Authors - Uttam Patole, Manish Shrivastava, Archana Ugale, Abhishek Deshmukh Abstract - Poor water use and late illness detection force farmers to struggle, which hurts their crops and squanders resources. This study presents a computerized precision farming system combining machine learning, humidity sensors, and robot automation to improve irrigation and disease control. The technology operates in two phases: first, humidity sensors calculate the ideal watering amount by assessing soil moisture and applying a Random Forest model. Second, disease detection sensors identify agricultural diseases and forecast disease outbreaks using a Gradient Boosting Regressor (GBR), thereby guiding robot pesticide spraying. By integrating machine learning with real-time environmental monitoring, the technology minimizes human interaction while assuring exact irrigation and tailored pesticide administration. This strategy avoids the usage of pesticides and excessive water consumption while enhancing resource efficiency and crop health. The proposed system is scalable and adaptive to varied farming situations, making it a good alternative for modern precision agriculture. By combining predictive analytics and automation, the model enables data-driven decision-making in farming, supporting sustainable agricultural practices and boosting overall output.
Authors - Rahul Dhaigude, Ruby Chanda, Shrikant Ghadge Abstract - The healthcare industry is a complex and vital sector that plays a fundamental role in society, focusing on the well-being and health of individuals. In recent years, the integration of Artificial Intelligence (AI) into hospital logistics has emerged as a transformative force in the healthcare industry. This study investigates the application of AI in hospital logistics to enhance tracking and overall operational efficiency. A questionnaire was designed to gather quantitative and qualitative data on various aspects of hospital logistics, AI integration, challenges, and potential improvements. Qualitative data from open-ended questions and interviews were analyzed thematically to extract key themes and insights. The results show that AI has become an essential tool for optimizing logistics processes, improving patient outcomes, and enhancing the overall patient experience. However, careful planning, implementation, and consideration of data privacy and staff readiness are critical for successful AI adoption in hospital logistic.
Authors - T. Sruthi, Sheshikala Martha Abstract - Extracting meaningful knowledge & actionable comprehensions from complex, multi-dimensional, and heterogeneous biomedical data residues a significant challenge in healthcare. Modern health care systems generate various types of medical data that are often intricate, diverse, and typically unstructured. These large datasets are often difficult to interpret and process. Traditionally, data mining techniques have been employed to extract features from such data, with prediction or clustering models built on top of those features. However, this approach faces numerous challenges, particularly when dealing with complex data and limited domain expertise. Recent advancements in deep learning, however, have introduced new and effective methods for building learning models from these complex datasets. In this paper, we discuss various clinical data types and their relevant features that can serve as inputs to deep learning net-works, contributing to the creation of a more reliable and sustainable healthcare system.
Authors - T. Aruna, Dhanya Kulkarni, Riya Javali, Rakshan Kulkarni, Suneeta V. Budihal, Shamshuddin K Abstract - In 5G networks, dynamic slicing is a major improvement, which makes it possible to allocate specific resources to satisfy the various quality-of-service (QoS) requirements of applications including ultra-reliable low-latency communications, enormous IoT, and enhanced mobile broadband. However, managing these slices in response to dynamic and varied traffic patterns requires real-time flexibility, which poses considerable hurdles. The methods for effective dynamic network slicing are examined in this research, with an emphasis on resource allocation optimization, QoS adherence, resource waste reduction, and network stability. To accommodate upcoming developments, the suggested solutions seek to improve 5G network performance, scalability, and adaptability.
Authors - Ajay Talele, Madhav Jagtap, Radhika Gadewar, Akanksha Katore, Sufiyan Sajan, Deepika Sidral, Yash Shinde, Gaurav Desale, Saburi Nikam, Shubham Landge, Chetan Channa Abstract - Farm trade is a key component of economic viability, but farmers usually struggle with issues of market access restriction, price volatility, and dependency on middlemen. This article introduces a Java-based mobile market that is capable of empowering farmers by creating direct links to customers, thus ensuring transparency and profitability. The new platform incorporates real-time price feeds, demand forecasts, and electronic secure transactions to form a highly efficient and consumer-friendly trading system. Further, the system incorporates IoT-based weather updates and agricultural advisory services to enable farmers to make appropriate decisions. The system utilizes a recommendation engine employing machine learning to maximize prices and predict demand trends. The system is highly secure because transactions are encrypted and involves a strong mechanism for user verification. Fair trade is promoted while minimizing post-harvest losses; the marketplace helps farmers make electronic payments, which enhances financial inclusion and sustainable agriculture. System analyses and case studies attest to its capability to revolutionize agricultural commerce, close the urban-rural digital divide, and enhance farmers’ economic performance.
Tuesday August 25, 2026 9:30am - 11:30am IST Virtual Room BGOA, India
Authors - Hemender Sai, Bipin Sai Bhaskar, P.Saranya Abstract - Tuberculosis continues to be a major global health concern, particularly in remote regions with limited access to healthcare. Early and precise diagnosis is essential to prevent its spread. This project utilizes high-performance computing (HPC) to tackle two major challenges: handling inconsistent medical data and enhancing TB detection. Due to the disproportion between healthy and TB-positive samples, Generative Adversarial Networks (GANs) are employed to create synthetic images, expanding the dataset and improving its diversity. This enriched dataset is then used to train Convolutional Neural Networks, which are highly effective in medical image processing. By leveraging HPC, we accelerate the CNN training process on large-scale, augmented datasets, significantly cutting down computation time while preserving accuracy. This approach enhances TB detection by integrating GAN-based data augmentation with CNN models, ensuring a quicker and more reliable diagnosis.
Authors - Shankar Lingam. M, Raghavendra GS, Sakthi Kamal Nathan Sambasivam Abstract - In the digital era, the integration of ICT policies and e-Governance has emerged as a critical driver for public sector modernization. ICT policy sets the groundwork for effective e-Governance systems by enabling the digital transformation of government functions and services. However, the realization of e-Governance goals faces significant challenges, particularly in fostering inter-governmental cooperation. These challenges arise from varying policy frameworks, technological disparities, and governance structures across different levels of government. In this context, navigating the intricacies of inter-governmental relationships is essential to ensure seamless information exchange and collaborative governance. The evolving digital landscape introduces both opportunities and complexities, particularly in terms of data privacy, cybersecurity, and digital inclusion. This paper presents a comprehensive review of ICT policies and e-Governance frameworks, with a focus on overcoming inter-governmental challenges in the digital era. Our methodology includes a scoping review of key studies and case analyses, such as the work of Obi (2007) on global perspectives of e-Governance [Obi, T. (2007). E-Governance: A Global Perspective on a New Paradigm], Prasad (2012) on India’s ICT policy for digital democracy [Prasad, K. (2012). E-Governance Policy for Modernizing Government through Digital Democracy in India], and Manda (2017) on South Africa’s smart governance approach [Manda, M. I. (2017). Towards "Smart Governance" through a Multidisciplinary Approach to E-Government Integration]. The findings highlight the importance of fostering interoperable and inclusive e-Governance systems to navigate the inter-governmental challenges posed by ICT adoption. Key implications for policy include the need for harmonized digital policies, the development of interoperable infrastructure, and the emphasis on inclusivity to bridge the digital divide. The results underscore the need for collaborative frameworks that enable effective governance across multiple levels of government, thus ensuring that the digital transformation of public services benefits all stakeholders.
Authors - Manisha Mane, Gargee Nitin Rangnekar, Gayatri Kishore Kshirsagar, Adarsh Suresh Nikam Abstract - Monitoring and prediction of space weather have gained tremendous significance with the increasing reliance of the telecommunication and aviation sectors on satellite communication and navigation systems. The two sectors are very vulnerable to space weather occurrences because they can always interfere with the functioning of satellites, high-frequency radio communication, and GPS accuracy. To mitigate these exposures, we recommend that telecommunication and aviation companies utilize a machine learning Space Weather Dashboard to facilitate real-time data visualization and predictive analytics assistance. The proposed architecture employs Azure Workspace for data storage and management, Unreal Engine 5 for the production of high-fidelity graphics, and machine learning models developed in Python. Our approach is based on the utilization of Long Short-Term Memory networks (LSTMs) for historical space data, Convolutional Neural Networks (CNNs), and Recurrent Neural Networks (RNNs). Various types of weather data points like solar X-ray flux, solar wind speed, coronal mass ejections, interplanetary magnetic field measurements, solar energetic particles, ionospheric data, and auroral data are utilized to enhance prediction precision. The dashboard allows for actionable insights to be built for industry professionals and real-time monitoring of critical space weather parameters. Additionally, in consideration of how it can improve their contribution to operation safety, the research here addresses the forecasting power of some of the machine learning architectures used for space weather. Our comparative research affirms that improved forecasting results in effective warning and risk assessment. This is a wonderful benchmark for the aviation and telecommunications industries, improving situation awareness and round-the-clock operating continuity.
Authors - Surya K, Rohit Kumar Abstract - The work deals with conceptual representation of sensitive issues in society as images in media using generative AI. Conceptual representation of sensitive information creates an overall societal impact. Conceptual processing involves understanding the represented information and the way how people respond and interact to the displayed information. Some of the issues like menstrual cycle representation, sex education, domestic violence, sex abuse and mental health awareness are difficult to represent conceptually and respectfully in the media. The goal of this paper is to use generative AI and represent these issues clearly to the society using media. Our work aims at understanding the challenges in representation of these images conceptually as well as providing an overview of the generative AI tools that can be used for implementing the solutions. A case study of a generative AI tool is used to understand the underlying problem of conceptual and respectful image generation for representing it in media. There are many generative AI tools for image generation, and we have chosen the Google Gemini AI as it gives more creative images compared to other tools [13]. In this paper, some of the sensitive issues are taken into consideration for representation in media in a conceptual and respectful manner.
Authors - T. A. Alka, M. Suresh, Aswathy Sreenivasan Abstract - This study aims to explore recent trends in rural entrepreneurship (RE). 439 documents available in the Scopus database are used for the trend analysis by using the R programming Biblioshiny package for bibliometric analysis. The result shows that rural entrepreneurship has been trending since the 1970s, and there is scope for further research. Rural entrepreneurs contribute to the society's upliftment, and development and ultimately result in the growth of the nation. The study is focused on the Scopus database; other databases are not considered. The study identified major themes, such as the rural agriculture entrepreneurs' contribution to developing the rural economy and the need for rural entrepreneurship education among college students in developing countries. Trend topics in this field highlight the rural entrepreneurship contribution to the development of the rural economy, agriculture, rural development, fostering innovation, promoting sustainable development, developing a sound entrepreneurial ecosystem, the role of rural entrepreneurship in the growth of developing countries, etc. The study identified North America and Asia connection, intra-European and transatlantic collaboration, Europe-Middle East, and Asia-North America regional collaboration. The study is relevant even when comparing recently published papers to map the trends; the themes used in research-related articles are always changing. This study gives insights to policymakers to help them with planning, policy formulation, program support, etc., to the development of rural entrepreneurs and also offers future research direction through thematic analysis.
Associate Professor and Head, Department of Computer Science & Engineering (Artificial Intelligence), Vishwakarma Institute of Information Technology, Pune, India
Tuesday August 25, 2026 12:28pm - 12:30pm IST Virtual Room BGOA, India
Authors - V. V. mandhare, Poonam Bhokare, P.S. Vikhe, Chandrakant Kadu Abstract - These days, billions of people use the internet worldwide. Technology for intrusion detection may be novel. a security technology generation that keeps an eye on the system to stave against malicious activity. The IDPS tracks malicious user behavior over time using a neighborhood procedure grid. Because of the rhetorical alternatives, the system suggests a security system during this project called the Intrusion Detection and Protection System at call level, which builds user pro- files to monitor usage activities. The proposed work is evaluated using intrusion detection systems and forensic techniques. The bottom paper includes a review of the literature on the Intrusion Detection System (IDS) and Internal Intrusion Detection System (IIDS). Internal Intrusion Detection System (IIDS), which employs predetermined algorithms or approaches to distinguish unauthorized user activity or attacks over a network, was designed during this research.
