current state, challenges, and future directions of Kubernetes scalability, emphasizing the need for intelligent, adaptive, and secure scaling solutions.
Authors - Suruchi Pandey, Hemlata Gaikwad, Yograj Ingale, Neha Sharma Abstract - This article looks at how organisational culture, resource allocation, and decision-making procedures reflect sustainable leadership principles in different settings. A comparative investigation shows that military commanders place a higher priority on mission success and national security than do business executives, who place more emphasis on profitability and shareholder value. Nonetheless, there are similarities between the two fields, including the value of making moral decisions, flexibility in the face of change, and an emphasis on long-term goals. The role of innovation in sustainable leadership is also examined in this article, with particular attention paid to how strategy development and technology support organisational resilience. Additionally, it looks at how sustainable leadership affects worker engagement, emphasising how crucial it is to develop a feeling of dedication and purpose. This article seeks to provide a more comprehensive knowledge of successful leadership techniques by exploring the subtleties of sustainable leadership in business and defence environments. Regardless of the particular difficulties they encounter, executives looking to im-prove the sustainability and resilience of their organisations may gain a great deal of insight from identifying the parallels and variations across these industries.
Authors - G. Ram Sundar, Sindhu Chandra Sekharan, Taruni Mamidipaka, Yoga Sreedhar Reddy Kakanuru, Priyadharshini M Abstract - The traditional sarees in India represent a rich history both culturally and artistically, as their patterns are created from regional influences along with modern fashion. Making sarees requires detailed skill, and traditional techniques are highly laborious and time-consuming. In this research, we have developed Vastra Kalpana, an AI-driven saree design generator that uses generative deep learning models to automate textile pattern creation. Users can now provide voice commands, and they are converted to text prompts by our integrated OpenAI Whisper speech-to-text software, which are then transformed into structured textual descriptions. These descriptions serve as instructions for the Stable Diffusion's high-resolution saree design generator. Our research results suggest that this automated saree design generator is both solution oriented and efficient, proving the generative techniques offer a novel approach for saree design while tackling the challenges of overreliance on handmade designs. This research mainly contributes to the field of fashion design and establishes a framework for future advancements in automated textile pattern generation.
Assistant Professor, Department of Computer Science and Engineering (Artificial Intelligence & Machine Learning), Vishwakarma Institute of Technology, Pune, India
Thursday August 27, 2026 12:28pm - 12:30pm IST Virtual Room CGOA, India
Authors - Prranjali Jadhav, Varsha H Patil Abstract - Recommendation systems play a crucial role in personalized content delivery across various domains such as e-commerce, streaming platforms, and healthcare. This survey presents a comprehensive analysis of recommendation frameworks, emphasizing their architectures, methodologies, challenges, and future directions. The provided framework integrates hybrid models, deep learning, collaborative filtering, and content-based filtering, processed through a multi-layered architecture. The data processing layer handles preprocessing, feature extraction, and data collection, while the model selection layer chooses an appropriate recommendation technique. The recommendation engine ranks and scores predictions before delivering final recommendations. A critical component is the user feedback & continuous learning module, incorporating explicit and implicit feedback to dynamically update the model. Challenges such as scalability, data sparsity, and real-time adaptation are explored, along with emerging advancements like knowledge graphs and reinforcement learning. The paper highlights future research opportunities to enhance recommendation accuracy and user experience.
