Current Issue
Browse the latest research articles published by the International Journal for Interdisciplinary Sciences and Engineering Applications.
Research Articles
This project presents a scalable solution for multi-class sentiment analysis of Twitter data. It classifies tweets into Extremely Negative, Negative, Neutral, Positive, and Extremely Positive categories using FastAPI for backend API development and a TensorFlow deep learning model for accurate real-time sentiment prediction.
Ground moving target indication (GMTI) in synthetic aperture radar (SAR) remains challenging when weak or slowly moving targets are embedded in stationary clutter. This paper presents a unified analysis of adaptive and improved simplified fractional Fourier transform (SFFrFT) approaches optimized using Harris Hawks Optimization and Hybrid Black Widow–Elephant Herding Optimization for enhanced target detection and localization.
Automated fraud detection in transaction data uses machine learning pipelines to identify suspicious patterns, improve prediction accuracy, and support faster financial risk analysis.
AI powered interview preparation with code execution supports students and job seekers by combining guided practice, coding questions, live code execution, and performance-focused preparation features.
The retail industry faces increasing challenges in matching supply with demand due to evolving consumer behaviors, market volatility, and supply chain disruptions.
The Sales Performance Dashboard is an advanced business intelligence solution designed to provide comprehensive analytics and visualization for automobile sales data. In today's competitive automotive market, organizations require real-time insights into their sales performance across multiple dimensions including dealer regions, vehicle body styles, temporal trends, and individual model performance.
Customer attrition prediction is a critical challenge in the telecommunications industry due to its direct impact on customer retention and organizational revenue. Traditional churn analysis approaches primarily rely on descriptive statistics and reactive strategies, which are insufficient for identifying high-risk customers at an early stage. This paper proposes an explainable retention intelligence framework for telecom customer attrition prediction using ensemble learning techniques and interactive business intelligence visualization.
The aim of this project, titled “End-to-End Predictive Analytics Pipeline for Customer Behaviour Using Python and Neural Networks”, is to develop a system that predicts customer behaviour and satisfaction using advanced machine learning techniques. The system performs preprocessing, feature engineering, and trains a neural network model to accurately predict customer satisfaction levels and behavioural trends.
In today's competitive business environment, understanding customer behavior and predicting customer value are essential for effective marketing and business growth. This project focuses on Customer Segmentation and Customer Lifetime Value prediction using data modelling techniques. The system analyzes customer transactional data using Recency, Frequency, and Monetary analysis, applies K-Means clustering, and includes an interactive dashboard for efficient visualization.
The Random Access Channel procedure in Fourth Generation Long Term Evolution technology plays a critical role in establishing initial communication between User Equipment and the evolved NodeB. This paper presents an overview of the RACH procedure, highlighting its significance in uplink synchronization, connection establishment, and mobility management.
In the contemporary digital economy, the spread of product options has posed a big problem upon consumers, which has been referred to as information overload. This study introduces a powerful personalized e-commerce recommendation engine that aims at reducing decision fatigue and improving user experience using data science techniques.
Rapid advancement of digital technologies has significantly transformed traditional grocery shopping methods into modern, efficient online platforms. This paper presents the design and development of a web-based online grocery shopping system that enables customers to browse products, manage shopping carts, and place orders through an intuitive interface.
As internet-based services and online communication grow rapidly, computer networks are becoming more susceptible to cybercrime, including distributed denial-of-service attacks, unauthorized access attempts, and traffic surges. This study introduces a real-time network security monitoring and auto-mitigating firewall system using Python and IPTables.
Stock market prediction remains one of the most challenging problems in financial analytics due to the highly volatile, nonlinear, and dynamic nature of stock price movements. This paper presents a machine learning-based approach for stock market prediction and trend analysis with an interactive Power BI dashboard.
This study suggests an Artificial Intelligence-powered Augmented Reality Smart Shopping Assistant that offers a real-time virtual try-on experience with computer vision and web-based technologies. The proposed system combines body landmark detection, automatic background removal, and dynamic image overlay features to enable users to view clothing items on their bodies using a web camera interface.