Authors - Dev Gandhi, Adarsh Srivastava, Rachit Soni, Daksh Parekh, Lokesh Heda Abstract - Recognizing people in images and videos is the main objective of computer vision-based person identification. The past ten years have seen a great deal of research in human detection. As single-stage algorithms, YOLO is a desirable choice for object detection because it offers faster results than two-stage algorithms. The advantage of this approach is that it provides both a manual for choosing the most effective human detection methods for real-world applications and a thorough analysis of current methods. This research paper's objective is specifically to evaluate and compare the performance of YOLOv3, YOLOv4, and YOLOv5 models on various images to detect human in visual scenes. Additionally, this paper discusses various parameters according to which the model’s efficiency is determined. Our experimental results demonstrate that YOLOv5 achieves higher accuracy, precision, and recall as compared to other YOLO models; the highest accuracy attained by YOLOv5 is 0.94 with F1 score of 0.96.
Authors - Shubhamm Kumaar, Akshat Sharma, Sukrati Chaturvedi Abstract - By connecting Agentic AI with Model Context Protocol Servers (MCPS) it is made possible to work towards autonomy of decision making. work-flow automation. Agent AI which is powered by large language models can provide a live context of information using MCPS. This work mostly focuses on the making of smarter workflows which are regular and pleasing. traditional automation. This integration improves using a decentralized multi agent framework. Makes decisions better and works smoother. Healthcare, manufacturing, smart, and other fields. cities. Tesla's efficient production and Singapore's Smart city are real-life examples. Nonetheless, various obstacles regarding scalability, ethics, computation, and more can hinder its application. However, challenges like scalability, ethics, and computational demands remain. This review synthesizes current research, applications, and future directions, underscoring the promise of this technology for innovative automation solutions.
Authors - V. V. mandhare, Hemlata Mali, P.S. Vikhe, M. R. Bendre, M. R. Parkhe Abstract - This paper introduces a smart and efficient Hall Ticket Generation System that integrates QR code technology to improve the process of generating and managing hall tickets for academic examinations and events. Built using Java and MySQL, the system is designed to simplify administrative work, enhance security, and promote eco-friendly practices by eliminating the need for paper-based hall tickets. The core functionality of the system is its ability to generate dynamic QR codes for each hall ticket, which can be quickly scanned using an Android- based scanner application. This allows invigilators to instantly verify the identity and credentials of candidates, reducing the chances of fraudulent entries or unauthorized access to the examination hall. By transitioning tothisdigitalmodel,institutionsbenefitfromreducedpaperusage,lower administrative costs, and improved operational efficiency. The system is not only secure but also user-friendly, making it easier for both staff and students to manage examination logistics. In Automatic Question Paper Generator Module, which addresses the common challenges associated with manually preparing exam questions. To overcome these limitations, the system uses a key word-based randomization algorithm to create question papers swiftly and securely. This method ensures that each set of questions is unique, well-distributed across topics, and free from duplication. The system is capable of storing and managing multiple question paper sets, making it easier to conduct exams across different classes while ensuring comprehensive curriculum coverage. Overall, this project offers a complete solution that not only simplifies exam management but also enhances the quality and integrity of academic assessments through automation and intelligent design.
Authors - Madhu Shukla, Vipul Ladva, Simrin Fathima Syed, Neel H. Dholakia Abstract - Cardiovascular disease continues to be one of the leading causes of death globally, demonstrating the critical role of efficient and reliable prediction models. Here in this study a dataset that is integrated from five heart disease datasets originated from publicly available sources such as UCI for Heart Attack risk prediction and analysis with 1,888 instances are used. Fourteen primary factors that include age,abnormality of cholesterol level, type of chest pain, as well as exercise related parameters from demographic and clinical dimensions were investigated in order to find the association with heart disease. Significant trends were found using data visualization to show high heart risks with certain chest pain types and high maximum heart rate. A correlation matrix illustrates important inter-feature relationships and sheds new light on the predictabilitheckability of the features. This study highlights the possibility to exploit demographic and clinical information for early detection of high risk individuals and in the future, to allow medical interventions and improve health state.
Authors - Bhukya Rakesh, Dasari Yaswanth, Kumari Nidhi Lal Abstract - Mobile Ad Hoc Networks (MANETs) are dynamic, decentralized networks that require efficient routing mechanisms to ensure reliable communication. Traditional routing protocols struggle with issues such as high node mobility, energy constraints, and unpredictable topology changes. This research explores the integration of artificial intelligence, specifically neural networks, to enhance routing efficiency in MANETs. Our approach leverages deep learning models to predict optimal routes by analyzing network parameters such as node density, mobility patterns, and link stability. The proposed AI-driven routing mechanism dynamically adapts to network variations, improving packet delivery ratio, reducing latency, and optimizing energy consumption. Comparative evaluations against conventional routing protocols, such as AODV and DSR, demonstrate significant improvements in network performance. The results highlight the potential of neural networks in revolutionizing adaptive routing for MANETs, paving the way for more intelligent and resilient communication systems.
Authors - Ayush Tiwari, Ayushi Tomar, Prabhjot Kaur Abstract - Online banking fraud constitutes illegal access to accounts for the payment transfer purposes. There are several reasons for difficulty in the detection of such crimes-from the imbalance of data to the fraudsters' techniques that are ever-changing. The set of tools used for this purpose are machine learning, economic optimization, and risk assessment. By combining these techniques, a maximum reduction in the losses due to fraud and false positives will be achieved. The machine learning models, when validated against real datasets, were able to reduce losses by 52%, allowing for just 0.4% false positives. The improvement in behavior analysis for fraud detection is conducted through transaction clustering, sliding window aggregation, and adaptive classifiers. The algorithms such as KNN, SVM, Logistic Regression, Local Outlier Factor, and Isolation Forest are used to predict fraud in credit card transactions so that it is accurately detected, false alarms being at a minimum, and customers are availed against unauthorized charging.
Tuesday August 25, 2026 12:30pm - 2:30pm IST Virtual Room BGOA, India
Authors - Nisarg Chaudhari, Urva Dave, Zeel Patel, Manasvi Vachhani, Dweepna Garg, Bhavika Patel, Kashyap Patel, Parth Goel Abstract - Fertilizer recommendation plays a crucial role in optimizing crop yield while minimizing resource wastage. The integration of machine learning techniques enables precise fertilizer prediction based on soil and environmental conditions, leading to improved agricultural productivity. However, traditional methods often result in overuse or underuse of fertilizers, negatively impacting soil health and crop growth.This study employs various machine learning algorithms, including RandomForest, XGBoost, LightGBM, HistGradientBoosting, CatBoost, and Neural Networks, to classify and recommend fertilizers based on soil parameters. The models were trained on a synthetic fertilizer dataset containing diverse soil compositions and fertilizer requirements.Experimental results indicate that Neural Networks outperform tree-based models, achieving the highest testing accuracy of 84.10%, demonstrating strong generalization capabilities. Accuracy and loss trends over epochs confirm stable learning, while a confusion matrix reveals minimal misclassifications. This study highlights the effectiveness of deep learning in optimizing fertilizer recommendations, contributing to more sustainable and efficient agricultural practices.
Authors - Nirav Narayan, Martin Parmar, Parth Shah, Mrugendra Rahevar Abstract - Water quality is a critical concern for human health, ecosystems, and sustainable resource management. Traditional water quality monitoring methods are expensive, time-consuming, and often lack real-time data availability. This research proposes a Secure Water Quality Monitoring framework integrating IoT, LoRa WAN, and secure data transmission to enable real-time, cost-effective monitoring in remote and urban areas. The system employs low-power LoRa technology for long-range communication, ensuring reliable data transmission even in connectivity-challenged regions. ESP32 microcontrollers process sensor data, measuring key parameters such as pH, turbidity, TDS, EC and DO. Security is ensured using AES-128-bit encryption and SHA-256 hashing, safeguarding environmental data against tampering. The proposed framework addresses challenges in conventional monitoring, such as high costs and limited scalability, by offering a low-cost, energy-efficient, and scalable approach. This study demonstrates the potential of IoT-driven smart monitoring framework to enhance water resource management and environmental sustainability, paving the way for future advancements in sensor technology, energy efficiency, and data security.
Authors - Amrithesh TV, Sajin John Shaji, Sangeeth Gopinath Abstract - Family businesses play significantly in the world economy; yet, the majority can somehow manage to survive through leadership change from impacts of family relationships. Unlike some business corporations with formalized transitions, family companies dominantly depend on family relationships and informal decision-making, hence bringing about business instability as well as conflicts. This research examines just how internal family systems indeed influence successor choice, and shape leadership transition on business resilience. Through in-depth qualitative research with strict thematic analysis of family firm owner interviews, the research discovers a number of important succession determinants, such as traditional heirarchy, mentering, and resistance to modernization of the modern kind. The research does uncover that in fact successor selection is usually highly influenced by family influence, in contrast to planned planning, and therefore jeopardizing some degree of business stability. This paper suggests the adoption of comprehensive succession planning and participative decision-making. Systematic leadership development programs, coupled with open communication, will ensure successful leadership succession and long-term business success.
Associate Professor and Head, Department of Computer Science & Engineering (Artificial Intelligence), Vishwakarma Institute of Information Technology, Pune, India
Tuesday August 25, 2026 2:30pm - 2:32pm IST Virtual Room BGOA, India
Authors - Ketki Kshirsagar, Samarth Chikane, Dev Desai, Avdhut Hande, Aniruddha Deobhankar, Arun Govind, Shubham Derkar, Arjun Gupta Abstract - The one of the most important sensory organ of our body is an eye, it is the reason why people can enjoy its beautiful surroundings. What if we would not have this important organ?, the answer is quite obvious, it would be very challenging, he would be isolated. There are millions of people across the globe living such miserable lives. So to overcome this challenge we come up with an idea to build an assistive aid for blind needy. The project is AI-Powered Smart glasses for visually impaired. This basically notifies the blind person about the obstacle in front of him/her. This tool has the ability to tell the user the distance of the obstacle and what particularly the obstacle is like the tree is 60 cm away. This paper includes the brief information about how this glasses work. The glasses are full advanced technical tools like ultrasonic sensors, IR sensors, ESP32 microcontroller, ESP32 cam module, earpiece for user’s enhanced listening. The software we used is YOLOv5 for object recognition, Tensor Flow Lite, text-to-speech software, MATLAB, etc.
Authors - Sparsh Kumar, Satyadhyan Chickerur, Prashanth Kumar Malkiwodeyar Abstract - Accurately capturing unsteady flow phenomena and complex fluid dynamics is essential for understanding and predicting arterial blood flow behavior. This study leverages computational fluid dynamics (CFD) and machine learning (ML) to detect regions of elevated shear stress in the aorta, focusing on the dynamic responses of arteries under varying hemodynamic conditions. The geometric model of the human aorta was sourced from the Vascular Database, which is supported by SimVascular and includes all necessary boundary conditions. Simulations were performed using the Navier-Stokes equations within SimVascular to generate vtk files containing velocity and pressure data. Since these files could not be directly used for ML training, additional postprocessing was conducted in ParaView. By applying specific functions, we extracted key metrics such as pressure, velocity, wall shear stress, and other parameters at multiple spatial coordinates. This resulted in a CSV dataset comprising 23,777 points with corresponding attributes for further analysis of high wall shear stress regions. A Random Forest classifier was trained on this dataset to predict regions of high wall shear stress of the aorta by analyzing attributes like pressure, velocity, and wall shear stress, providing precise coordinate-based predictions of areas with abnormal hemodynamic stress. In our analysis, regions of elevated shear stress were detected at 1,288 points, representing 5.42% of the total dataset. By integrating CFD with ML, we successfully identified regions of high wall shear stress in the aorta, enhancing clinical diagnostic accuracy and offering a data-driven alternative to traditional cardiovascular diagnostic techniques. This study underscores the value of combining CFD and ML to reduce reliance on traditional cardiovascular tests, streamlining diagnosis and therapeutic planning while reducing costs and complexity.