Authors - Amit Budhodkar, Rupali Umbare, Nihar Ranjan, Shubham Udgirkar, Sakshi Suryawanshi, Pradnya Aher Abstract - As mock interviews are essential for job interview preparation, the resources available cannot evaluate both technical and non-technical skills. This paper outlines the AI Mock Interview Platform which interfaces with learners and simulates truthful interviews by assessing non-technical competencies such as body language, confidence, emotional expression, and technical knowledge. The platform is capable of dynamically generating interview questions relevant to a candidate's particular role using AI technologies. In addition, AI technologies enable real-time feedback provision. With regard to feedback generation, the system utilizes AI technologies to consider specific features unique to the candidates’ voices, movement, and gaze direction. It utilizes Dlib’s human body posture detection library and video sentiment analysis for facial expression recognition with AffectNet dataset for Convolutional Neural Network faces as well as videos. With the Courses feature, learners can focus on varied topics and the platform automatically selects suitable instructional content consistent with the candidates’ preferred method of learning
Authors - SHALINI S, C NANDINI, LAKSHMI MR, KOUSTAV BISWAS, L DIVYASHREE, MITAYI AJAY KUMAR, MONIKA V Abstract - This research work uses Artificial Intelligence for early detection and targeted intervention in the case of Autism Spectrum Disorder (ASD). Through sophisticated language analysis and pattern identification of interactions, the platform detects signs of autism to facilitate early intervention. Relying on these findings, the platform tailors developmental programs in pivotal areas of communication, daily living, and adaptive learning through fun, interactive modules. An integrated chatbot powered by AI improves user experience through conversational assistance, responding to questions, and assisting individuals with autism, as well as their caregivers. Ongoing interaction develops a greater familiarity with the resources available on the platform and encourages active involvement in skill development exercises. Structured with users from every age group, the platform places strong emphasis on ethical use of AI and protecting data, offering a secure and reliable environment. Through its fit to the singular developmental path of each user, it fosters autonomy, skills development, and social integration. The platform is an integrated system of care and empowerment for the autism community. It seeks to respond to the broad range of individual needs on a universal, adaptable, and empathetic level, facilitating personal development and increased autonomy for individuals on the spectrum.
Authors - Abhay Pratap Singh, Aanya Mittal, Ashmit Tyagi, Kanan Agrawal, Avdhesh Gupta Abstract - Communication is one of the major attributes of human life[1]. The system discussed in this research paper focuses on developing a novel and efficient way of communicating between a deaf-mute person and any other person who is normal (does not have deaf or dumb handicaps). Advanced technologies used in the design will support the conversion from voice to Indian Sign Language using Natural Language Processing Machine learning algorithms and computer vision techniques and vice versa. It translates audio messages into sign language images with text in real-time, trying to basically eliminate the conventional dependency on interpreters as a means of communication for every person. The main idea of the research is the solution of urgent problems connected with communication of the deaf-mute people and, at the same time, to be able to solve this task with the use of modern technologies, keeping in mind the principles of inclusiveness and independence. The design, implementation, and potential of the system to improve the living standards of deaf-mute people by bringing them closer to society are discussed. The aim is to plead for inclusion by raising the awareness of educators, policymakers, and the public at large regarding the demand for communication resources addressed specifically to the deaf and mute community. The consciousness of the demand for ISL interpreters and the promotion of video datasets will be helpful in bridging the gap in communication, as seen in the research on the lack of certified ISL interpreters and the demand for automated sign recognition systems.
Thursday August 27, 2026 12:30pm - 2:30pm IST Virtual Room CGOA, India
Authors - Debabala Swain, Monalisa Swain, Sharmistha Roy, Debabrata Swain, Jayanta Mondal, Prachee Dewangan Abstract - The information stored or transmitted digitally is vulnerable to unauthorized access. The authentication of digital images is a critical issue in the era of digital advancements, given the ease with which any image can be altered. Consequently, methods for verifying the credibility of images are gaining widespread recognition due to their relevance in various societal domains, such as government, military, forensics, and electronic commerce. The significance of protecting images from manipulation has escalated, recognizing that even a minor tampering incident could lead to severe consequences. Hence, safeguarding images from alterations has become increasingly essential. Literature has seen the development of numerous approaches to ensure the genuineness and integrity of digital images. This study offers a comprehensive overview of both domain-based and AI-based watermark techniques for authenticating images, providing the capability to detect tampering and pinpoint the specific manipulated areas within an image.
