← Back to Archives
VOLUME 13, ISSUE 11, NOVEMBER 2025
Data-Driven Retail Sales Forecasting Through Machine Learning Approaches
Pratham Mehta, Aditya Balaji, Dr. Golda Dilip
PERSONALIZED LEARNING PATH RECOMMENDATION SYSTEM
Prithivi Raj.S, Goutham Krishna, Abhinav Manoj, Bharath G, Anand V D, Neelam Sanjeev Kumar
Audio Deepfake Detection Using MFCC-SVM, CQCC-GMM, and Spectrogram-CNN: A Comparative Study
Abhiram Sri Paravastu, B. Harshat, H. Keerthi Lakshmi, Abhishek Harish, Vivek S Nair, Neelam Sanjeev Kumar
CosmoScan – A Galaxy Type Identifier Using Computer Vision
Jaishree Baskaran, Kirthika Hariram, Charulatha.R.T
AI-Driven Communication Analysis for Biodiversity Conservation: A Fruit Bat Prototype
Ankit Sharma, Owais Kadri, Shourya Kapoor, Surya Amrit, Yogesh Choudhary, Neelam Sanjeev Kumar
Intelligent Prioritisation of Tomato Leaf Disease Diagnosis Using Symptom Hierarchies and Computer Vision
Himanshu, Nishant Kumar, Ishaan Chandola, Neelam Sanjeev Kumar
AI Discharge & Readmission Prevention System
Ragul M, Derick Dilip, Ms. D. Kalpana
FINANCIAL FRAUD DETECTION
Sai Chandana Y, Rama Devi DP, Neethu Jimmy Joy, Neelam Sanjeev Kumar
Predictive Mental Health Risk Assessment in Undergraduate Students Using Personalized Adaptive Machine Learning
Saai Jaswant, Ahamed Saif, Aishwarya Balaji, Sree Mouthika, Neelam Sanjeev Kumar
IoT-Based Smart Waste Management System for Smart Cities: A Synthesized Framework Addressing Security and Interoperability Challenges
Prof. Rana Afreen Sheikh, Rushikesh R. Gunjarkar, Snehal R. Nagrale
ATS Resume Classification Project Analysis
Vibhu Sahu, Kunal Meshram, Anshita Tripathi
Comparative Predictive Modeling of Dry Eye Disease: An Integrated Approach Using Decision Tree and Random Forest Techniques
Mohammed Ihsan N, Sharan R, Dr. Paavai Anand
Forest Fire Severity Prediction using Random Forest and Neural Network Stacking with SMOTE
Mani Shankar B, Jai Vignesh K, Arjun Ashkar N C, Dr G. Paavai Anand
Machine Learning Approaches to Passenger Survival Prediction: A Titanic Dataset Analysis
ESWARAMUTHU M, M KIRITHIKA, ABHISHEK R, LAVANYA S, MOHANA KRISHNAN B, VAISHNAVI S, Dr. M. ULAGAMMAI
A Hybrid Machine Learning Approach for Apple (AAPL) Stock Price Prediction Using Ensemble Methods and Anomaly Detection
Gautam R, Shreya R, Dr G. Paavai Anand
Comparative Analysis of User-Based and Item-Based Collaborative Filtering Using the MovieLens Dataset
M. V. Karthikeya, Vishnu Vardhan, D.Nanda Kishore, Subhrajit Panda, Sai Tejas, A. Narendrasai B Sharath Reddy, Dr. M. ULAGAMMAI
Song Skip Prediction Using XGBoost: A Machine Learning Approach for Music Recommendation Systems
Surya Sivakumar, Koduri Pranav, Tharun kumaar SD, Dr G.Paavai Anand
Sentiment Analysis on Customer Reviews Using Fine-Tuned DistilBERT Transformer Model
Caroline Vineeta S L, Vedika Singh, Asritha D, G. Paavai Anand
INNOVATIVE MPPT TRACKING TECHNIQUES FOR INTEGRATED WIND AND PHOTOVOLTAIC ENERGY SYSTEMS
M. Divakar, U. Lakshmi
Password Strength Prediction using Regression-Based Machine Learning Models with Entropy and Caesar Cipher–Driven Synthetic Dataset
Dr G. Paavai Anand, Roshni Y
A Machine Learning Approach for E-Commerce Counterfeit Product Detection Using Transactional and Behavioral Data
Nilesh J, Ashwin C, Anoop Mahesh, Dr G. Paavai Anand
Industrial pollution Monitoring System
E.Sajjan, M.Shivaleela, T.Priyanka, L.Praveen, S.Sanjay
Automatic Vehicle Speed Regulation System
E.Sajjan, A.Shruthi, Ch.Sai Teja, G.Aakash, J.SaiKiran
Toxic Comment Classification Using Ensemble Machine Learning Techniques
AATHITYA.A, KRISHITH TP, SARVESH S, KEVIN BENJAMIN SAMUEL, PRANAV M, VIGNESH D, Dr. M. ULAGAMMAI
Optimized Ensemble Learning Integrated with Anomaly Detection for Road Accident Severity Prediction
Rohith M, Sharan S, Suryaprakasam B, Dr. G. Paavai Anand
INSURANCE CLAIM STATUS PREDICTION
Venkatapathy R, Senthilnathan M, Rohith G, Dr. G Paavai Anand
A Bibliographic Survey on Solar Policies and Regulations for the state of Himachal Pradesh
Jasmine Kaur
OPTIMUM CHARGING METHOD FOR E-VEHICLES
Ananya C, Dr. E. Latha Mercy
Intelligent Switching System for Seamless Solar Inverter Transition Between Grid-Tied and Off-Grid Modes
Rajeswari Ramachandran, Sureshkumar Raghavan
The Use of Interactive Whiteboards in Education: A Quantitative Analysis of Teachers’ Perceptions in Greece
Dr. Konstantinos N. Domouchtsis, Dr. Apostolos Ch. Klonis
AI-Driven Plant Disease Prediction Using Deep Learning and Web-Based Deployment
Sujana Mallavaram, Pranesh S, Chandrika Banerjee, Shalini. M, Dr. Ulagammai. M
