IJSRET » Blog Archives

Author Archives: Kajal Tripathi

The Generative AI Industry is Flawed!

Uncategorized

The Generative AI Industry is Flawed!/strong>
Authors:-Isha Syed, Aryan Purohit, Yash Malusare

Abstract-Generative Artificial Intelligence (GenAI) has evolved rapidly, creating transformative opportunities across sectors, particularly in healthcare and marketing. Despite the promise of improved patient care, streamlined medical workflows, and enhanced customer engagement, GenAI faces significant challenges. Key obstacles include high computational costs, data-privacy concerns, and ethical accountability in content generation. Moreover, the open-source initiatives by leading firms like Meta have intensified competition, pushing GenAI models toward commoditization, impacting revenue structures and sparking a “race to the bottom” in pricing. The market is further complicated by monopolistic dependencies on critical hardware providers, particularly Nvidia, which dominate GPU supplies essential for AI training. With a rapidly growing market projected to reach trillions by 2030, the industry must navigate these barriers to realize the full potential of GenAI. This study explores GenAI’s current applications, fiscal and ethical challenges, and the strategic imperatives needed to foster sustainable, profitable growth within an increasingly crowded and commoditized industry landscape.

DOI: 10.61137/ijsret.vol.10.issue6.325
55

Published by:

The Impact of Behavioural Features on Predicting Academic Success: A Machine Learning Approach

Uncategorized

The Impact of Behavioural Features on Predicting Academic Success: A Machine Learning Approach
Authors:-Nidhi Kataria Chawla, Chietra Jalota

Abstract-To discover hidden patterns from educational data, researchers are developing methods by using educational data mining. Dataset and its features/attributes determine the eminence of data mining techniques. Student’s academic performance model by using a new class of features i.e., behavioural features was built in this research paper. These are significant features as they are associated with the learner interactivity in e-learning system. Data was collected from an e-Learning system called Kalboard 360 using Experience API web service called (xAPI). After data preprocessing and feature selection, machine learning algorithms such as Decision Tree, Support Vector Machine and Artificial Neural Network were used to build the model. It is clearly visible from the results that there is a sturdy association between learner behaviours and its academic achievement. Results with above-mentioned classification methods using behavioural features attained up to 25% enhancement in the accuracy as compared to the results when same classification methods were applied on the data set without behavioural features.

DOI: 10.61137/ijsret.vol.10.issue5.324

Published by:

Surface Water Cleaning Robot (SWCR) for Sustainable Environmental Protection

Uncategorized

Surface Water Cleaning Robot (SWCR) for Sustainable Environmental Protection/strong>
Authors:-T. Anilkumar, V. Abhiram, K. Sampath Kumar, R. Yashwanth Sai Ganesh, U. Bhavani Prasad, P. Aditya Raj

Abstract-The emphasis of the project is centered upon the designing and advanced construction of an ecological water cleaning system that has wireless control features which integrates advanced environmental monitoring and robotics that is operated remotely towards achieving environmental sustainability. Consequently, due to the growing concern of water pollution, there is an increasing demand of deploying an easy system which will eliminate the waste and pollutants from the water bodies in an efficient manner. The system consists of a and a rotary bracket which consists of a substantial floating platform mounted on a 12V battery, four 500 RPM DC motors and L298N motor driver for river surface navigation. The operation of this device centers around the use of an ESP32- CAM module which acts as a camera that streams real time images to the operator for effective monitoring of the device and waste collection process. This system solves the problem of debris reduction in water bodies and enhances water reclamation curbing the risks of cash intensive manual cleaning. If the technology comes into practice it is going to improve environmental protection by introducing a new approach to environmental management and promoting sustainability strategies in the protection of water bodies.

DOI: 10.61137/ijsret.vol.10.issue5.323
55

Published by:

Real-Time River Health Monitoring using Custom Dataset, YOLOv8, and Crowdsourced Solutions: A Comprehensive Review

Uncategorized

Real-Time River Health Monitoring using Custom Dataset, YOLOv8, and Crowdsourced Solutions: A Comprehensive Review/strong>
Authors:-Assistant Professor Mrs. Vandana Navale, Yashi Solanki, Riddhi Khot, Pradnya Nalawade, Aakanksha Wadekar

Abstract-Water pollution is still a global problem, especially in urban waters. The routine process of monitoring water bodies is slow and resource intensive. This paper reviews modern approaches to monitoring water health using proprietary data, deep learning models such as YOLOv8 for pollution detection, and public service centers for initiating cleanup projects. The review describes the collection of user data and examines how the proposed system combines public research with machine learning techniques to develop good and measurable solutions to problems. It also investigates the role of public services in promo ng knowledge and environmental financing.