Authors - Pooja Singh Chaudhary, Nirav Bhatt, Purvi Prajapati Abstract - Object Detection and the Object Pursuit are the fundamental and the emerging tasks in the Machine learning and in the computer vision to detect the object and then too track the object in all the real and the dynamic environments. The latest trends which are emerged in this area, highlighting the embedding of deep learning techniques has transformed the field of object detection and tracking. Methods like Convolutional Neural Networks, Deep SORT, You Only Look Once and Region-Based Convolutional Neural Networks have significantly improved accuracy and efficiency. We examine the shift towards more robust and measurable and the scalable solutions, with particular focus on multi-object tracking, real-time processing, and handling challenging Challenges like occlusion, variations in scale, and varying in illumination. The survey also addresses key challenges that remain, including computational efficiency, accuracy in complex scenarios, and the development of algorithms. Furthermore, we discuss the applications of object detection and pursuit across industries like autonomous driving, robotics, surveillance, and augmented reality, while offering insights into future research directions that may overcome existing limitations and drive the field forward. These recent advancements, combined with the evolution of tracking algorithms, have made it possible to detect and track objects in real-time with high precision.
Authors - Akhil K J, Saurabh Shrivastava, Harish R Abstract - The increasing concerns over online privacy and the growing prevalence of internet censorship have driven many users to seek greater anonymity through tools like proxies and virtual private networks (VPNs). While peer-to-peer (P2P) networks provide a decentralized way for users to communicate securely across multiple nodes, they are not immune to security threats. One of the major vulnerabilities in P2P networks is the risk of man-in-the-middle (MITM) attacks, where malicious actors intercept communication between nodes. In these attacks, attackers can manipulate, inject, or even remove data being transmitted, compromising the integrity of the information. Another rising threat within these networks is cryptojacking—a tactic where attackers surreptitiously use a website’s resources to mine cryptocurrency, often without the knowledge or consent of the website visitors. This malicious practice has gained attention due to its increasing prevalence on popular sites. In the context of P2P networks, the exploitation of exit nodes poses a significant risk, as attackers can inject mining scripts into the HTTP responses sent from these nodes. These risks highlight the need for robust security protocols to safeguard decentralized networks and prevent malicious interference, ensuring the security, privacy, and integrity of online communication systems. Effective measures are vital to protecting users and maintaining trust in these technologies.
Thursday August 27, 2026 12:30pm - 2:30pm IST Virtual Room CGOA, India
Authors - Kamini Solanki, Nilay Vaidya, Jaimin Undavia, Krishna Kant, Jay Panchal, Anjali Mahavar Abstract - The rapid growth of internet-based applications, such as social media platforms and blogs, has led to an increase in comments and reviews about everyday activities. Sentiment analysis involves collecting and analysing people's opinions, thoughts, and perceptions on various topics, products, services, and subjects. These opinions can provide valuable insights for businesses, governments, and individuals in making informed decisions. However, the process of sentiment analysis faces several challenges that make it difficult to accurately interpret sentiments and determine the correct sentiment polarity. Sentiment analysis extracts subjective information from text using natural language processing (NLP) and text mining techniques. This article provides an in-depth overview of the methods used to perform sentiment analysis, along with its applications. It also evaluates and compares different approaches, discussing their advantages and limitations. Finally, the article examines the challenges in sentiment analysis and proposes future directions for the field. Sentiment analysis, also referred to as opinion mining, is a vital area of research in natural language processing (NLP) that focuses on identifying the sentiment expressed in text. This paper reviews various sentiment analysis techniques, explores its broad range of applications, and discusses the challenges within the field. The goal is to provide a thorough understanding of the current state of sentiment analysis and its potential future developments.
Authors - Botcha Divya, Yelavarti Kalyan Chakravarti, V. Esther Jyothi, A. Satya Kranthi Abstract - Software-Defined Networking (SDN) has transformed contemporary network topology by separating the control plane from the data plane, allowing the network to be centrally and dynamically managed. Its central design, however, also presents enormous security threats that must be mitigated using efficient Intrusion Detection Systems (IDS). This paper proposes an intelligent IDS framework for SDN networks utilizing machine learning algorithms. The proposed method employs the UNSW-NB15 dataset, preprocessing with advanced methods, SMOTE-Tomek resampling, and multi-class classification by XGBoost for attack detection and classification of different attacks. Interactive Streamlit-based dashboards and packet simulation allow real-time observation, filtering of attacks, and visualization of anomalies in detail. Experimental results demonstrate enhanced detection accuracy of 84% using the top 20 features selected that outperform conventional classifiers in precision and responsiveness. The addition of real-time prediction counters, attack distribution graphs, and downloading capability allows for tremendous flexibility when used in live SDN contexts. The project tries to minimize the theoretical/practical implementation gap found among existing IDS models and live deployments with its suggested scalable, interpretable, and effective intrusion detection solution.