Authors - Vanishree Pabalkar, Ruby Chanda, Yash Yadav, Megha Patil Abstract - Sentiment analysis, is termed as opinion mining, is a significant tool to assess customer’s opinions and expressions by analyzing textual data from various digital platforms. In marketing, sentiment analysis provides invaluable insights into customer feedback, helping companies to customize the products and services, and marketing strategies to meet consumer needs. This paper explores the application of sentiment analysis specifically through a case study of the Samsung Galaxy S24 Ultra. The study involves collecting data from various sources, like the news forums, and news articles, and employing natural language processing (NLP) techniques to classify and analyze sentiments into positive, negative, or neutral categories. The outcome conveys the essence of sentiment analysis in identifying consumer preferences and issues, such as high prices or software problems, which directly impact marketing strategies and product development. By using sentiment analysis, companies like Samsung can make data-driven decisions to retain satisfied customers and ensure brand loyalty. This study also highlights the issues and constraints of current sentiment analysis methods, that include the need for improved accuracy in sentiment classification and the handling of complex linguistic nuances. Future research directions include enhancing ML tools to classify the sentiment detection and exploring the use of sentiment analysis in real-time applications to provide instant feedback for marketers. The implications of sentiment analysis extend beyond marketing into areas like public relations, customer service, and product innovation, making it an indispensable tool in today's digital age. As digital communication continues to grow, the role of sentiment analysis is expected to expand, offering inputs into consumer behavior and enabling more personalized, effective strategies.
Authors - Kumar Rahul, Ramjee Prasad Gupta, Neeraj Arora, Surender Kumar Kulshrestha Abstract - Artificial Neural Network (ANN) plays an important role in shaping modern social media platforms. The networks assist in presenting content recommendations, analyzing user engagement, detecting sentiment, and automating moderation processes. Through the processing of user data, ANNs enhance personalization, improve advertising strategies, and detect harmful content, ensuring a seamless and engaging user experience. This paper explores the diverse applications of ANNs in social media, highlighting their impact on user interaction, content curation, and platform security. This review highlights key advancements, challenges, and open issues related to data variations, and evaluation criteria in social media analysis. Additionally, a structured framework is proposed for future studies focused on leveraging ANNs to gain social media insights.
Authors - Tejas Nadagadalli, Vishwanath Baligar Abstract - The assistance of deep learning for medical image diagnosis is often crucial in the timely treatment of patients suffering from diseases like brain tumors and lung cancer. This paper evaluates the performance of VGG 16 and Efficient-Net deep learning models for the classification of MRI and CT scans of the brain and lungs. A new low complex algorithm referred to as PDBS was promoted to increase the efficiency of model optimizations. The lesion detection models were assessed for accuracy, training time, and level of generalization attained. Experimental results highlight that the PDBS model consistently outperformed traditional CNN architectures, achieving higher classification accuracy with 97% testing accuracy for brain MRI scans and 96.5% for lung CT scans while maintaining efficiency. These results above illustrate the depth of the contribution offered by deep learning methods to the enhancement of medical image analysis to support clinical workflow.
Authors - Sweety Singhal, Uma Sharma Abstract - Healthcare domains deal with massive amounts of sensitive data, such as patient health records, diagnostic results, and clinical notes, which must be secured under privacy regulations (like HIPAA). Traditional security technologies cover many problems but struggle against more advanced iterative threats. Implementing deep learning algorithms has resulted in the creation of texts of advanced encryption techniques, which can discover obvious patterns and improve the security of health systems. When combined with Natural Language Processing (NLP), these algorithms can also anonymize and de-identify patient information, allowing healthcare providers to share and collaborate on data without violating patient confidentiality. This will prove valuable for further medical research and improving the quality of telemedicine, where better information transfer is key for better treatment results. Key Approaches: Data encryption, differential privacy, tokenization, and access control are all essential methods of protecting healthcare data, and NLP plays a vital role in ensuring that sensitive health information is handled securely.
Authors - Madhuri Badole, Rohit Rathod, Pavan Bachhav, Devanshu Parulekar, Harsh Kulkarni Abstract - Brain tumor diagnosis is a critical area of medical research as it directly impacts patient survival and treatment. Early diagnosis is essential for improving prognosis and facilitating therapy. Magnetic Resonance Imaging (MRI) is particularly reliable for detecting brain tumors due to its superior image quality and contrast. This study provides a comprehensive review of recent advancements in brain tumor diagnosis methods. Evolutionary algorithms based on natural selection principles offer optimal strategies for image processing, segmentation, feature extraction, and classification in tumor diagnosis. Random Forest algorithms are used in this study to classify MRI images, distinguishing between normal brain tissues and malignancies. Additionally, the study explores hybrid models that integrate evolutionary algorithms with neural networks (CNN) to enhance accuracy. This research offers insights into the benefits and limitations of these approaches, paving the way for further neuropsychology research.
Authors - Pritee Parwekar, Chinmayee Ambarish Parwekar, Kshitij Bhushan Abstract - As growing dependence on wind and solar energy brings about new challenges in both maximizing energy generation and ensuring the stability of grids due to the intermittent nature by virtue of dependence on variable atmospheric conditions, optimization of hybrid solar-wind plant output is achieved through the present study by means of a DE algorithm for maximum energy yield enhancement and grid robustness. The approach includes a simulation of a small hybrid energy system, which consists of a 10 m² solar panel and three wind turbines, each with a capacity of 2 kW, over a period of 24 hours. Using real meteorological data in the form of solar irradiance and wind speed profiles, the differential evolution (DE) algorithm minimizes two most important parameters: tilt angle of solar panels from 0 to 90 degrees and the spacing of the wind turbines, variable from 5 to 50 meters. The objective function is to maximize energy output in total and minimize hour-by-hour power oscillations, a surrogate for grid stability. The results indicate that the optimized configuration with tilt angle 15.23° and turbine spacing of 35.67 m produces a total of 135.82 kWh, up 12.7% from the baseline (120.45 kWh at tilt angle 30° and 10 m spacing). Moreover, the stability penalty, expressed as the sum of hourly output differences, reduces from 48.73 to 42.19, reflecting better grid compatibility. These results underscore the potential of DE as a successful method for the optimization of renewable energy, with a real application to harmonize energy production and stability in hybrid systems. This research supports more efficient and trustworthy renewable grids and contributes to sustainable energy infrastructures transition
Authors - Amogh M, Satyadhyan Chickerur, Prashanth Kumar Malkiwodeyar Abstract - Cardiac arrhythmias pose a significant challenge in clinical diagnostics, necessitating accurate and efficient detection methods. This study explores the classification of arrhythmias using advanced machine learning models, including Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, and Physics-Informed Neural Networks (PINNs). A dataset of 16,000 simulated ECG signals, generated using SimVascular, provided the foundation for training and evaluation. CNNs achieved high accuracy in spatial feature extraction, while LSTMs excelled in capturing temporal dependencies in sequential ECG data. PINNs emerged as the most robust model, achieving a training accuracy of 97.8% and a testing accuracy of 97.2%, leveraging domain-specific constraints from the FitzHugh-Nagumo equations. The results highlight the complementary strengths of these models, with PINNs offering superior interpretability and physiological consistency. Future work will focus on integrating multi-modal data and developing real-time systems to advance arrhythmia diagnostics and improve cardiac care outcomes.
Tuesday August 25, 2026 3:30pm - 5:30pm IST Virtual Room BGOA, India
Authors - Siddhant Sawant, Sajal Nampalliwar, Ansh Masand, Variza Negi Abstract - Financial literacy is a crucial yet often overlooked skill in formal education. Our research presents an AI-driven framework that democratises financial education through four interactive modules: budgeting tools, a newsletter, a virtual market simulator and a structured course. At the core lies an Intelligent Learning Agent. A chatbot first gauges prior knowledge, then steers each learner to the appropriate level. Machine-learning, NLP and reinforcement-learning techniques continually adapt the pathway in response to engagement signals. We detail the conceptual framework and system architecture, illustrating how AI delivers scalable, personalised financial learning.
Authors - Saraswati Patil, Mustafa Limdiyawala, M.S. Dawngliana Fanai, Meghaj Kharwadkar, Shivshankar Mahajan Abstract - Natural disasters such as earthquakes, floods, and tsunamis pose severe threats to human lives, infrastructure, and economies. Effective prediction and response strategies are vital for minimizing their impact. This paper introduces an AI-driven Disaster Prediction and Relief Dashboard, an integrated platform leveraging machine learning and geospatial mapping to forecast natural disasters and optimize relief operations. Using Random Forest and Gradient Boosting algorithms trained on historical data, the system predicts the likelihood, magnitude, and severity of disasters. Geospatial visualization highlights high-risk zones and delivers real-time situational awareness for authorities. Additionally, the platform streamlines relief management by dynamically allocating resources based on predicted disaster severity and location. By integrating predictive analytics with operational planning, the system enhances preparedness and responsiveness, contributing to more resilient disaster management.
Authors - N.N.S.S.S. Adithya, P. Vanishree Sah, B. Jyothirmai, Nidhi Mishra, D. Indira Abstract - Early prediction of lung cancer is crucial for reducing the death rate. Artificial intelligence, particularly deep learning, is employed to analyze CT scan images for more accurate automated prediction of types of lung cancer. This process of prediction is called classification. Lung cancer classification can be done with pre-networks such as VGG16 and ResNet50.But the main drawback of these techniques is that cancer cannot be detected on Histopathology images (i.e. image of tissues). As VGG16 and ResNet50 are designed for more general usage, they are not suitable for analyzing Histopathology image. This study involves the development of a customized neural network model which can solve the problem of analyzing Histopathology images. This CNN model can help us detect lung cancer at a very early state in lung tissues. Detecting lung cancer at a very early stage can help doctors to cure the patient and save the life of a patient.
Authors - Sneha S. Biradar, Suvarna Kanakaraddi, Neha Tarannum Pendari Abstract - This personalized music therapy framework for children with Autism Spectrum Disorder (ASD) involves data collection (AQ scores, age, gender, and demographics), severity identification, and customized music creation. Using a publicly available dataset, children were classified by severity, allowing the design of therapeutic music with varying duration, tempo, and complexity. Compositions were set to 15 minutes for low severity, 30 minutes for moderate, and 50-60 minutes for high severity. Preliminary results suggest this approach boosts engagement and may improve cognitive, emotional, and social outcomes, demonstrating the potential of combining advanced analytics with personalized music therapy for ASD.
Authors - Ruby S Chanda, Vanishree Pabalkar, Priya Pradipkumar Tiwary Abstract - Companies are increasingly using data analytics and AI to personalize interactions and provide tailored recommendations. Additionally, there is a growing focus on emotional intelligence and understanding customers' needs beyond their transactional behavior. Some real-world examples are – Netflix uses AI to analyze viewing history and preferences, recommending personalized content, Spotify leverages data on listening habits to create tailored playlists and discover new music. The need for AI models in NPS and CX enhancement arises from the increasing complexity of customer interactions and the vast amount of data generated. AI can help in predicting consumer’s future behavior by analysing their demographic and purchase data and identifying patterns. This empowers businesses to create more tailored and meaningful customer experiences, resulting in greater satisfaction and loyalty. This project aims to develop a Generative AI-based Net Promoter Score (NPS) predictor to enhance Customer Experience (CX) in the retail industry. By leveraging advanced AI techniques like VAEs and GANs, the model will be able to analyze vast datasets of consumer behavior and demographics, providing more accurate and personalized NPS predictions.
Authors - Shriraj A. Patil, Chudaman D. Sukte, Jayesh R. Patil, Chinmay R. Mhaske, Mandar Dakhorkar, Manohar K. Kodmelwar Abstract - This paper presents a novel IoT-based fruit-picking system that integrates Reinforcement Learning (RL), Transfer Learning (TL), and Neuroevolution to address the inefficiencies of current robotic harvesting methods. As demand for efficient agricultural practices rises, traditional fruit-picking systems face significant challenges, including operational inefficiencies, fruit damage, and limited adaptability to diverse environments. Our proposed solution leverages RL to optimize picking strategies through adaptive learning, enhancing the robotic arm's efficiency over time. TL is employed to improve fruit recognition capabilities, utilizing pre-trained models for accurate ripeness detection, even with limited training data for specific fruit varieties. Additionally, Neuroevolution evolves control strategies for the robotic arm, enabling it to adapt to dynamic harvesting conditions. Comprehensive simulations demonstrate significant improvements in picking accuracy, efficiency, and adaptability compared to existing methods. The findings highlight the potential of integrating these AI models within IoT frameworks to revolutionize fruit harvesting, ultimately contributing to smarter farming practices and enhanced agricultural productivity. This research underscores the interdisciplinary nature of modern agriculture, combining advancements in AI, robotics, and IoT technologies to provide innovative solutions for the challenges facing the agricultural sector today.