DOI: 10.61137/ijsret.vol.10.issue5.322
55

Published by:

Agriculture Sustainability: A Comprehensive Review

Uncategorized

Agriculture Sustainability: A Comprehensive Review/strong>
Authors:-Rajat Kumar, Gurshaminder Singh

Abstract-Agricultural sustainability is essential for meeting global food demands, safeguarding the environment, and ensuring economic stability. This review delves into the various dimensions of sustainable agriculture, covering practices, technologies, policies and their collective effects on biodiversity, soil heath, and climate resilience. A central focus is on blending traditional agricultural knowledge with contemporary innovations to create sustainable practices that support biodiversity and soil vitality while adapting to climate challenges. The role of agroecology, which emphasizes ecological principles in agricultural settings, is highlighted as a key approach in promoting biodiversity and minimizing environmental impact. Additionally, the review stresses the importance of robust policy framework that support sustainable practices, ensure resource management, and address climate impacts. The paper also examines the main challenges hindering sustainable agriculture, such as resource depletion, land degradation, water scarcity, and economic pressures. These issues are interconnected with socio-economic factors, including access to resources, income stability, and social equity, all of which shape agricultural sustainability and impact communities reliant on farming. Resource depletion and land degradation are particularly emphasized, as they reduce productivity and soil health, leading to less resilient agricultural systems. To combat these challenges, the review suggests innovative solutions aimed at fostering resilience and sustainability. These include precision agriculture, which leverages data and technology for efficient resource use, crop diversification to reduce vulnerability to climate shifts, and regenerative farming practices that enhance soil health and sequester carbon. The potential of agroecology and regenerative practices is especially emphasized for their ability to restore ecosystems while boosting productivity. Policy interventions, particularly those that support sustainable practices, incentivize research and development in agro-innovations, and provide farmers with training and resources, are crucial for advancing sustainable agriculture.

DOI: 10.61137/ijsret.vol.10.issue5.321
55

Published by:

Adoption of Artificial Intelligence: Benefits, Challenges, and Future Prospects

Uncategorized

Adoption of Artificial Intelligence: Benefits, Challenges, and Future Prospects/strong>
Authors:-Malvika Singh

Abstract-Artificial Intelligence (AI) has emerged as one of the most transformative technologies of the 21st century, reshaping industries, driving operational efficiencies, and fostering innovation. The adoption of AI spans numerous sectors, such as healthcare, finance, retail, and manufacturing, where it is optimizing processes, enhancing decision-making, and delivering personalized services. However, while AI adoption holds significant promise, it also presents notable challenges, including ethical concerns, data privacy issues, skills gaps, and high implementation costs. This paper explores the advantages of AI adoption, the barriers it faces, and future trends that could shape its progression. By examining case studies and identifying key trends, this paper aims to provide a comprehensive overview of the adoption of AI and its potential for transforming industries worldwide.

DOI: 10.61137/ijsret.vol.10.issue5.320
55

Published by:

Energy Theft Detection in Smart Grids Using Graph Neural Networks (GNNs)

Uncategorized

Energy Theft Detection in Smart Grids Using Graph Neural Networks (GNNs)/strong>
Authors:-Assistant Professor Dr. Pankaj Malik, Himisha Gupta, Anoushka Anand, Siddhesh Bhatt, Devansh Gupta