Authors - Deepesh Sudhan Arunachalam, Dennis Andrew, K. S. Gayathri, A. Shahina, V. Durgadevi, A. Saravanan Abstract - This work introduces a deep learning-based framework for 3D pressure mapping to assess sleep quality and body strain. 2D pressure maps suffer from loss of depth information, poor spatial context, posture misclassification errors, and limited accuracy in capturing regional pressure variations. To overcome these limitations, the framework constructs 3D pressure maps that enable precise region-wise pressure estimation with anatomical landmarks to analyze body strain. Sleep quality is monitored by tracking frequent posture changes with converting pressure maps to point clouds achieved 99.26% accuracy with PointNet and 99.49% with PointCNN.
Assistant Professor, Department of Computer Science and Engineering (Artificial Intelligence & Machine Learning), Vishwakarma Institute of Technology, Pune, India
Thursday August 27, 2026 2:30pm - 2:32pm IST Virtual Room CGOA, India
Authors - Yash Tekade, Mayur Shinde, Bhumika Lipane, Nikita Patil, Suhasini Bhat Abstract - The paper presents the idea and methodology of development of a real-time threat detection system designed to enhance women's safety across various environments using AI technology and CCTV surveillance. The system consists of features like real-time person detection, gender classification, and SOS gesture recognition, all connected to an alert system for law enforcement authorities. It effectively identifies potential threats, including a lone woman at night or a woman surrounded by men, enabling proactive actions before incidents escalate. Additionally, the system maps hotspot areas where previous incidents have been recorded, allowing authorities to allocate resources efficiently. It also alerts security personnel about low-light conditions in an area, ensuring surveillance even in challenging environments. By combining these capabilities, the system aims to create a safer atmosphere for women, promoting proactive measures that can significantly reduce crime rates and contribute to enhance overall safety strategies.
Authors - Ritveek Rana, Manisha Manoj,vAnitha Dhanasekaran Abstract - This research endeavors to apply artificial intelligence to estimate past energy statistics and forecast future energy consumption patterns in India. The research utilizes energy indicators such as access to electricity, the share of renewable energy, CO2 emissions, and economic development to develop a model to forecast future energy needs and renewable energy share. The future energy consumption patterns and the share of renewable energy are forecast using regression analysis. The intention is to provide insights into energy transition required in order to ensure sustainability by reducing the reliance on fossil fuels and increasing renewable sources.
Authors - G.B. Sambare, Sankarsha Shelke, Sahil Wawdhane, Harshad Wable, Abhinav Thube Abstract - The KYC Powered by Blockchain for decentralized, secure, and more efficient Know Your Customer (KYC) system using blockchain. This system solves the inherent inefficiencies of traditional KYC by allowing institutions to share validated customer data, mitigating redundancy among KYC providers, and reducing both costs and compliance time. Tamper-proof architecture of blockchain allows for strong data privacy, security, and compliance of AML and GDPR regulations. Customers gain full control over their personal data, with the ability to grant and revoke access dynamically, reducing risks of breaches and fraud. The framework integrates off-chain storage for sensitive data and combines advanced cryptographic methods like AES and ECC for encryption and security. Smart contracts automate data handling and permissions management, ensuring secure, transparent, and immutable data sharing across institutions.