Authors - Gatla Vijayendher, K Sai Karthikeya, E Jayanth Madhav, Tadepalli Satya Kiranmai Abstract - The increasing number of tumor cases has caused an alarming situation in the health care space. The tumor detection and diagnosis is a very computationally heavy and requires multiple medical imaging devices such as Magnetic Resonance Imaging (MRI) and Computed Tomography (CT). It is very vital in order for early detection of tumor which can be done by precisely measuring their size which can improve a treatment by a huge factor in the patients. Existing diagnostic approaches often face challenges due to the diverse appearances of tumors and the constraints of current models. This study examines the different cutting-edge deep learning methods, with a focus on utilizing Generative Adversarial Networks (GANs) to enhance tumor identification across various categories, types, and imaging techniques. We also investigate the role of data augmentation strategies in enhancing the model performance. Furthermore, we examine the integration of Convolution Neural Networks (CNNs) to achieve accurate and robust results while preserving the data privacy. The goal of this study is to understand the detailed current scenario of the early detection and ways about the different techniques in various kinds of tumors which are present in the human body.
Authors - Amol More, Sanjeev Kumar, Sandeep Kore Abstract - This study carefully checks how well three different types of heat sinks move heat under controlled laboratory conditions. The main focus is on Copper Pin Fin Heat Sinks, Aluminum Phase Change Material (PCM) Pocketed Heat Sinks, and Aluminum Plate Fin Heat Sinks (PFHS). These were put through a wind tunnel test that simulated forced convection. This gave a thorough comparison of how well they kept heat in. To make the experiments work, heat was applied to the bottom of the heat sinks with 10W, 20W, and 30W of power, to represent various thermal loads. The speed of the air was changed from 1 m/s to 5 m/s to see how speed affected how well heat was removed. It was also improved by making changes like adding a copper plate to the aluminum fins and making holes in both the shield connection and the pin fins which were used in the experiment. By changing the surface area and turbulence, these changes are meant to see if they can improve the rate of heat transfer. The study's results should give us useful information about how to build heat sinks so they work best in a wide range of situations, from home electronics to industrial systems. It is expected that the results will help make thermal management solutions that work better, which will improve the performance and life of heat-sensitive parts.
Wednesday August 26, 2026 9:30am - 11:30am IST Virtual Room BGOA, India
Authors - Ayinampudi Siva Rama Raju, Suneetha Dwarapu, Gaddiboyina Sai Jahnavi, Tummalapalli Sai Sri Varshit, Kalimahanthi Sai Nikhil Kartikeya Abstract - Sentiment analysis, a key task in natural language processing (NLP), identifies emotional tone in text. This study compares two sentiment classification approaches: a lexicon-based method using VADER (Valence Aware Dictionary and sEntiment Reasoner) and a transformer-based deep learning method with RoBERTa (Robustly Optimized BERT Pretraining Approach). Using the Amazon Fine Food Reviews dataset of 568,454 customer reviews, the analysis categorizes sentiments as positive or negative. Preprocessing steps, including text normalization and handling missing values, ensure data reliability. VADER efficiently processes short, informal texts using a predefined lexicon but struggles with complex linguistic structures and contextual subtleties. RoBERTa leverages transformer-based architectures to capture intricate word relationships, enabling superior accuracy and nuanced sentiment detection in contextually rich texts. A comparative evaluation demonstrates that RoBERTa outperforms VADER by a significant margin, underscoring the strengths of deep learning for detailed sentiment analysis. These findings emphasize the trade-offs between speed and contextual depth in sentiment analysis models and provide valuable insights for customer feedback interpretation, opinion mining, and broader NLP research.
Wednesday August 26, 2026 9:30am - 11:30am IST Virtual Room BGOA, India
Authors - Ruby S Chanda, Rahul Dhaigude Abstract - In order to predict stock values, this study investigates the combination of machine learning with sentiment analysis. The quick spread of news and the growth of social media sites like Twitter have made public opinion a bigger factor in financial markets. This study extracts market sentiment from tweets and news stories using Natural Language Processing (NLP) techniques, namely the VADER sentiment analysis tool, which has been tailored with financial lexicons. To forecast stock price fluctuations for firms like Amazon and Tesla, sentiment data is included into a Generative Adversarial Network (GAN) model together with technical indicators like moving averages and Bollinger Bands. The model is evaluated using performance metrics like Root Mean Square Error (RMSE), demonstrating its ability to capture price trends and market sentiment dynamics. While results highlight the potential of GANs for real-world applications in financial trading, the study also acknowledges limitations such as data quality and model uncertainty. Future directions include improving sentiment algorithms and incorporating additional market factors..
Authors - Aanjaneya K, Anjana P, S Sameera, Ajith Sundaram Abstract - The environment-related worries of Generation Z together with sustainability-based activities established them as leaders who champion green consumerism. Digital natives of this generation opt to make buying choices on social media platforms according to their established reputation. The platforms of Instagram together with YouTube and LinkedIn function as essential spaces for spreading sustainability content which affects how people behave regarding their purchasing choices. Social media promotes consumer engagement through direct communication and enables fast information flow about green events so it stands as a key factor in developing positive green purchasing attitudes. Current research analyzes the impact of social media information sharing on Gen Z sustainable buying motivation through an investigation of green-value and subjective-norms as intervening variables. This research depends on the Stimulus-Organism-Response (SOR) model to see how social media leads consumers toward buying green products. This research study addresses the mental factors behind environmentally conscious buying to provide concrete recommendations for business organizations and government institutions. Companies can use the research results as a foundation to create better sustainability-oriented marketing plans that aim at Gen Z consumers.
Wednesday August 26, 2026 9:30am - 11:30am IST Virtual Room BGOA, India
Authors - Jishnu Prakash k, Rithish S V, Liz Maria Liyons, Vijval Srinivasan, Deepthi L R Abstract - The rapid spread of misinformation and rumors on social media platforms, particularly Twitter, poses significant risks to public perception and decision-making. This study presents a comprehensive approach to analyzing and mitigating rumor propagation by identifying key influencers and optimizing propagation time within online communities. Our dataset consists of over 800,000 nodes with interactions categorized as retweets, mentions, and replies, each assigned an influence score to quantify user impact. Using the Infomap algorithm, we initially detected 13,500 communities and filtered them to retain 67 influential clusters with a higher number of nodes and stronger influence scores. To analyze the spread of rumors, we developed an algorithm that tracks propagation within these communities, leveraging top influencers as initial spreaders and computing the average propagation time. Furthermore, we introduced a node deletion strategy to iteratively remove high-impact influencers, reducing the overall propagation time and limiting misinformation spread. Finally, we adjusted the propagation times by normalizing them with the earliest influencer timestamps to ensure precise measurement. Our findings highlight that targeted removal of key spreaders significantly disrupts rumor diffusion, providing insights into optimizing influence-based network interventions for misinformation control.
Authors - Yukta, Nishant Yadav, Sukrati Chaturvedi Abstract - The rapid expansion of social media platforms and online news consumption has led to an increased spread of fake news, posing significant challenges to society by influencing public decision- making and causing severe consequences in domains such as politics and healthcare. Traditional methods for identifying fake news are often too slow to combat its swift dissemination. Therefore, identifying and addressing misinformation is crucial for maintaining the accuracy and trustworthiness of information disseminated on social media. Natural language processing (NLP) and artificial intelligence (AI) play a vital role in this endeavor by facilitating the effective examination of extensive datasets to uncover patterns that are suggestive of misinformation. Machine learning models, particularly transformer-based architectures, enhance fake news detection by improving accuracy and interpretability. The integration of Explainable AI (XAI) methods further enhances trasparency and trust in these models
Authors - Praful Sambhare, Nitin Choudhary, Abhay Rahangdale, Atharva Rane, Sahil Raina Abstract - The integration of artificial intelligence (AI) within the fashion industry is becoming increasingly prevalent, with the aim of delivering a shopping experience that is personalized, seamless, and engaging. As an example of this trend, GlamBot is an AI- driven fashion assistant, a sophisticated fashion assistant that employs a range of advanced AI methodologies. These methodologies include Natural Language Processing (NLP), image- based similarity searches, and voice recognition technologies, all of which together transform the interactions that users have with fashion platforms. The functionality of GlamBot significantly improves the user experience by providing tailored fashion recommendations. This is achieved through the analysis of text inputs, the execution of visual similarity searches, and the processing of voice commands. Consequently, fashion discovery is rendered more intuitive and accessible for users, thereby facilitating a more engaging interaction with the fashion domain. . By analyzing and understanding user preferences, GlamBot builds personalized profiles that evolve over time to deliver increasingly accurate recommendations . GlamBot’s image-based search feature allows users to upload pictures of fashion items they like. Using advanced ResNet50 image recognition models, GlamBot analyzes these images and provides visually similar product recommendations, bridging the gap between users’ visual preferences and available fashion products .
Authors - Merlin Priya Jacob, Sukhada Aloni, Hetal Rawat, Shrishti Sakore, Shreya Naik, Shubham Pardhi Abstract - Using Ethereum, IPFS, and the MERN stack, this project offers a transparent and safe blockchain-based examination system. The solution guarantees tamper-proof question paper management by utilising IPFS (via Pinata) for decentralised storage and Ethereum smart contracts. Instructors submit tests to IPFS, ensuring data integrity by storing their cryptographic hashes on the Ethereum blockchain. While MetaMask allows for safe user interaction and authentication, Hardhat makes it easier to design and deploy smart contracts on a testnet. A strong online application is powered by the MERN stack, with Node.js managing database functions and instructor authentication. Exam papers are safely retrieved by authorised superintendents, guaranteeing regulated access. Academic assessment integrity is improved by this decentralised method, which reduces the possibility of paper leaks and unauthorised changes while offering a scalable and effective substitute for conventional test systems.
Authors - Sarbjit Kaur, Jasmeen Gill Abstract - Wireless sensor networks (WSNs) play a vital role in sensing environmental conditions in far-flung areas. However, their energy consumption remains a critical issue, affecting the network's lifetime and coverage area. Clustering has emerged as an efficient strategy to prolong sensor network lifespan, and the Fruit Fly Algorithm (FFA) and Ant Colony Optimization (ACO) are promising techniques for cluster formation and efficient path establishment, respectively. In this study, we propose an innovative approach that combines FFA for cluster formation and ACO for path establishment. This novel algorithm is implemented in MATLAB and evaluated in both homogeneous and heterogeneous environments. We compare our proposed algorithm with the Biogeography-Based Optimization Algorithm (BOA) and the Low Energy Adaptive Clustering Hierarchy (LEACH) algorithm. Our results indicate that the proposed algorithm significantly outperforms both BOA and LEACH in terms of network lifetime and coverage area, particularly in heterogeneous environments.
Authors - M Jayaram, Kodari Madhavi, Amboth Anil Kumar, Pachipala Naveen, Gajula Rithvik Abstract - Human-Computer Interaction prioritizes universally accessible systems, crucial for individuals with physical disabilities. This study introduces an Eyes as Interfaces: A Novel Eye-Tracking Mouse Cursor System, a hands-free solution enabling seamless digital environment interaction. For people those with paralysis, muscular dystrophy, or spinal injuries, this technology provides independent computing access, eliminating dependency on external assistance. A Convolutional Neural Network (CNN) is the system's backbone that provides real-time pupil detection, mapping eye gaze to exact cursor movement. Advanced image processing like this guarantees smooth operation regardless of changing lighting and user conditions. By accurately mapping eye movements to cursor actions, users can navigate and communicate with computer interfaces, opening avenues for information access, communication, and work participation. This encourages independence and enables users to access the web on their own, performing tasks like document creation and web navigation. Beyond personal benefits, this technology promotes inclusivity by bridging the digital divide, allowing for real-time, unrestricted participation in learning, employment, and social activities. Its smooth integration in widespread digital platforms ensures that it carries the highest level of potential in changing lives of people with mobility disabilities Worldwide.
