Daily Archives: June 26, 2026

Authors: Bala Sundara Rao Kimmoju, Y.Jagadeesh Kumar, P. Pradeep

Abstract: Climate change has significantly increased the occurrence of extreme weather events such as floods, cyclones, droughts, heatwaves, and heavy rainfall, creating a strong need for accurate and reliable forecasting systems. Traditional climate prediction methods often fail to effectively capture the complex spatial and temporal relationships present in large-scale climate data and generally lack interpretability. This project proposes an Explainable Hybrid Deep Learning Framework for Climate Trend Analysis and Extreme Weather Prediction that integrates Convolutional Neural Networks (CNN) for spatial feature extraction, Long Short-Term Memory (LSTM) networks for temporal sequence learning, and an Attention Mechanism for identifying important climatic features. To enhance transparency and trustworthiness, Explainable Artificial Intelligence (XAI) techniques such as SHAP and Grad-CAM are incorporated into the framework. The system utilizes climate parameters including temperature, humidity, rainfall, wind speed, atmospheric pressure, cloud cover, and satellite imagery collected from multiple sources. Data preprocessing techniques such as normalization, missing value handling, and feature engineering are applied to improve data quality and model performance. The hybrid CNN-LSTM architecture effectively learns spatiotemporal climate patterns, enabling accurate climate trend analysis and extreme weather forecasting. Experimental results demonstrate improved prediction accuracy, reduced false alarm rates, and better interpretability compared to traditional forecasting approaches. The proposed framework supports real-time climate monitoring and provides reliable, transparent, and efficient forecasting solutions for disaster management, agriculture, environmental monitoring, and public safety applications.

DOI: http://doi.org/10.61137/ijsret.vol.12.issue3.194

Published by: vikaspatanker

Authors: Simhachalam Patnana, S.Sudeer Kumar, Y. Jagadesh Kumar

Abstract: Automatic image recognition has become a fundamental component of modern intelligent systems, finding applications in areas such as food recognition, healthcare imaging, smart surveillance, object detection, and visual analytics. However, traditional image classification techniques often face challenges due to image noise, class imbalance, varying lighting conditions, complex backgrounds, and diverse visual patterns, which reduce classification accuracy and prediction reliability. To address these challenges, this project proposes a Smart Vision Fusion Architecture for Multi-Class Image Recognition (SVFA-MCIR), an intelligent hybrid framework that combines deep learning and machine learning techniques for efficient multi-class image classification.The proposed framework incorporates image preprocessing, enhancement, augmentation, and feature optimization techniques to improve dataset quality and model performance. Existing image recognition models such as CNN, EfficientNet + XGBoost, and DenseNet + XGBoost are initially evaluated to analyze their classification capabilities. To further enhance recognition accuracy and classification stability, the proposed system integrates ResNet50 and XGBoost into a unified hybrid architecture. ResNet50 is utilized to extract high-level visual features and complex image representations, while XGBoost performs optimized multi-class classification using the extracted deep feature vectors.Experimental results demonstrate that the proposed SVFA-MCIR framework achieves superior performance in terms of recognition accuracy, prediction robustness, feature learning capability, and computational efficiency when compared with existing approaches. The framework provides a scalable, adaptive, and intelligent solution for modern image recognition applications and contributes to the advancement of smart vision systems through accurate and reliable multi-class image classification.

DOI: http://doi.org/10.61137/ijsret.vol.12.issue3.195

Published by: vikaspatanker

Authors: Mrs. Vidyashree B.P Assistant Professor, Manjunath A J, Pradeep Nagavath, Shreyank S D, Poorna Chandra Thejaswi M D

Abstract: Vehicle theft remains a significant concern worldwide, especially in urban areas where vehicle density is high and traditional security systems are often insufficient. This project presents a cost-effective and intelligent Vehicle Theft Protection System that enhances vehicle security through biometric authentication and real-time user intervention using GSM communication. The core of the system is built around the Arduino UNO microcontroller, interfaced with a fingerprint sensor module (R305S), a GSM module (SIM800L), and a relay module to control the ignition system. Authorized users register their fingerprints in the system memory. Upon an unauthorized access attempt, the system sends an SMS alert to the vehicle owner, who can remotely allow or deny engine start. The proposed system provides a high-speed, reliable, and cost-effective solution for automotive embedded security applications.