Authors - Sakshi G. Wagh, Snehal S. Shirsath, Vaibhavi V. Pujari, Shrirang A. Sonawane, Milindkumar B. Vaidya Abstract - Diagnosing brain tumors is a complex task due to their intricate characteristics and variability in presentation. Timely and accurate detection plays a vital role in ensuring effective treatment and improving patient prognosis. This study presents the development of an automated system for brain tumor detection and segmentation using Convolutional Neural Networks (CNNs). The model is trained on annotated MRI datasets to distinguish between normal and tumorous brain tissues with high accuracy. The proposed approach involves a comprehensive pipeline that includes image preprocessing to enhance MRI quality, training a CNN-based model for tumor recognition, and applying post-processing techniques to refine the output. By automating the diagnostic process, the system aims to support radiologists by increasing accuracy, reducing diagnostic delays, and minimizing manual interpretation errors. Furthermore, the project incorporates various image processing techniques and data augmentation strategies to strengthen the model’s performance and generalizability across diverse imaging conditions. The result is an intelligent and accessible diagnostic tool intended to assist healthcare professionals in delivering more precise and efficient brain tumor diagnoses, ultimately contributing to better clinical decision-making and patient care.
Authors - Monali P. Deshmukh, Dhanashri Arjun Ghadage, Mrunali Sunil Rangankar, Prajakta Dattatray Supugade, Deep Isane Abstract - This research paper presents an AI-assisted web-based coding platform, Code Understanding using Sherlock, that integrates a real-time compiler with an AI-powered chatbot. The chatbot provides contextual assistance based on selected code snippets or general programming queries. Users can toggle between a standard chatbot mode and a code-aware mode, where the chatbot analyzes selected code portions to answer relevant questions. The system enhances the coding experience by providing explanations, debugging help, and execution functionalities. By leveraging AI and NLP techniques, the chatbot can understand syntax, logical structures, and common programming errors, offering detailed feedback and solutions. The platform streamlines the development process by reducing debugging time and enhancing code comprehension. Additionally, the system provides a seamless file management experience, enabling users to create, edit, and organize their projects efficiently. This integration fosters an interactive learning and development environment, making it valuable for both beginners and experienced programmers.
Authors - Anandhukrishna A S, Santanu Mandal, Raghu Raman Abstract - Digital Twin (DT) technology offers unprecedented capabilities that are transforming supply chain management (SCM), delivering a new level of system-wide resilience, real-time insights and predictive analytics. Yet, the nascent research still lacks in terms of cohesion, with most studies being heavily centralised around technical solutions and missing strategic, managerial and empirical perspectives. In light of this gap, the current study provides a thorough bibliometric analysis of 99 peer-reviewed articles published from 2016 to 2024 selected from Scopus with analytical tools of Biblioshiny R package. The results clearly showed that there was a higher growth of DT-related SCM research after 2020, indeed due to significant intercontinental disruptions and the demand of resilient, sustainable, and intelligent infrastructure systems. Resilience in the supply chain, sustainability, interoperability, and AI-driven optimization are core themes. Importantly, while China, Germany, and the USA dominate in terms of number of papers produced, institutions such as The Hong Kong Polytechnic University are also leading in productivity metrics here. However, the analysis reveals important gaps — notably a lack of cross-border cooperation and empirical case studies as well as longitudinal research. Less developed but rich prospects, like the integration with blockchain, extended reality and physical internet also emerge as compelling themes. This work offers actionable insights into the way forward for researchers, policymakers, and industry leaders, calling for cross-disciplinary partnerships, real-world pilots, and frameworks for applying overarching compliance. This also advances the role of Digital Twins as a strategic enabler of a resilient and future-ready supply chain.
Thursday August 27, 2026 3:30pm - 5:30pm IST Virtual Room CGOA, India
Authors - Jhalak Bansal, Janvi Jain, Sukti Jain, Harsh Chaudhary, Vikas Srivastava Abstract - Traffic accidents, a leading cause of death worldwide with nearly one million fatalities annually (WHO), are often driven by fatigue-related drowsiness. Our project introduces a real-time drowsiness detection system leveraging technologies like OpenCV, Python, and machine learning to enhance safety and accuracy. Using a camera, the system monitors facial features and eye movements, Using facial landmark detection to identify 68 key points, the system calculates the Eye Aspect Ratio (EAR). Extended periods of eye closure activate an alert, and GPS-enabled location tracking enhances response by sending automated emails with the vehicle’s real-time location to pre-registered contacts. The methodology integrates image processing, real-time facial landmark detection, and a dynamic scoring system to evaluate drowsiness. With an accuracy target of over 85%, the system addresses the limitations of existing solutions while introducing innovative location-based intervention. Results highlight its potential to reduce drowsy driving incidents, ensuring safer roads.