Authors - Kailash Agarwal, Parikshit N. Mahalle, Bhagwan D. Thorat Abstract - Heart failure is a serious medical condition that affects millions worldwide, and early prediction is essential for timely intervention and better patient outcomes. While machine learning models have demonstrated strong predictive capabilities in healthcare, many high-performing models, such as Support Vector Machines (SVM), function as black boxes, making them difficult to interpret in clinical settings. This study examines how Explainable AI (XAI) techniques can enhance transparency in heart failure prediction.Using a publicly available dataset from Kaggle, we preprocess the data with label encoding, feature scaling and Hyperparameter tuning before training various machine learning models for binary classification. Our results indicate that the Support Vector Classifier (SVC) with a Linear kernel achieves the highest predictive accuracy. However, to improve interpretability, we compare its performance with explainable models like Decision Trees and apply post-hoc explanation techniques such as SHAP (SHapley Additive Explanations) and Permutation Importance.Through this comparative analysis, we highlight the trade-off between model accuracy and interpretability, offering insights into the feasibility of XAI-driven models in real-world clinical decision-making. Our findings reinforce the importance of developing AI systems that not only perform well but also provide understandable and trustworthy insights for medical professionals.
Authors - Bhagwan Thorat, Omkar More, Prathamesh Medage, Aditi Mali, Janhavi Maske, Sanika Maind Abstract - Traditional healthcare systems have primarily focused on physical health, often overlooking mental well-being. With the rapid advancement of technology, integrating AI-driven emotion detection into mental health management can offer valuable insights. This paper presents a comprehensive mental health management system that utilizes Haar Cascade classifiers and a Keras deep learning model for real-time emotion recognition via OpenCV. A Flask-based web interface, built using HTML, CSS, and Python, enables users to monitor their emotional states and facilitates therapist booking and automated receipt generation. By leveraging facial expression analysis, the system provides a data-driven approach to mental health assessment, enabling early intervention. The platform also ensures accessibility and efficiency, reducing the burden on healthcare providers. Experimental evaluations demonstrate the system’s effectiveness in accurately detecting emotions and its potential in AI-assisted psychological support. Future enhancements will focus on multi-modal emotion detection, incorporating natural language processing (NLP) and IoT-based physiological monitoring for a more holistic approach to mental health assessment. This research contributes to the growing field of AI-powered mental health solutions, bridging the gap between technology and psychological well-being while promoting early detection and accessible care.
Authors - Sonali Patil, Siddhesh Arun Patil, Ayush Patil, Piyush Pawar, Siddhesh Sandeep Patil Abstract - Around 970 million people face mental health disorders across the world and depression affects 75% of these people because of their sleep disturbances. The connection between persistent sleep problems and depression emerges when affected individuals become twice as likely to develop depression thus establishing sleep as a major sign for mental health forecasting. Current approaches to this problem deal with three key issues which are dataset biases, small available sample sizes along with the reliance on self-reported symptoms instead of actual physiological signals. Our deep learning solution relies on multi-channel Convolutional Neural Networks (CNNs) to analyze wearable sensor data because it tackles existing analysis limitations. DreamT-150 contains heart rate (HR), blood volume pulse (BVP) and electrothermal activity (EDA) measurements from 150 sleep patients. Three models including MultiChannelCNN and MultiChannelEfficientNet and MultiChannelResNet analyzed the signals which appeared as time-series graphs. The best model proved to be EfficientNet-B0 because it demonstrated superior generalization. The pre-trained layers from EfficientNet adjusted the vulnerability of training loss which led to stable model performance. The research demonstrates sleep-derived physiological signals' usefulness for non-invasive mental health predictions which can lead to real-time monitoring systems. The upcoming research aims to boost both dataset range and better models for clinical adoption requirements.
Wednesday August 26, 2026 12:30pm - 2:30pm IST Virtual Room BGOA, India
Authors - Bhoomi C. Parikh, Zankhana Shah Abstract - Stress is any type of mental imbalance that can lead to mental disorders ranging from low to high severities which can be classified as acute and chronic stress conditions. Chronic stress leads to hyperactivation of the sympathetic nervous system, resulting in physical, psychological, and behavioural problems. Currently, there is no recognised standard for stress assessment. Thus depressive disorders leading to stress are a flight or fight response to the stimulus generated by human nervous system caused due to any unacceptable behaviour or circumstance. Throughout this response adrenaline hormones are secreted that leads to increased respiration and heart rates, along with increased muscle activity. Such type of biological alterations prime the organism for a physical response that affects human body mechanisms in terms of sleep abnormalities, digestive disorders or work imbalances in routine lives. Thus WESAD is a multimodal wearable dataset which combines both affective states(baseline, depression and happy) and other sensor modalities such as blood pressure, ECG, skin conductivity , EMG, breathing, and three-axis acceleration. There are also other classification parameters based on physiological changes which are also found in WESAD dataset and by using different types of Machine Learning Classifiers analysis is done . The algorithms with the highest accuracy can be used for developing a novel and a hybrid model which can categorise stress based on Heartrate Variability and stating HRV as a biomarker for stress detection. Both characteristics related to time and frequency of heart rate are categorized in the research study. In the context of the three-class classification based on three affect states , baseline, stress, and amusement result up to 99% was obtained. Use of two affective states like stress and amusement gave an accuracy up to 84% using DT classifier.
Authors - Payel Das, Siri Kethineedi Abstract - This study explores the elements of consumer resistance to sustainable marketing with an integrated theoretical approach adopting cognitive dissonance theory, institutional theory, and theory of planned behaviour. Although awareness of sustainability is increasing, consumers frequently do not accept sustainable products because of psychological discomfort, institutional barriers, and perceived behavioural limits. Using interpretive structural modelling (ISM), this study elucidates the hierarchy relationships among the main barriers, such as greenwashing, lack of transparency, price sensitivity, norm conformity, and instantaneous gratification. The most impactful of these drivers were identified as greenwashing and transparency deficits, both of which contribute to distrust and ultimately erode consumer confidence. Weak regulations and social norms that perpetuate these problems are demonstrated by Institutional Theory, while price premiums and limited access reduce perceived behavioural control and are described in the theory of planned behaviour. This study proposes a multi-tiered effort for policymakers and businesses to address resistance. Transparency will be enforced through independent certifications and stringent sustainability standards regulated by the regulatory frameworks. To regain consumer trust, companies must embrace true sustainability and communicate honestly. Price premiums can also be lowered through innovation, subsidies, and supply chain efficiencies to help play a role in them become more affordable. Understanding these barriers allows businesses to understand how they can build consumer trust, policymakers to enact effective regulations, and society as a whole to begin moving toward more sustainable consumption habits.
Authors - Payel Das, Sonali Bolisetty, Digumarthi Iswarya Abstract - This study seeks to identify the success enablers of omnichannel retailing by using Interpretive Structural Modeling (ISM) for building a hierarchical framework. From findings in a consumer survey with 108 consumers and expert evaluations, the research uncovers user enablers such as technological infrastructure, data analytics capability, personalization, mobile optimization, and seamless integration. The study is based on Service-Dominant Logic (SDL) and the Technology Acceptance Model (TAM) to investigate theory around foundational, operational, and experiential issues that result in customer engagement and brand loyalty. The findings underscore the critical importance of strong technological infrastructure and the use of real-time data in helping with friction reduction between digital and physical touchpoints. Using AI-powered analytics, it can improve personalization, which affects perceived system usefulness and thus, customer satisfaction. In addition, the study emphasizes the need for brand consistency and proper employee training to provide trouble-free service experiences. We also explore privacy and security concerns and their impact on consumer trust and omnichannel adoption. To policymakers, this research calls for prescriptive regulations that will protect data privacy and grow the space of technological innovation. For practitioners, it provides actionable insights to maximize omnichannel universality, improve customer pursuits, and develop sustainable total brand loyalty. In doing so, with the introduction of SDL and TAM, the study contributes to theoretical knowledge and proposes a comprehensive framework for the businesses who are dealing with the complexities of omnichannel retailing.
Authors - Ketki Kshrisagar, Chinmay Kalbhor, Sudarshan Chitte, Atharva Chivate, Pragati Chopade, Sanika Chougule Abstract - This paper describes the design and implementation of an automated control system for grain storage temperature and humidity. Operations begin using a microcontroller, Arduino Uno, and DHT11 or thermocouple sensors, for real-time environmental conditions, whereas the temperature and humidity are controlled through a Peltier module and a USB spray humidifier, to give the ideal storage conditions. Another complementing feature is an I2C LCD, which visualizes real-time parameters for the users locally to monitor environmental conditions. Besides, the system also includes a Blynk app, which allows the users to monitor and control it through a phone interface from a remote location. The main function of this is to act as a standalone and inexpensive system, which is primarily aimed at reducing grain spoilage and ensuring quality. Test results have confirmed that it was able to provide applicable environmental control for various storage scenarios
Wednesday August 26, 2026 3:30pm - 5:30pm IST Virtual Room BGOA, India
Authors - Fathima Mariya A K, Sarath S, Jyothisha J Nair, Sunitha E V Abstract - Thermal images often hide mixed signals, making accurate analysis challenging. However, segmentation and analysis are significantly compromised with the task of mixed pixels (a pixel containing the signals from several endmember sources). This study proposes a hybrid approach combining gradient-based thresholding (80 percentile and 85 percentile) and different clustering techniques (K-Means, Variational Bayesian GMM, Dirichlet Process GMM and Constrained GMM) to boost precision in mixed pixel identification. Results show that the gradient threshold has a positive effect on detection error (20.77 percentile), closely matching the values of K-Means (20.82 percentile) and Constrained GMM (20.69 percentile). The deviation from those methods to VBGMM and DP GMM is more moderate by 13.60 percentile. This study confirms the usefulness of an integrated approach for a more accurate interpretation of thermal images. Deep learning and multi-spectral will be researched to boost segmentation accuracy in the future.
Authors - Rashmi S. Bhumbare, Pallavi S. Gaikwad, Anjali M. Gutte, Araju M. Shaikh, Gayatri K. Chaudhari Abstract - In this paper, ensuring secure, transparent, and tamper-proof elections is a critical challenge in modern democratic processes. Traditional voting systems, including paper ballots and electronic voting machines (EVMs), suffer from issues such as fraud, lack of transparency, and centralized control. This project presents a Blockchain-Based Voting System, implemented as an Android application using Java/XML, with SHA- 256 encryption ensuring vote security and Firebase Realtime Database handling user authentication and data management. The system leverages blockchain technology to record votes in an immutable and decentralized ledger, preventing manipulation and unauthorized access. The implementation includes secure voter authentication, encrypted vote submission, blockchain-based integrity verification, and real-time result compilation. This approach eliminates traditional vulnerabilities such as vote tampering, duplicate voting, and unauthorized system access. Furthermore, the decentralized nature of blockchain ensures transparency, allowing voters to independently verify their votes while maintaining anonymity.
Authors - Beena B.M, Devika Madhusoodanan, Nithin Sagar, Vismaya R, Hridyalakshmi Santhosh Abstract - Energy conservation in cloud data centers remains one of the biggest research challenges today. Energy efficiency has become an important concern in the management of contemporary data centers due to the rapidly growing computational needs and the environmental impact of power consumption. This study examines various power management techniques, including Dynamic Voltage and Frequency Scaling (DVFS), Dynamic Power Management (DPM), and Adaptive Voltage Scaling (AVS), to optimize CPU power consumption. Using frequency data from historical and current CPU usage, these algorithms control CPU frequency settings and assess their impact on energy consumption and performance. The results indicate that DVFS reduces power consumption by 25-30%, DPM achieves energy savings of 28-35%, and AVS provides savings of 35-40% by dynamically adjusting both voltage and frequency. A performance matrix evaluates the power savings and utilization efficiency of these strategies to determine the most suitable approach. AVS was found to be 5-10% more energy efficient than DVFS and DPM, demonstrating its advantage in real-world applications. Furthermore, AVS exhibited the highest precision (96%) to adapt to workload fluctuations, compared to 95% for DVFS and 92% for DPM. This study focuses on adaptive power management and provides key findings on algorithmic solutions for energy efficiency in software-defined cloud infrastructures. The findings contribute to reducing data center energy consumption while maintaining performance, aligning with the UN Sustainable Development Goals by promoting sustainable and eco-friendly cloud operations.