Abstract-Energy theft poses significant challenges to smart grid operations, leading to substantial financial losses and grid instability. Traditional machine learning approaches often fall short in detecting energy theft due to the complex and interconnected nature of smart grid systems. This paper proposes a novel approach to energy theft detection using Graph Neural Networks (GNNs), leveraging the inherent graph structure of smart grids. By representing the grid as a graph, where nodes correspond to smart meters and transformers, and edges represent electrical connections, GNNs capture both the local consumption patterns and the relationships between grid components. The proposed model aggregates node and edge features to identify anomalous consumption behaviors indicative of energy theft. We apply both Graph Convolutional Networks (GCN) and Graph Attention Networks (GAT) to enhance detection accuracy by considering both the structural and consumption-related features of the grid. The model is trained and evaluated on real-world and simulated smart grid datasets, showing improved performance over traditional classification models such as support vector machines and random forests. Evaluation metrics including precision, recall, and F1-score demonstrate the model’s robustness, even in the presence of noisy data and imbalanced class distributions. This research highlights the potential of GNNs to enhance energy theft detection in smart grids, providing a scalable and interpretable solution that can adapt to evolving grid conditions. Future work includes expanding the model to incorporate temporal data for real-time detection and exploring reinforcement learning for adaptive theft prevention strategies.

DOI: 10.61137/ijsret.vol.10.issue5.319
55

Published by:

Dietary Interventions for Speech Delay and Hyperactivity in a Child with Machine Learning and AI Applications

Uncategorized

Dietary Interventions for Speech Delay and Hyperactivity in a Child with Machine Learning and AI Applications/strong>
Authors:-Sujatha Mudadla

Abstract-This study investigates the role of specific dietary changes in addressing speech delay and hyperactivity symptoms in my son. Recognizing nutrition and maternal health as influential factors in child development, I explored how targeted dietary adjustments might enhance speech clarity, attention, concentration, and behavior. The study also explores maternal influences, including anemia and stress during conception, and their potential impacts on gut health and speech development. Additionally, I examined the effectiveness of repeated oral teaching methods, such as memorizing rhymes and vocabulary, for reinforcing neural pathways. To extend the research, I explore how machine learning (ML), deep learning (DL), computer vision, and generative AI can be applied to monitor, predict, and enhance the intervention’s effectiveness.

DOI: 10.61137/ijsret.vol.10.issue5.318
55

Published by:

Heart Disease Prediction Using Machine Learning Techniques in Python: A Review

Uncategorized

Heart Disease Prediction Using Machine Learning Techniques in Python: A Review/strong>
Authors:-Tanmay Deshmukh, Supriya Kharatmol, Professor Nishant Patil

Abstract-As the global incidence of heart disease escalates daily, there is an urgent imperative to accurately predict and diagnose these conditions efficiently. Heart illness, also referred to as cardiovascular disease, is a broad category of conditions that affect the heart, including congenital abnormalities, vascular problems, and cardiac arrhythmias. In recent decades, it has emerged as one of the world’s top causes of death. Thus, it is imperative to create accurate and trustworthy techniques for early disease detection .Heart illness, also referred to as cardiovascular disease, is a broad category of conditions that affect the heart, including congenital abnormalities, vascular problems, and cardiac arrhythmias

DOI: 10.61137/ijsret.vol.10.issue5.317
55

Published by:

Space Debris Tracking and Prediction Models

Uncategorized

Space Debris Tracking and Prediction Models/strong>
Authors:-Sakshi Khedekar, Jayesh Jadhav, Jiya Mokalkar, Pratik Patil, Professor Manisha Mali

Abstract-In a growing risk for space activities intentionally located or accidentally resulting from the creation of space debris, monitoring and forecasting are indispensable for the protection of both crewed and uncrewed space missions. The paper presents the assessment of eight most widespread space debris tracking and prediction models: TLE based SGP4, ORDEM, MASTER, Debrisat, SDebrisNet, SDTS, CARA, SSN. For every model, a multi-faceted approach with respect to its various characteristics, accuracy, complexity, data requirement, adaptability, reliability, and usability is employed. This appraisal provides the benefits and associated drawbacks of each methodology in tackling the major issues of data, computation and construction of the complete system. The research further considers the progress of tracking devices and existing systems as well as possibilities of their improvement for the realtime challenges. The comparative assessment of the models presented in this paper will help to strategically improve current approaches to space debris control instruments, thus supporting safety and long-term operating trends in outer space. This study has been carried out in order to devise strategies that will fit the growing and dynamic endeavors of exploring space, by tracking debris with the utmost efficiency and precision.

DOI: 10.61137/ijsret.vol.10.issue5.316
55

Published by:
× How can I help you?