Credit risk assessment is important in finance because it affects how banks decide on loans and manage risks. This project builds a system using machine learning and visualization tools to predict loan default risk. Python is used for data cleaning and model building, while Power BI dashboards present risk charts, borrower groups, default trends, and key metrics for decision support.
This project presents a scalable solution for multi-class sentiment analysis of Twitter data. It classifies tweets into Extremely Negative, Negative, Neutral, Positive, and Extremely Positive categories. The system uses FastAPI for backend development and a TensorFlow deep learning model for accurate real-time sentiment prediction.
This work presents the design and implementation of a scalable web-based attendance management system with integrated analytical capabilities to improve accuracy and efficiency. The system enables instructors to record attendance in real time, calculates attendance percentages, identifies absenteeism trends, and generates visual reports for academic decision support.
This project presents the design and development of a full-stack e-commerce web application named Nexacart, aimed at delivering a robust, user-friendly, and secure platform for online transactions. The system integrates user authentication, product catalog management, shopping cart, wishlist, order processing, and administrative controls.
The Smart Crime Analytics and High-Risk Zone Forecasting System is an intelligent data analytics platform developed using Streamlit, Pandas, and Folium to analyze historical crime datasets and forecast high-risk areas through geo-visualization and temporal trend analysis. The platform supports crime record uploads, filtering, heatmaps, marker clusters, and monthly crime trend insights.
Predicting ICU readmissions is critical for improving healthcare outcomes and reducing costs. This study uses electronic health records to predict ICU readmissions through preprocessing methods such as age mapping, normalization, and binary encoding. The approach prepares structured data for machine learning models to support better patient management.
This study presents a data-driven machine learning approach for ransomware detection using executable file records. The proposed system extracts Portable Executable header features and applies a Random Forest classifier to distinguish between legitimate and malicious files. SMOTE-TOMEK resampling and LIME explainability are integrated to improve reliability and transparency.
This paper presents an AI-powered content-based book recommendation framework that uses Natural Language Processing techniques to generate meaningful recommendations from textual book descriptions. The system applies TF-IDF for feature extraction, cosine similarity for measuring semantic relationships, and an interactive Streamlit application for real-time recommendation visualization.
This work presents an advanced control strategy for improving the speed control performance of a three-phase squirrel cage induction motor using a hybrid Fuzzy-PID controller integrated with vector control. Simulation results show faster response, reduced overshoot, improved stability, and strong robustness against load disturbances and parameter variations.
This paper proposes the design, creation, and testing of an Online Court Booking and Management System for legal firms. The platform uses the MERN stack and integrates real-time WebRTC video consultations, structured case management, secure document handling, role-based access control, and automated email and SMS notifications.
This project proposes an Intrusion Detection and Prevention System using Python and Snort for efficient detection and mitigation of DNS flood and DDoS attacks. The system combines Snort-based intrusion detection with Python-based traffic analysis and automation to provide real-time detection, alert correlation, malicious IP blocking, rate limiting, and firewall rule updates.
This project presents a real-time anomaly detection system for IoT sensor data using LSTM Autoencoder neural networks. The system processes multivariate sensor streams including temperature, humidity, air quality, light intensity, and loudness, then identifies deviations from normal patterns using reconstruction error thresholding through an interactive Streamlit web application.
This paper presents a Workforce Harmony Predictor System, a machine learning-based solution designed to identify employees who are likely to leave an organization. The system uses features such as satisfaction level, evaluation score, workload, tenure, and salary, with a Random Forest model and Streamlit web application for single and batch predictions.
This project presents a Customer Churn Prediction and Retention Dashboard developed using Streamlit. The tool enables users to upload customer datasets, preprocess data, apply multiple machine learning models, predict churn, compute churn probabilities, recommend retention actions, and visualize churn distribution, feature importance, and retention strategies.