Authors - Ajay Menon, Anjali Sivan, Navya S, Sandhya G, Astha Santhosh T Abstract - The micro, small and medium enterprise sector plays a crucial role in Kerala’s rural economy and makes a substantial contribution to socio-economic development and job creation. This study explores business continuity intentions among women-led micro enterprises in rural Kerala, using thematic analysis of in-depth interviews with six units from agro-processing, dairy and fisheries sectors. Drawing insights from qualitative data, this study uses the Theory of Planned Behaviour (TPB) to show that continuity intentions are strongly influenced by perceived behavioural control, strong family support, and positive attitudes. However, institutional inefficiencies and financial limitations create significant obstacles. This study also introduces ‘Team-led resilience’ and ‘Gendered leadership dynamics’ as critical factors, highlighting collaborative support and autonomous female leadership. These findings highlight the importance of financial literacy, access to credit, and supportive government policies, which will also help to expand the traditional TPB framework, emphasizing the importance of social and financial resilience.
Wednesday August 26, 2026 3:30pm - 5:30pm IST Virtual Room BGOA, India
Authors - M. Suresh, T. A. Alka, Aswathy Sreenivasan Abstract - The main purpose of this study is to theoretically explore the evolution of trends in teaching entrepreneurship through a Bibliometric analysis. The final number of documents selected is 1375, which are analysed through the Biblioshiny package under R programming. The results show that there are technology-related and non-technology-related trends that have evolved in teaching entrepreneurship. Major trends are happening in teaching methods, learning, courses, global reach, teamwork, and the emergence of technology trends such as artificial intelligence, virtual reality, etc. Bibliometric results draw that the major themes evolved in this domain are related to innovation trends in teaching entrepreneurship for shaping entrepreneurs for tomorrow, transformation, learning culture, technology trends, academic entrepreneurship in the covid-19 pandemic, learning types, concepts, skills required, sustainability and teaching entrepreneurship, entrepreneurialism and thinking in teaching entrepreneurship. The major future research avenues are; entrepreneurial intention; effectuation; entrepreneurship, business model innovation; innovation; digital transformation, and entrepreneurial university; academic entrepreneurship; innovation. The limitations of the research are; the Scopus database is only used for the search. Only the documents in the English language and final publication stage papers were selected. The inherent drawbacks of the bibliometric methodology may influence the results. The study offers theoretical implications for future research work including Sci-Val future research topics and practical implications by offering insights to entrepreneurs, investors, researchers, academicians, policymakers, etc. The novelty and the originality of the study are underlying in the in-depth theoretical exploration through a comprehensive literature review of the past thirty years.
Authors - Ganga S, Nitharshana P, Varun Madhusoodan, Rojalin Patri Abstract - Advancement of financial technology has resulted in the emergence of automated investment solutions, such as robo-advisors. While Gen-Z investors are typically receptive to digital innovations, their adoption of robo-advisory services remains an underexplored area. This study investigates the primary factors affecting Gen-Z's inclination to use robo-advisors, applying the Technology Acceptance Model (TAM). A quantitative methodology was utilized, with data collected from 161 respondents and analyzed through multiple regression techniques. Findings indicate that trust and attitude have a significant impact on the adoption intent of robo-advisory services in investment decisions made by Gen-Z investors. The results suggest that fostering trust and shaping positive perceptions of robo-advisors are more crucial for adoption than enhancing usability. This study contributes to fintech literature and offers insights for financial institutions and policymakers aiming to increase robo-advisory adoption among young investors.
Authors - Apolinar P. Datu, Annaliza C. Sinfuego, Garry C. Bayran, Dominic T. Urgelles, Julius R. Beltran, Rossana B. Liray, Janina Odette S. Vidallon, Erwin Joel B. Layug Abstract - The rapid transition to online learning, catalyzed by the global pandemic, has necessitated a critical examination of its implications within the context of general education. This study investigates the multifaceted factors influencing the implementation, delivery, and reception of online classes in general education programs across selected higher education institutions. Employing a mixed-methods research design, quantitative data were gathered through structured surveys while qualitative insights were obtained via in-depth interviews with students and faculty members. Results indicate that technological accessibility, digital competency, instructional quality, learner motivation, and institutional support are central determinants of effective online learning. The research highlights disparities in students’ digital readiness and access to conducive learning environments, which significantly affect their academic engagement and performance. Moreover, pedagogical adaptability and the integration of interactive tools were found to be critical in maintaining student interest and participation in virtual settings. The findings underscore the necessity for higher education institutions to invest in sustainable digital infrastructures, provide continuous faculty development programs, and adopt inclusive, student-centered online learning strategies. This study contributes to the growing body of literature on e-learning by offering empirical evidence on the challenges and enablers of online education in general education curricula. It also presents actionable recommendations aimed at enhancing the quality and equity of online instruction. In doing so, the research supports the advancement of resilient and adaptive educational systems capable of meeting the evolving demands of 21st-century learners.
Wednesday August 26, 2026 3:30pm - 5:30pm IST Virtual Room BGOA, India
Authors - Deepti Maheshwari, Shailesh Gahane Abstract - Website phishing poses a massive security threat that continues to increase in prevalence. Internet scammers exploit human faith by running imitation websites which aim to obtain confidential user information. An extensive review of multiple phishing detection techniques and hybrid detection models appears in this paper which brings together different detection methods to speed up and increase the accuracy of breaking down phishing-related websites. The paper investigates how machine learning (ML), artificial intelligence (AI) and heuristic-based approaches and anomaly detection should be implemented within hybrid systems which detect phishing behavior. Multiple studies from the literature receive analysis through which we identify their research approaches as well as their outcomes together with their limitations along with their contributions to the field. The evaluation will demonstrate how hybrid models can boost the detection of phished emails while detailing methods to strengthen model performance and flexible design and growing capability.
Authors - Rinkle Solanki, Shailesh Gahane Abstract - Surgical site infections (SSIs) represent a major concern for the healthcare industry, highlighting the need for timely intervention and effective predictive strategies.This paper presents an external validation framework for machine learning algorithms designed to identify and forecast SSIs in advance. We use sophisticated algorithms to create accurate predictive models using a variety of variables, including clinical features, microbiological data, and patient demographics. Across a range of patient demographics and therapeutic circumstances, thorough external validation is carried out. Our results demonstrate how effective this strategy is at precisely identifying SSIs, enabling prompt interventions, and improving patient outcomes. Surgical care procedures could be improved and medical expenses could be decreased by using validated models.
Authors - G Jeyashaathvee, Anish Pranav, Sindhu Chandra Sekharan, Jesline D, Ajanthaa Lakkshmanan Abstract - The challenge of extracting useful insights from unstructured data in the presence of big digital information is still present today. AI Intensive Timestamp based Summarization proposes a novel framework to automatically extract and summarize pivotal events, along with their corresponding timestamps from different data sets. Making use of Natural Language Processing and deep learning machine learning approaches, the system checks text data for temporal markers, extracts salient events and produces verbose summaries. The method proposed shall make the historical analysis easier and trend detection along with automated reporting of large dataset summaries more concise. Technique is implemented using Named Entity Recognition for date extraction and transformer models in case of summarization Experiments show that AI-based timestamp summarization is efficient in enhancing IR results and autodoc reliability. Making This Research Unique and Contributing to the much Expanding AI-driven text analysis filed, a Scalable in nature for timestamp extraction on domain wise basis.
Authors - Ankita Mehta, Shailesh Gahane Abstract - This paper discusses about, we display a profound learning-based approach bipolar clutter discovery utilizing Convolutional- Neural Systems (CNN) and Long Short-Term Memory (LSTM) systems. To assess the model’s performance, two particular datasets Twitter information and survey data were analyzed. Preprocessing steps, counting information enlargement, normalization, and the application of the Adam optimizer, were joined to upgrade the model’s adequacy. The model’s exactness and misfortune were measured for both datasets, and it was watched that the survey dataset given superior execution, yielding higher precision and lower misfortune compared to the Twitter dataset. These discoveries propose that the questionnaire-based information may be more reasonable for solid bipolar disorder location within the given show. The inquire about illustrates the potential of combining CNN and LSTM for mental wellbeing examination, highlighting the significance of information determination in accomplishing ideal comes about.
Thursday August 27, 2026 9:30am - 11:30am IST Virtual Room BGOA, India
Authors - Sarika G. Songire, Deepa S. Deshpande Abstract - Deep learning algorithms have completely transformed medical diagnostics by enabling accurate and efficient identification of brain tumors. Brain disorders often arise due to increased in excessive and improper cell counts, which can damage neural structures and, in severe cases, lead to malignant brain cancer. The reducing mortality rates requires prompt intervention and early discovery. This research work presents an architecture of deep neural network specifically designed for the purpose of detecting brain tumors from MRI images. The proposed model is evaluated against existing research using a similar dataset to assess its effectiveness. For comparison, performance parameters including area under the curve (AUC), recall, accuracy, precision, and loss are used. According to experimental results, the suggested CNN model accomplishes 98.61% accuracy, 99.70% AUC, 99% of both precision and recall, and 0.41 loss in a data set of 3,264 MRI images. These findings indicate that the suggested model surpasses current models and provides a dependable and effective technique for prompt brain tumor diagnosis.
Authors - Ayush Itkhede, Aryan Fulsunge, Sarthak Bomanwar, Sujal khobragade, D.M.Shinde Abstract - The integration of embedded systems into brick manufacturing has led to significant improvements in efficiency, quality control, and resource management. This paper presents an embedded system designed to optimize brick production by monitoring critical parameters such as temperature, humidity, and material composition in real-time. The system achieved a 32% reduction in production cycle time, from 6.5 hours to 4.4 hours per batch, and improved material handling speed by 41%, from 120 kg/hour to 169 kg/hour. Quality control metrics showed a 78% reduction in defect rates, from 5.2% to 1.15%, while energy consumption decreased by 29.3%, from 17.4 kWh to 12.3 kWh per 1,000 bricks. The system also reduced waste by 64%, from 8.9% to 3.2%, and improved material utilization rates from 83.5% to 94.8%. These results demonstrate the potential of embedded systems to revolutionize brick manufacturing, making it more sustainable and cost-effective.
Authors - Kamini Solanki, Rahul Vaghela, Jay Panchal, Anjali Mahavar, Jaimin Undavia, Nilay Vaidya Abstract - Diabetes is a chronic illness that affects the body's ability to metabolize glucose, which is the sugar that provides energy to the body. Type 1 and type 2 diabetes are the two main types of the disease. The immune system of a person with type 1 diabetes attacks and destroys the pancreatic cells that make insulin, which causes blood sugar levels to rise. Type 1 diabetes symptoms include excessive thirst, frequent urination, extreme hunger, weight loss, fatigue, impaired vision, delayed healing, tingling or numbness in the hands and feet, and recurring infections. AI algorithms may be used to collect and analyze medical data to support early diagnosis and treatment. Typically developing in childhood or adolescence, type 1 diabetes can be managed with the injection of insulin or an insulin pump. A person with type 2 diabetes either develops an inability to use insulin or ceases making enough of it to regulate blood sugar levels.
Authors - Dipali Wankhade, Shailesh Gahane, Mrunal Meshram Abstract - Oral cancer is one of the fatal diseases present in society. Its late-stage diagnosis puts it under the list of diseases with high mortality rates. Deep learning has conceptualized the revolution in the field of medical imaging with impressive depth and accuracy in diagnosis and early detection. This review highlights the advancements in deep-learning-based methodologies for oral cancer detection and classification, including convolutional neural networks (CNNs), recurrent neural networks (RNNs), deep reinforcement learning (DRL). Multi-modal learning and hybrid models combining histopathological and radiological data have increased the precision of tumor segmentation and subtype classification. The observational data will help provide insights into the factors responsible for explaining the model's decisions and the associated risks, which are most relevant for potential applications. The development of transfer learning and self-supervised learning also seems to have significantly solved some of the most serious challenges regarding the volume of clinical data. Future studies can harness standardized practices for data collection, deploy technically sound explainable AI frameworks, and conduct clinical validations as the closing gap between tremendous strides made in deep learning and practical real-world implementation. This review provides a wide-ranging overview of such AI-driven methodologies, focused on the critical challenges and future directions for improvements in early oral cancer detection and reduced mortality rates.
