Category Archives: Uncategorized

Automated Student Attendance Using Computer Vision
Authors:-Reshi, Merinkanth, Yogeswari, Gayathri, HOD/IT Dr.P.Sachidhanandam

Abstract-The implementation of an Automated Student Attendance System utilizing Computer Vision optimizes attendance management by leveraging facial recognition technology to automate the marking process. This approach minimizes manual intervention while enhancing accuracy and efficiency. A web-based interface facilitates seamless attendance tracking, record maintenance, and real-time monitoring. Furthermore, the system incorporates email notifications to provide timely updates and allows direct downloads of attendance logs for administrative convenience. Security measures are reinforced through the identification and image capture of unauthorized individuals, with automated email alerts dispatched to administrators for enhanced surveillance. By integrating artificial intelligence, this system ensures a robust, reliable, and autonomous attendance tracking solution, significantly improving record-keeping efficiency within educational institutions and organizations.

DOI: 10.61137/ijsret.vol.11.issue2.350

Published by: Kajal Tripathi

Bianchi Type-III Cosmological Model with f(R, T) Gravity Based on Lyra Geometry
Authors:-L. S. Ladke, B.V.Bansole, V. P. Tripade

Abstract- This paper is devoted to the study of Bianchi type-III cosmological model with gravity in the presence of perfect fluid based on Lyra geometry. We formalize the gravity equations based on Lyra geometry. To solve the field equations, obtained by considering Bianchi type-III space-time, we used physical condition that the shear scalar σ2 is proportional to scalar expansion . The behavior of the model has been discussed by studying the physical and kinematical properties of the model.

DOI: 10.61137/ijsret.vol.11.issue2.331

Published by: Kajal Tripathi

AI-Powered SQL Assistant: Transforming Natural Language into Optimized SQL Queries
Authors:-Atharv Deshmukh, Manali Gawade, Ronit Fulari

Abstract-Databases have a steep learning curve, littered with schema design, SQL di- alects, and performance optimizations. Formulating efficient SQL queries is a chal- lenging process to tinker with, and this is one of the reasons many developers and analysts are blocked. Here we present an AI-based SQL Assistant that utilizes cutting-edge AI models to transform natural language requests into fast SQL code. It lives on top of a variety of SQL dialects, including Spark SQL, PostgreSQL, and MySQL, and provides schema suggestions and smart executor queries. It can use a feedback loop with machine learning methods to improve performance after the sys- tem is deployed based on users adapting the system to query patterns. We provide experimental evidence to show that the proposed solution not only improves query performance and execution time but also accuracy so that the database interactions are smooth for non-experts.

DOI: 10.61137/ijsret.vol.11.issue2.349

Published by: Kajal Tripathi

Harnessing Computer Vision for Precision Agriculture: Advancements in Crop Monitoring, Yield Prediction, and Disease Identification
Authors:-Swaraj Kawade, Anurag Kawade, Rohit Kashid, Aditya Jedhe, Rajdeep Jagtap

Abstract-The use of modern technologies such as computer vision and artificial intelligence individually or in tandem are modifying farming methods. These technologies are assisting in agricultural practices as the world population is expected to peak at 9.7 billion by 2050. The objective of this paper is to analyze major advancements within the years 2020 to 2024 regarding the role of computer vision in agriculture, specifically in crop health monitoring, yield prediction, and early detection of plant diseases. Some impressive progress includes: Autonomous weed management systems with 96% accuracy (Praveenraj et al., 2024). Near 99% accuracy in crop yield predictions from machine learning models (Sharma et al., 2023). Deep learning algorithms correctly identifying plant diseases at a rate of 99.35% (Li et al., 2021). Additionally, the development of autonomous vehicles has contributed to safety within agricultural fields, while AI image generation contributes to predicting potential yield imagery, along with real-time field monitoring robotic systems. These developments are positive strides towards sustainable agriculture. On the other hand, these systems always carry limitations like dealing with dynamic field conditions, insufficient amounts of data, and the need for high-end processing units. This paper evaluates the advancement to date, what it means in terms of practical agriculture, and how we can further progress:

DOI: 10.61137/ijsret.vol.11.issue2.348

Published by: Kajal Tripathi

Image Fusion of MRI and CT Scan for Brain Tumor Detection Using VGG-19
Authors:-Professor Kirti Digholkar, Shreyas Depura, Adwait Mali, Vedant Latthe, Rohan Patil

Abstract-For a patient’s prognosis, the careful examination of image bio-analytics, including Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) imaging, is crucial for the patient’s tumor detection. Moreover, interpreting these images manually remains challenging owing to the required expertise and time needed to properly analyze the images. To address this issue, we propose an image improvement model that enhances the accuracy of MRI and CT scans using Wavelet- based fusion and the VGG-19 architecture. Image fusion, or the merging of medical images, synergistically uses and adapts the various modalities’ strengths and weaknesses. In our research, we apply the Wavelet approach to MRI and CT images by splitting them into frequency sub-bands. Structural details are important for the image’s low-frequency LL band. The VGG- 19 network which consists of several convolutional layers and pooling layers is then used to merge the LL bands and form the fused images. Our method undergoes a series of preprocessing, feature extraction, and fusion stages on brain MRI and CT scans. This method saves time for medical practitioners and enables efficient tumor identification through automation, improving the overall quality of patient care.