This project presents a web-based AI-powered travel itinerary planner built as a Streamlit application. It uses LangChain and a Large Language Model accessed through the Groq API to generate personalized travel plans based on destination, trip duration, interests, and travel style, with dynamic pricing support.
ResilienceNet Forecaster is an AI-driven disaster risk management platform that integrates machine learning, explainable AI, geospatial hazard mapping, and intelligent resource allocation. The system predicts disaster risk levels, estimates potential impact, visualizes hazard maps, and converts risk scores into emergency response plans.
This paper proposes an automated stress detection system built upon BERT, a deep learning model that analyzes user-generated text to classify stress and non-stress conditions. The system uses preprocessing, tokenization, fine-tuned training, and a Streamlit interface for real-time analysis and probability visualization.
Athlon Predict Pro is a sports analytics platform for cricket that uses machine learning to predict player performance and optimize team strategies. It processes historical ODI match data and uses models such as XGBoost, Random Forest, and SVM for performance forecasting, team selection, injury risk prediction, and strategy support.
This project develops a Smart Coding Interview Preparation Portal with live code execution to overcome the limitations of traditional interview preparation platforms. The system integrates real-time code execution using the JDoodle API, AI-driven personalized learning, support for multiple programming languages, and an analytics dashboard to track progress, accuracy, and interview readiness.
This project proposes an AI-based Osteoporosis Detection System that analyzes clinical patient data and DXA images to classify bone health into Normal, Osteopenia, and Osteoporosis categories. The system uses machine learning models on clinical attributes and a convolutional neural network for DXA scan image analysis, while also providing personalized lifestyle recommendations.
This project presents a Smart Crop Recommendation and Yield Prediction System using machine learning to support data-driven agricultural decisions. The system analyzes soil parameters such as nitrogen, phosphorus, potassium, and pH, along with environmental factors like temperature, humidity, and rainfall, to provide crop recommendations through an interactive Streamlit web application.
This project presents a Smart Heart Disease Prediction System, an AI-powered web application that predicts the likelihood of heart disease using clinical parameters such as age, sex, chest pain type, blood pressure, cholesterol, ECG results, maximum heart rate, exercise-induced angina, old peak, and ST slope. It provides risk classification, probability scores, personalized health insights, and preventive recommendations.
This project introduces Support Bot AI, an intelligent conversational system designed to understand, analyze, and respond to customer queries in an emotionally aware manner. The chatbot integrates natural language processing for intent recognition and sentiment analysis to detect user emotions and provide empathetic, context-aware responses through a Streamlit-based interface.
This study proposes an explainable machine learning framework for predicting stroke occurrence based on clinical, demographic, and lifestyle factors. The system uses attributes such as age, hypertension status, heart disease history, average glucose level, smoking behavior, and BMI, with SMOTE balancing, multiple machine learning models, and SHAP analysis for interpretable prediction.
This project presents an interactive Streamlit-based web application for advanced demand forecasting in retail supply chain management using a pre-trained XGBoost model. The system predicts daily item-level sales, provides monthly trends, store and item breakdowns, actual versus predicted comparisons, RMSE metrics, and downloadable forecast results to support inventory optimization.
This paper presents an AI-based brain stroke classification system using Vision Transformer models to classify MRI scans into Normal and Haemorrhagic categories. The system compares CNN, ResNet-18, and ViT architectures, applies data augmentation, and deploys the prediction workflow through a Streamlit web application for real-time clinical decision support.
This paper presents a Workforce Harmony Predictor System, a machine learning-based solution designed to identify employees who are likely to leave an organization. The system uses satisfaction level, evaluation score, workload, tenure, and salary with a Random Forest model and Streamlit web application for single and batch predictions.
The Sales Performance Dashboard is an advanced business intelligence solution designed to provide comprehensive analytics and visualization for automobile sales data. The system uses Power BI and predictive forecasting to analyze dealer regions, vehicle body styles, temporal trends, individual model performance, and future sales patterns.
Issue Files
Archives