Authors - Ankita Mehta, Shailesh Gahane Abstract - In this research, we focus specifically on mental disorder which is bipolar disorder using machine learning techniques, utilizing a simple dataset from Kaggle for training and evaluation. The study involves applying various ML models, including R. Forest, XG-Boost, and S. Vector Machines (SVM), on the dataset. We assess the performance of these models using evaluation MSE (how far actual value to predicted value), Precision, and Fl Score to determine their effectiveness in predicting bipolar disorder. However, through extensive experimentation, we found that the combination of (CNN) and (LSTM) networks outperformed the other algorithms, achieving an overall accuracy of 95%.
Thursday August 27, 2026 9:30am - 11:30am IST Virtual Room BGOA, India
Authors - Vijeta V Shettar, SR Nirmala, Satish Chikkamath, Suneeta V Budihal Abstract - This study investigates the performance of various speaker recognition models, including SpeakerNet, TiTANet Small, and TiTANet Large, using the IITG-MV dataset for speaker verification tasks. The preprocessing steps involved resampling, noise reduction, segmentation, and volume normalization to prepare the audio data for input into the models. The models were evaluated based on their ability to correctly verify whether two audio samples belong to the same speaker or not. The evaluation metrics, derived from the confusion matrix, revealed that SpeakerNet outperformed both TiTANet variants, achieving an accuracy of 85%, while TiTANet Small and TiTANet Large achieved accuracies of 75% and 80%, respectively. Despite lacking graphical visualizations, the confusion matrix provided a comprehensive view of the models’ performance, showing how each model handled correct and incorrect speaker match predictions. The results highlight that SpeakerNet is the most effective model for speaker recognition in this setup, demonstrating superior accuracy and robustness in identifying speaker-specific features. These findings can guide future research in optimizing speaker recognition models for real-world applications involving speaker verification.
Authors - Madan Kumar Sharma, Nadir Kamal Salih Idries, Abdullah Said Alkalbani, Satyanarayana Degala, Gopal Rathinam, Ankit Sharma Abstract - Multi-resonance (MR) Microwave sensors have emerged as a promising sensing device for mineral-based materials characterization due to their high sensitivity and precision. This research presents a novel microwave sensor for mineral-based material characterization. The sensor's resonating structure comprises a 3 × 3 array of circular-shaped complementary split-ring resonators (CSRR) coupled with four rectangular defected structures etched around the CSRR array. This unique configuration enhances electromagnetic interaction with the material under test (MUT), leading to precise characterization based on S-parameter analysis. To validate the sensor’s efficacy, simulation-based investigations were conducted on the mining-based materials, including chrome, copper, and quartz. The obtained results demonstrate distinct resonance shifts and attenuation variations corresponding to each mineral, highlighting the sensor’s capability to differentiate and analyze their dielectric properties. The proposed MR-sensor design provides a robust and efficient method for non-destructive material characterization, offering potential applications in the mining industry, quality control, and geophysical exploration.
Authors - Geethu Lakshmi G, P. Nagaraj, P. Chinnasamy Abstract - Lung cancer detection constitutes a paramount process in the diagnosis and management of one of the predominant contributors of cancer-induced humanity worldwide. The significance of early screening is underscored by its essential role in enhancing survival rates through the identification of disease during a stage amenable to treatment. Diagnostic methodologies, including imaging modalities, are routinely utilized for diagnosis. Furthermore, advancements in the realm of molecular biology have facilitated the emergence of biomarkers and genetic examines, thereby enabling a more accurate identification of lung cancer. The prompt and defined detection of lung cancer facilitates timely therapeutic interventions, which significantly influence both the efficacy of treatment and overall outcomes for patients. The primary objective of this research is to develop an optimization-based hybrid deep learning methodology for the detection of lung cancer. The initial phase involves pre-processing of input images through techniques such as color space transformation, data augmentation, resizing, and normalization. Subsequently, features derived from Slime Mould Algorithm-based Convolutional Neural Network (SMA-CNN) are employed for the detection of lung cancer, with CNN being trained utilizing SMA, extracted from pre-processed images. Finally, the Squeeze-Inception V3 model, which integrates SqueezeNet and Inception V3, leverages SMA to train the classifier. Consequently, the proposed SMA-based hybrid SqueezeNet-Inception V3 is utilized to classify instances as normal or abnormal. Empirical results designate that SMA-based hybrid SqueezeNet-Inception V3 attained an accuracy of 97.3%, a specificity of 96.1%, and a sensitivity of 98%, thereby underscoring its efficacy in the detection of lung cancer.
Authors - Narayan Gupta, Pawan Kumar, Prince Kumar Singh, Priyabart Kumar, Parampreet Kaur Abstract - For investors, accurately predicting stock market prices is a critical part of financial analysis. This work studies a wide variety of machine learning algorithms specifically designed for the task of predicting stock prices using a variety of techniques and the latest technologies available as of the time of study. The study critically compares a suite of algorithms, including Support Vector Regression (SVR), Random Forests, Decision Tree models, and Long Short-Term Memory (LSTM), each with differing strengths. Moreover, it investigates a variety of approaches that are focused on understanding the complex connections found in the past price data. The dataset used for this study includes extremely long-term stock price representatives over a time range from 2010 to 2024 for Tata Consultancy services (TCS), containing a wide range of information, including opening and closing trading prices, trading volumes, and a variety of calculated indicators reflecting market behaviour. To understand the first experiment carried out for this study, it can be seen that in combination mode, for the best-performing model, Random Forest has the best interpretability. In addition, both Support Vector Regression (SVR) and Decision Tree algorithms deliver both impressive short-term prediction results and clear explanations behind their decisions.
Authors - Sandeep Shinde, Samarveer Moray, Aditya Sakhare, Prathamesh Salokhe, Kedar Sathe Abstract - The exponential growth of digital documents, particularly PDFs, presents significant challenges in efficient information retrieval and extraction. Traditional methods often struggle with the complexity and variability inherent in large PDF documents. Recent advancements in Natural Language Processing (NLP), especially Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG), offer promising solutions. This paper presents a comprehensive system for efficient information extraction from large PDFs using RAG and LLMs. We propose a robust and scalable pipeline addressing challenges such as document segmentation, dynamic retrieval, and response contextualization. Through extensive experiments across multiple domains—including legal analysis, technical documentation, and scientific literature—we demonstrate that the proposed methodology significantly outperforms existing approaches in terms of accuracy, scalability, and efficiency. Our research lays the groundwork for integrating RAG and LLMs in various domains, offering a valuable tool for extracting knowledge from complex documents.
Authors - Rahul Pethe, Parag Puranik, Abhay Kasetwar Abstract - Wireless Sensor Networks (WSNs) have been designed and developed by numerous researchers over time, evolving based on emerging demands and environmental conditions. Over the years, various enhancements and improvements have been proposed, with multiple protocols introduced to address design challenges. In parallel, Mobile Ad Hoc Networks (MANETs) have witnessed tremendous growth in this wireless era. While many researchers have focused on protocols like DSR and hybrid approaches for energy-efficient clustering, these solutions have often proven to be time-bound and context-specific. To overcome these limitations and enhance the performance and lifetime of WSNs, we propose a new scheme based on the AODV (Ad hoc On-Demand Distance Vector) protocol within a mesh networking framework. Our approach achieves optimal results, including 100% throughput, minimal jitter, and significant improvements in network life-time.
Authors - Pranav Mittal, Prerna, Dhruv Bansal, Nikhil Panwar, Krish Tyagi Abstract - Stock market prediction is an intricate process in financial analysis, as its main aim is to predict the price trends and thus help to make trading strategies. But the stock market is so unpredictable with various factors affecting it, and thus predicting whether the value will rise or not becomes a difficult task. This study aims to predict the stock market prices with historical data that will be achieved via Deep Learning techniques in particular LSTM networks. Due to its strength in learning long-term dependencies and preserving the sequence information, LSTM (one of the variants of RNNs) is ideal for time-series data. Our approach leverages a dataset of daily stock prices from various financial indices over multiple years. The data is preprocessed using normalization techniques to improve model accuracy. The LSTM model is then compared to a traditional feed-forward neural network to demonstrate the superiority of LSTM in predicting shortterm stock trends. The models are optimized using the Adam optimizer, The results indicate that LSTM significantly outperforms the conventional models in forecasting accuracy. Moreover, this research introduced a hybrid LSTM-CNN method to extract features and prediction firmly. This study will contribute to financial forecasting by utilizing deep learning techniques and real trading scenarios. The research work was carried out using the programming language known as Python, deep learning tools such as TensorFlow and Keras, data management libraries such as Pandas and NumPy, and data representation software such as Matplotlib. It was found that the LSTM model has a much higher level of success when time series patterns are approximate than any other type of neural network, hence stock price predictions are more accurate. This study helps in how LSTM networks can be useful for forecasting in finance hence would be helpful to traders and market analysts.
Authors - Sheetal Phatangare, Komal Potdar, Yash Mahajan, Mandar Pandagale, Vivek Nikam Abstract - University students frequently encounter challenges in retrieving relevant academic information due to the limitations of traditional search engines. This research introduces KnowledgePilot, the first 1-bit Large Language Model (LLM) specifically designed to support university students. Leveraging the BitNet b1.58 architecture, which employs ternary parameterization (-1, 0, 1), KnowledgePilot achieves high performance with reduced computational costs, making it both resource-efficient and fast. The system integrates Retrieval Augmented Generation (RAG) pipelines, enabling it to access external academic data sources, thus minimizing hallucination issues common in LLMs and providing accurate, context-specific responses. The research also encompasses the development of tools for file conversion, dataset creation, model pretraining and fine tuning. Comprehensive evaluations will measure the system’s performance and user satisfaction, demonstrating its potential to significantly enhance student access to academic resources, while setting the stage for future advancements in low-bit AI technologies for education.
Authors - Yash Chavan, Arnav Sonawane, Arpit Pattiwar, Aditya Nagdive, Kaushalya Thopate Abstract - In today's fast-moving world, consumers rely on packaged foods. This is extremely important for easy access to detailed and personalized nutritional information. This project focuses on developing mobile applications for barcode scanning. This includes extensive food details, including ingredients, nutritional value, allergen warnings, and personalized consumption recommendations based on a person's health. Applications written with Python and Kivy provide a seamless user experience, allowing individuals to scan barcodes and upload images of ingredients to extract and analyze related information. Additionally, it includes optical character detection (OCR) using Tesseract, which extracts text from photos to allow users to analyze the ingredient list and nutritional name, even if barcode scans are not possible. By taking into account user nutritional limitations or illnesses such as diabetes, lactose intolerance, or gluten sensitivity, this application provides tailor-made health advice and helps individuals make found food decisions appropriately. A secure user authentication system improves the experience by storing your preferences and receiving recommendations created by tailors. The main goal of this project is to enable consumers to choose food in real time and promote healthier consumption habits. The combination of barcode scanning, OCR, and a structured database causes applications to close the gap between the complexity of food indicators and user understanding. Future improvements include mechanically learning-based ingredients, integration into real-time product databases, and expansion of several platforms beyond Android. This initiative represents an important step in using technology to improve consumer health awareness and ensure safer and sounder decisions for food consumption.
Thursday August 27, 2026 12:30pm - 2:30pm IST Virtual Room BGOA, India
Authors - Evangeline R C, Krupa Nirmal, Laasya P, Aishwarya K Abstract - Real-time pedestrian trajectory prediction is essential for enhancing safety and urban mobility, particularly in dense and dynamic environments. This paper introduces a video data processing system that accurately predicts pedestrian movement by analyzing sequences of video frames in real time. The system effectively handles challenges such as overlapping individuals, partial occlusions, and diverse walking behaviors, making it suitable for real-world deployment. The architecture is designed to be both modular and scalable, allowing for seamless integration into various applications such as traffic management, urban planning, and pedestrian safety enhancement. A user-friendly interface provides real-time visualization of the predicted trajectories, enabling accessibility for both technical and non-technical stakeholders, including urban planners and public safety officials. Extensive experiments conducted on multiple diverse datasets demonstrate the system’s reliability and accuracy across various conditions, including crowded scenes and irregular pedestrian movement. The system successfully captures complex behavior patterns and provides predictive insights that can help reduce pedestrian-related accidents. This research contributes significantly to the field of intelligent pedestrian monitoring systems. By combining real-time responsiveness with accurate trajectory prediction, the proposed system supports the development of smarter and safer urban infrastructure, fostering proactive decision-making and improved pedestrian safety in modern cities.