Published by: vikaspatanker

Authors: Pilla Rushitha, Puppala Pradeep, Yerrapatruni Jagadeesh Kumar

Abstract: Natural disasters such as floods, earthquakes, cyclones, droughts, landslides, and wildfires continue to pose significant threats to human life, infrastructure, and environmental sustainability. The growing complexity of climate patterns and environmental changes has increased the need for intelligent disaster prediction systems capable of providing accurate and timely forecasts. This project presents a Neural-XGBoost Hybrid Framework for Disaster Prediction and Management that integrates deep learning-based feature extraction with the robust classification capability of Extreme Gradient Boosting (XGBoost). The proposed approach utilizes disaster-related environmental and meteorological data, including rainfall, temperature, humidity, wind speed, and atmospheric conditions, to identify potential disaster events. Data preprocessing techniques such as cleaning, normalization, and feature selection are employed to enhance data quality and model performance. The neural network component automatically learns complex patterns and hidden relationships within the dataset, while XGBoost performs efficient multi-class disaster classification. Experimental evaluation demonstrates that the hybrid framework achieves superior prediction accuracy, improved generalization capability, and reduced overfitting when compared with conventional machine learning approaches. The system supports disaster preparedness, risk assessment, resource planning, and early warning mechanisms, enabling authorities to make informed decisions and minimize disaster-related losses. The proposed framework offers a scalable, reliable, and data-driven solution for modern disaster management applications.

DOI: http://doi.org/10.61137/ijsret.vol.12.issue3.196

Published by: vikaspatanker

Authors: Balla Revathi, Dhavala Shilpa, Yerrapatruni Jagadeesh Kumar

Abstract: Accurate product pricing has become a critical requirement for modern e-commerce platforms due to rapidly changing market conditions, customer preferences, competitor strategies, and fluctuating product demand. Traditional pricing methods often rely on static rules and historical analysis, making them ineffective in responding to real-time market dynamics. To address these challenges, this project proposes an intelligent framework called Adaptive Commerce Intelligence Framework for Real-Time Product Value Forecasting Using Hybrid Predictive Learning Models, which integrates machine learning techniques with business intelligence analytics to support intelligent pricing decisions and real-time product value forecasting.The proposed system collects and analyzes various pricing-related parameters, including product base cost, competitor pricing, sales volume, stock availability, customer ratings, reviews, and market trends. Individual machine learning algorithms such as Linear Regression, Random Forest, Support Vector Machine (SVM), and XGBoost are initially trained and evaluated independently to assess their forecasting capabilities. These models are then combined into a Hybrid Predictive Learning Model that leverages the strengths of each algorithm to improve prediction accuracy, forecasting stability, and pricing adaptability.Random Forest and XGBoost effectively identify complex market patterns and pricing trends, while SVM captures non-linear relationships among pricing factors. Linear Regression contributes to understanding pricing dependencies and improving model consistency. The framework also incorporates real-time analytics, competitor monitoring, historical prediction tracking, interactive dashboards, and MySQL-based data management to enhance business intelligence and decision-making capabilities.Experimental analysis demonstrates that the proposed hybrid framework provides more accurate and reliable pricing forecasts compared to standalone machine learning approaches. By integrating predictive learning with adaptive commerce analytics, the system enables dynamic pricing optimization, improves market responsiveness, supports revenue growth, and enhances competitiveness in modern digital commerce environments.

DOI: http://doi.org/10.61137/ijsret.vol.12.issue3.197

Published by: vikaspatanker

Authors: Tosif Raza Mansoori

Abstract: In the modern business environment, organizations generate large volumes of transactional data that contain valuable information regarding profitability, operational efficiency, product performance, and market behavior. However, extracting meaningful insights from raw datasets remains a significant challenge. Business Intelligence (BI) and Data Analytics techniques provide effective solutions by transforming data into actionable information that supports strategic decision-making. This research presents a comprehensive Product Line Profitability and Margin Performance Analysis Dashboard developed for Nassau Candy Distributor. The primary objective of this project is to evaluate the profitability of product lines, analyze revenue distribution, identify high-performing products, assess regional performance, and provide business recommendations through data visualization. The dashboard was developed using Python and Streamlit, while Pandas, NumPy, Matplotlib, and Seaborn were utilized for data preprocessing, statistical analysis, and visualization. Several analytical techniques including Key Performance Indicator (KPI) evaluation, profitability analysis, division-wise performance assessment, revenue analysis, and Pareto Analysis were implemented to uncover business insights. The developed dashboard enables stakeholders to monitor revenue, profit, cost, margin percentage, and product performance through interactive visualizations. The findings demonstrate that a limited number of products contribute significantly to overall profitability, confirming the applicability of the Pareto Principle in business analytics. The proposed solution provides a scalable and user-friendly analytical framework that assists management in making informed business decisions related to pricing, inventory planning, product portfolio optimization, and strategic growth initiatives.

Published by: vikaspatanker

Authors: Sushma P S Assistant Professor, Chiranthan M Y, Jayanth K M, D P Rajashekar, Suresh B

Abstract: Power consumption and computational speed are important factors in modern digital systems. This project presents a 32-Bit Vedic ALU with Low Power Mode using System Verilog. The design employs the Urdhva Tiryakbhyam algorithm for fast multiplication and incorporates operand isolation and clock gating techniques to reduce power consumption. The ALU performs arithmetic, logical, and shift operations efficiently while maintaining high performance. The proposed system provides a high-speed, reliable, and power-efficient solution for embedded systems and processor applications.

Published by: vikaspatanker