Authors - Aoudumber Londhe, Ravindra Apare, Parikshit Mahalle, Ravindra Borhade Abstract - Aqua status quality prediction is a vital part of environmental monitoring, with significant implications for public health, ecosystem sustainability, and Aqua resource management. Traditional methods for evaluating aqua quality, is like taking the manual sample and to perform the laboratory analysis, are often labour-intensive and limited in scope. Recent developments in deep learning have transformed this domain by empowering the expansion of predictive models accomplished with analysing non-linear relationships in Aqua quality. Hybrid deep learning models, merging Recurrent Neural Networks (RNNs), Long Short-Term Memory (LSTM) networks, Convolutional Neural Networks (CNNs), and Gated Recurrent Units (GRUs), have verified superior performance in apprehending spatial and temporal dependencies in Aqua quality data. Optimization algorithms such as Particle Swarm Optimization, Grey Wolf Optimization, Sparrow Search Optimization (SSO), and Beluga Whale Optimization (BWO) have been integrated to enhance model accuracy and efficiency. Attention mechanisms and feature selection techniques have further improved model performance, while the integration of IoT has enabled real-time monitoring, addressing the limitations of traditional methods. Despite these advancements, challenges related to model interpretability, computational complexity and most important part data availability remain as it is. This review explores the pragmatic augmentation in hybrid deep learning models for Aqua quality prediction, focusing on their architecture, optimization techniques, and real-world applications.
Thursday August 27, 2026 3:30pm - 5:30pm IST Virtual Room CGOA, India
Authors - Rakhi Bharadwaj, Mohit Deo, Pratham Jain, Ashishkumar Jha, Harsh Bachhav Abstract - This study introduces a novel open-source educational website that makes use of AI-powered 3D environments and interaction with historical individuals to deliver immersive historical learning experiences. The site offers both contemporary views of these environments using 3D Gaussian splatting technology and offers precise historical recreations using Pixel Streaming. Interactive conversation with AI-powered historical individuals, dynamic quizzes to validate the knowledge of users, and AI-powered historical narratives are all among the offerings. To support knowledge about historical events and cultures from the past, the system merges interactive learning and storytelling for a fun and educational experience. Advanced natural language processing (NLP), speech-to-text, and AI-powered tour guides are all included as part of the platform architecture to provide personalized historical tours without necessitating complicated personal details.
Authors - A. Harshavardhan, Konkathi Nihal, Ramini Srinidhi, Konda Poojithasai, Gochika Bhanu prasad, Dhanraj Sai Ganesh Abstract - This paper presents a novel, dual-layer secure steganographic system that combines hybrid cryptography and steganography to ensure the confidentiality, integrity, and security of secret communications. Initially, the sender inputs their message and selects a cover image. The message is encrypted using a hybrid substitution (playfair cipher and columnar transposition cipher) and transposition cipher, and then embedded in randomly selected pixel positions of the image using Least Significant Bit (LSB) steganography. A position file that records these embedding locations is generated and encrypted. To obfuscate the presence of the stego-image, multiple duplicate images are created alongside the steganographic image. On the receiver's side, a ResNet50-based feature extractor followed by K-means clustering is used to identify the stego-image from the duplicates. The encrypted position file enables accurate message extraction and subsequent decryption. Experimental results show excellent performance with high imperceptibility (MSE: 0.0175, PSNR: 65.69 dB, SSIM: 0.9994) and strong resilience to brute-force and statistical steganalysis.
Thursday August 27, 2026 3:30pm - 5:30pm IST Virtual Room CGOA, India