Authors - Nisha Dubey, Randeep Singh Abstract - Facial profile classification and acknowledgment have different applications in security, perception, and identity affirmation. This paper proposes a novel approach utilizing Convolutional Neural Frameworks (CNNs) to classify and recognize facial profiles. The proposed system utilizes a significant CNN designing to remove solid highlights from facial profiles, taken after by a classification layer to recognize profile classes (e.g., cleared out, right, frontal). The illustrate is ready on a tremendous dataset of facial profiles and finishes tall accuracy in classification (95.2%) and affirmation (92.5%) errands. Test comes around outline the system's quality to assortments in lighting, pose, and expression. Besides, the proposed system outflanks existing techniques in facial profile classification and affirmation. This work contributes to the movement of facial examination development, engaging its course of action in real-world applications such as identity affirmation, get to control, and perception. In particular, facial profiles provide an interesting challenge due to the distinctive variety of lighting, attitude, expression and disorders. Furthermore, large data records and accessibility of arithmetic violations have made it possible to prepare violent .CNN models for facial profile classification and detection
Authors - Meena Rani, Randeep Singh Abstract - The exponential growth of smart devices and advanced image editing tools has made detecting and localizing image forgeries critical for ensuring digital content integrity. This paper focuses on developing a robust and scalable model for passive image forgery detection using convolutional neural networks (CNNs). Leveraging datasets like CASIA1 and MICC-F220, the study aims to identify tampered regions in digital images by analysing noise patterns, pixel-level anomalies, and compression artefacts. The proposed methodology integrates preprocessing, model training, and validation using diverse datasets to enhance detection accuracy and scalability. Compared to traditional techniques, the deep learning-based approach shows significant improvements in detecting complex forgeries, including splicing and copy-move manipulations. Applications of this research extend to digital forensics, media authentication, and cybersecurity. The findings underscore Deep learning systems' show promise to tackle new issues in picture forgery detection and localization
Authors - Kritika Benjwal, Rishika Agrawal, Rashi Gupta, Dinesh Kumar Saini Abstract - The ublic distribution system (PDS) plays a vital role in eradicating hunger and ensuring food security across the world[1]. However, there are certain challenges like beneficiary identification, inconsistent transaction, diversion of grains during procurement at different stages to the open market, ration shop owners selling subsidized goods at higher prices and most importantly the paper focusses on supply chain leakages.[2] This paper proposes a conceptual model for implementing blockchain in PDS and eradicating all the supply chain leakages and making PDS fair[3]. It first focusses on operations of PDS, then assessing all the possible loopholes and implementing blockchain for removing these loopholes. It proposes the idea in which it leverages the benefits of using smart contract and consortium-based ecosystem that can bring efficiencies in PDS.
Authors - Sandhya Borkar, Shital Patil Abstract - The rapid development of the Internet of Things (IoT) has made the provision for innovative solutions in various sectors, including fuel management. The research presents the design and analysis of a smart IoT-based system for fuel dispensing, aimed at improving the efficiency, transparency and security of fuel distribution. The proposed system uses microcontrollers, sensors and real-time data communication technologies to automate fuel dispensing, monitor fuel levels and prevent theft or misuse. Key components include flow sensors to measure fuel output, RFID modules for secure user authentication and cloud-based platforms for remote monitoring and control. Additionally, mobile applications provide users with instant transaction records, fueling history and alerts. The system undergoes performance evaluation to ensure precise fuel measurement and seamless data synchronization. The results demonstrate significant improvements in operational efficiency and customer satisfaction, reducing manual errors, fuel theft and fraud, operational downtime, high maintenance and labor cost. This smart fuel dispensing system offers a scalable and cost-effective solution suitable for fuel stations, logistics companies and industrial applications, contributing towards smarter resource management and enhanced energy distribution practices.
Authors - Pravin Game, Shubham Bhingardive Abstract - Deepfake technology, which enables manipulation of images and videos, poses serious threat to media integrity and cyber security. Existing detection models often struggle with accuracy due to data complexity and variability. This study introduces a hybrid deepfake detection model that combines MobileNetV2, EfficientNetB7 and Vision Transformer (ViT) to enhance feature extraction and classification. ViT provides strong pattern recognition, EfficientNetB7 offers scalable accuracy and MobileNetV2 ensures lightweight processing. The model is trained on a publicly available datastet from Yonsei University, consisting of real and fake facial images. Techniques such as data augmentation and image resizing improve generalization. Experimental results show that the proposed model achieves 94.64% accuracy, 93.55% precision, 96.67% sensitivity, 92.31% specificity and 95.08% F1 score outperforming precious methods. These improvements ase statistically significant (p < 0.05). The results highlight the effectiveness of multi-model fusion for robust deepfake detection offering a scalable and reliable solution for applications in digital forensics, information verification and cybersecurity.
Authors - Pravin Game, Shubham Bhingardive Abstract - The correct identification of heart disease is essential for successful treatment and management. In this work, we assess different machine learning algorithms' predictive power for diagnosing heart disease. On the dataset, we employed the method of principal component analysis (PCA) to choose features, we got top 9 principal components out of 13 features. Then, applied the hippopotamus optimization algorithm on that 9 principal components then trained and tested the model on eight different algorithms: Bagging, Boosting, Naive Bayes, K - Nearest Neighbors (KNN), Random Forest, Decision Tree, Support Vector Machine (SVM), and Logistic Regression(LR). The algorithm’s accuracy ranged from 86.81% to 94.53%, The most accurate methods were SVM, KNN and random forest. These findings show that machine learning algorithms may be able to help with heart disease and focus on the need of choosing suitable algorithms for exact and trustworthy clinical decision-making. Future research will concentrate on using sophisticated on feature selection and ensemble learning strategies to further increase model accuracy.
Authors - Ganesh Shivaji Pise, A D Londhe, Hrushikesh Jaivant Joshi, Bhagwan Dinkar Thorat, Yashita Parikshit Mahalle, Pankaj Chandre Abstract - Data leakage poses a significant threat to modern security systems, leading to unauthorized access and privacy breaches. Deep learning models have shown promise in detecting such anomalies; however, their black-box nature raises concerns regarding trust and interpretability. This paper explores the role of Explainable AI (XAI) in enhancing transparency and trust in deep learning-based data leakage detection. Various explainability techniques, including SHAP, LIME, and Grad-CAM, are integrated into a security framework to provide interpretability while maintaining detection accuracy. The proposed architecture bridges the gap between AI-driven security solutions and human decision-making, enabling security analysts and compliance officers to make informed assessments. Additionally, the study evaluates different XAI approaches based on accuracy, interpretability, and scalability to identify optimal techniques for real-world security applications. The findings highlight the importance of balancing explainability with performance to ensure robust and trustworthy cybersecurity solutions.
Authors - Priya Surana, Soham Jadhav, Janhvi Jathot, Arnav Joshi, Roshani Kadam Abstract - Landslides are natural disasters posing great risks to life, infrastructure, and the environment. Timely and accurate predictions are highly beneficial to reduce such impact. The advent of machine learning (ML) and deep learning (DL) has significantly improved the state of landslide prediction models. The review outlines the various ML and DL techniques adopted for landslide prediction and gives a brief account of methodologies, applications, benefits, and limitations. This is mainly the melding of ML and DL techniques, such as Random Forest (RF), Support Vector Machines (SVM), Convolutional Neural Networks (CNN), and Long Short-Term Memory (LSTM) networks, for the enhancement of the predictive ability of such models. Key challenges in landslide prediction, such as data availability, model interpretability, and computational complexity, alongside future directions, will be discussed to contribute to the robustness of landslide prediction models. Finally, inferences will be drawn as to the significance of hybrid ML-DL approaches in pushing forth landslide prediction models into better accuracies and reliability (Khuc, T.D., et al, 2023)(Wu, X., et al, 2023).
Thursday August 27, 2026 3:30pm - 5:30pm IST Virtual Room BGOA, India
Authors - A.Punidha, E.Arul, E.Yuvarani, S.Rajasakaran Abstract - With the rapid expansion of Internet of Things (IoT) and smart device ecosystems, security threats such as malware attacks have become a critical concern. Traditional signature-based malware detection methods struggle to detect evolving and polymorphic threats, necessitating the development of intelligent, data-driven cybersecurity mechanisms. This study proposes a novel malware detection framework that integrates optimized feature engineering and deep neural networks (DNNs) to classify malware in smart devices with high precision. The approach focuses on behavioral feature extraction, including API call sequences, network activity logs, and application permissions, followed by feature selection techniques to reduce dimensionality while retaining key discriminative attributes. A comparative analysis of various machine learning (ML) models, including Random Forest, Support Vector Machine (SVM), and Deep Learning models, demonstrates that the proposed feature engineering-enhanced DNN model achieves 96.1% accuracy, outperforming conventional methods. Extensive experimentation on a real-world dataset of 10,000 smart device applications showcases the robustness and scalability of our framework. This research contributes to enhancing security in smart environments by providing an adaptive and computationally efficient malware detection system.
Thursday August 27, 2026 3:30pm - 5:30pm IST Virtual Room BGOA, India
Authors - Jolou Vincent M. Jala, Everly A. Nacalaban, Nenon Roy A. Sandinao, Ryan Boyd D. Origines, Randy Joy M. Ventayen, Neilson D. Bation Abstract - Artificial intelligence (AI) has completely transformed enterprises and organizations all over the world due to its propensity to spur innovation and mimic operational efficiency (Jala, J.V.M. et al., 2024). Through its capability to boost learning outcomes, promote inclusion, and streamline operations, artificial intelligence (AI) is revolutionizing higher education. This study investigates the influence of artificial intelligence in higher education in the Philippines. The study specifically seeks to understand how artificial intelligence (AI) can be used in higher education institution in terms of personalized learning amidst large class sizes, access to education in rural areas, solving job skills mismatch, modernizing administrative procedures in inadequate resources institutions, artificial intelligence powered innovation and research, challenges of embracing artificial intelligence such as digital literacy and infrastructure, social and ethical implications and resistance to change and faculty development Moreover, this work also examines the ethical considerations in employing arti-ficial intelligence in higher education in the Philippines, precisely in terms of security and data privacy, fairness and algorithmic bias, digital divide, human supervision and accountability, autonomy and consent, influence on staff and faculty roles and intellectual property and academic integrity. To realize this objective, the proponents essentially examined 170 publications in the literature that were indexed by Scopus to look at artificial intelligence in the context of higher education. This finding highlights artificial intelligence’s essential role in embracing challenges and improving higher education in the Philippines while emphasizing ethical considerations such as fairness and data privacy. (Jala, D.J.V., 2025).
Authors - Kishan Raj, Samyuktha Vimal, V Shini Abstract - In recent years, the Indian online fashion e-commerce industry has experienced significant transformations due to fast-paced technology, consumer behavior changes, and a growing e-commerce landscape. Competition is fierce, especially in the e-lifestyle market as its payoff will exceed $30 billion by 2025, developing brand equity understandably becomes a key factor in maintaining long-term success. It is understood that brand equity is a significant factor as it directly affects consumers' trust, buying decisions, and consumer retention, which are also vital elements in preserving brand equity in an industry where distinctions matter. This study's goal is to examine the impact of marketing mix variables 7Ps including Personalization, which has just recently emerged as a particular critical factor in the e-commerce business, on brand equity in the Indian online fashion e-commerce industry. Quantitative research methods were adopted to analyze data provided by consumers regarding the influence of such factors. The analyses indicate that Personalization and Place affect brand equity substantially, whereas traditional elements such as Price and Promotion do not exert such influence in online fashion retailing. Indian consumers seem increasingly to make a shift toward a digital-first shopping experience, the marketing strategies of brands must follow suit by creating engaging, personalized, and innovative interactions. These research findings will provide some strategic recommendations to online fashion retailers, marketers, and industry gurus seeking to consolidate their brand positioning in this ever-evolving and highly competitive marketplace.