DOI: 10.61137/ijsret.vol.11.issue2.347

Published by: Kajal Tripathi

Next-Gen Health Solutions
Authors:-Assistant Professor Priti Bharambe, Vikas Mahandule, Shraddha Phulsundar, Priti Aivale, Shivanjali Shinde

Abstract-The integration of Information Technology (IT) in healthcare has transformed patient care, data management, and operational efficiency. This paper explores the role of IT solutions in enhancing healthcare delivery, focusing on electronic health records (EHRs), telemedicine, artificial intelligence (AI), and blockchain technology. IT innovations facilitate real-time data sharing, improve diagnostic accuracy, and enable remote patient monitoring, thereby increasing accessibility and reducing costs. Despite significant advancements, challenges such as data safety, compatibility, and ethical concerns persist. This study examines both the advantages and limitations of IT-driven healthcare, proposing strategies to optimize its implementation. By leveraging technology effectively, healthcare systems can enhance patient results and overall efficiency, clearing the path for a more connected and intelligent healthcare ecosystem.

DOI: 10.61137/ijsret.vol.11.issue2.346

Published by: Kajal Tripathi

A Machine Learning Approach to Heart Disease Prediction: 5-Fold Cross Validation and Hyperparameter Optimization
Authors:-Dr.N.Chandrasekhar

Abstract-The primary objective of this research is to develop an effective predictive model for heart disease using various Machine Learning (ML) algorithms. In this study, four different ML models—Gradient Boosting (GB), Random Forest (RF), LightGBM (LGBM), and AdaBoost (AB)—were implemented and evaluated for their prediction accuracy. To ensure the reliability and generalization of the models, 5-fold cross-validation was applied along with Grid Search Cross Validation (Grid Search CV) for hyperparameter tuning. This technique helped in identifying the optimal parameters for each algorithm, thereby improving their performance. Among all the models, Gradient Boosting achieved the highest accuracy of 95.08%, followed by Random Forest and LightGBM, both with 91.80%, and AdaBoost with 90.16%. These results highlight the effectiveness of ensemble-based ML models, particularly Gradient Boosting, in accurately predicting the risk of heart disease.

DOI: 10.61137/ijsret.vol.11.issue2.345

Published by: Kajal Tripathi

The Impact of Personalization on E-commerce Conversion Rates: An Empirical Analysis of 100 Respondents
Authors:-Hitesh Ramdasani

Abstract- This study investigates the impact of personalization strategies on e-commerce conversion rates, examining the perceptions of 100 online shoppers. Through a simulated survey approach, the research explores the relationship between the level of perceived personalization experienced by consumers and their likelihood of making a purchase. The findings from the simulated data analysis suggest a positive correlation between higher levels of perceived personalization and increased conversion likelihood. These results underscore the importance of implementing effective personalization techniques for e-commerce businesses seeking to enhance customer engagement and drive sales.

DOI: 10.61137/ijsret.vol.11.issue2.344

Published by: Kajal Tripathi

Student Council Election Portal
Authors:-Professor Bharati Bisane, Vaishnavi Rajendra Borse, Resham Sanjay Umale, Sharddha Bhagwan Borate

Abstract- Traditional paper-based voting has been used for many years, but it comes with several challenges, such as security risks, lack of transparency, human errors, and privacy concerns. To solve these issues, we propose a blockchain-based voting system for college elections. Blockchain technology is in high demand because it offers security, transparency, and decentralization. Our goal is to use blockchain to create a secure and tamper-proof election system at the college level. This system will also help developers build and deploy smart contracts, which ensure accuracy and provide quick voting results.
Smart contracts automate the voting process, making vote counting secure and preventing fraud. Blockchain stores all transactions in blocks within a decentralized network, ensuring that no single person can manipulate the results. Overall, th is system will make the voting process faster, more secure, and more reliable. It will also reduce costs since there is no need to print ballots.

DOI: 10.61137/ijsret.vol.11.issue2.343

Published by: Kajal Tripathi

Mall Customer Segmentation System for Retail Analytics and Personalized Marketing
Authors:-Dr.Prabakaran, S.M.Rafi Saddam, T.Narendra, B.Venkateswarlu, T.Venkatesu

Abstract- This paper presents a comprehensive customer seg- mentation system for retail businesses, specifically designed for shopping mall environments. Using advanced clustering tech- niques and RFM (Recency, Frequency, Monetary) analysis, we develop a robust framework for identifying distinct customer segments with similar purchasing behaviors. The system pro- cesses transactional data to create meaningful customer profiles, enabling businesses to implement targeted marketing strategies and improve customer relationship management. Our approach integrates data preprocessing, feature engineering, clustering algorithms, and interactive visualization to provide actionable insights. The implemented dashboard facilitates segment com- parison, geographical distribution analysis, and automated per- sonalized email campaigns tailored to each segment’s prefer- ences. Experimental results demonstrate the effectiveness of this approach in identifying five key customer segments with distinct behavioral patterns. The system’s practical application is validated through its ability to generate segment-specific market- ing recommendations and predict customer preferences, leading to more efficient resource allocation and potentially increased customer engagement. This research contributes to both the theoretical understanding of customer behavior modeling and provides a practical tool for retail analytics in real-world business environments.

DOI: 10.61137/ijsret.vol.11.issue2.342

Published by: Kajal Tripathi