Predicting Road Accident Risk
Sushanth. A, Moses. C M, Ashmith. S, Febi Andrew. R, Pranav Sakthi. S, Keshiv Raajh. SK, Joel Sam. S R, M. Ulagammai
Recognition of Handwritten Digit using Convolutional Neural Network in Python
Chandhan Sayee.R, Mohul Raj.T, Merlin.R. S, Sanjith Kumar.K, Nishanth.G. S, Monesh.V, M. Ulagammai
SMART GARBAGE MONITORING SYSTEM USING IoT
B. Sandeep Kumar, P. Vyshnavi, Sana, Y. Rahul Kumar, N. Pranay
IoT Based Surveillance Robot
D Jagan, M Archana, D Varsha, B Rahul, R Pavan Kumar
Microcontroller Based Automatic Engine Locking System for Drunken Drivers
B. Sandeep kumar, P. Udayasri, P. Shivashankar, E. Saikumar, G. Vishal
SMART FLOOR CLEANING ROBOT BY IoT
B. Hanumanthu, B. Prasanna, A. Ramya sree, K. Achala, A. Raghavendra
Vehicle-to-Vehicle (V2V) Communication: Design, Components, Benefits, Challenges, and Implementation Using Embedded Systems
G. Jagan, Afreen, D. Navya, D. Venu, M. Manoj
Enhancing Railway Accident Prevention Using Deep Learning, Machine Learning, And GPS Tracking: A Historical And Knowledge-Based Analysis
VIKAS CHANDRA GIRI, PARINEETA JHA
The Use of Educational Robotics in Greek Education: Trends, Challenges, and Future Prospects
Dr. Konstantinos N. Domouchtsis, Dr. Apostolos Ch. Klonis
Development of an Laser-Based Border Security System
K. Chiranjeevi, J Srujana, B Shailaja, M Mohanasrija, S Sairam
A Comparative Analysis of DTW Variants for Kinect-Based Motor Skill Assessment in Children with Developmental Coordination Disorder (DCD)
Jackie Ting Tiew Wei, Ting Huong Yong, Tiong Meng Chung
Physico-Chemical Characterization of Soil in the Vicinity of Banki Dam and Ghaghi Waterfall, Surguja District, Chhattisgarh, India
Mr. Govind Prasad, Dr. M.K. Maurya, Mr. Pramod Tigga, Miss Deepa Singh, Miss Khileshwari Rajwade
MQTT BASED COAL MINE MONITORING SYSTEM USING – IOT
K.Amarender, V Srikanth, P Pavitra, B Supriya, M Navya
Integrating Flow Theory in Designing Educational Virtual Worlds in the Metaverse
Dimitrios Magetos, Prof. Sarandis Mitropoulos, Prof. Christos Douligeris
Anti-sleep Safety System for the Vehicle Drivers
Tapan Kumar Mohanta, Jaykrushna Mohanta, Dhriti Rudrapal
IDENTIFICATION AND PREVENTION OF ACCIDENTS USING SMART HELMET AND GPS SYSTEM
P Thirupathi, A Aishwarya, J Vaishnavi, B Navya, P Sheshu Kumar
AI-Based Fraud Detection Using Deep Learning on Transaction Data
Abdul Hasham, Mubashir Ali Ahmed
Abstract
Data-Driven Retail Sales Forecasting Through Machine Learning Approaches
Pratham Mehta, Aditya Balaji, Dr. Golda Dilip
DOI: 10.17148/IJIREEICE.2025.131101
Abstract: This project investigates the efficacy of advanced Machine Learning models, including ensemble methods and deep learning architectures, in enhancing sales forecasting accuracy. By comparing their performance against classical time-series models across various datasets, we aim to demonstrate their superior ability to capture relationships and external influences, providing businesses with more reliable predictive tools for strategic planning and operational optimization.
Keywords: Machine Learning, Deep Learning, Sales Forecasting, Time-Series Models, Predictive Accuracy
Keywords: Machine Learning, Deep Learning, Sales Forecasting, Time-Series Models, Predictive Accuracy
Abstract
PERSONALIZED LEARNING PATH RECOMMENDATION SYSTEM
Prithivi Raj.S, Goutham Krishna, Abhinav Manoj, Bharath G, Anand V D, Neelam Sanjeev Kumar
DOI: 10.17148/IJIREEICE.2025.131102
Abstract: Personalized education has become a cornerstone of modern learning systems, enabling students to receive tailored recommendations based on their unique abilities, interests, and cognitive profiles. Traditional academic evaluation systems are rigid and often fail to capture the diversity of student skill sets. Consequently, learners are guided toward generalized career paths that may not align with their strengths or interests. This research proposes a Personalized Learning Path Recommendation System that leverages the power of machine learning (ML) to identify the most suitable learning domains for individual students. The system analyzes multiple attributes including Programming Score, Math Score, Logic Score, Creativity, Problem Solving Ability, Communication Skills, Interest in Technology, and Time Management. Using these parameters, four ML models — Random Forest, Decision Tree, K-Nearest Neighbor (KNN), and Support Vector Machine (SVM) — were trained and evaluated. The Random Forest classifier achieved the highest accuracy of 90%, outperforming other models in stability and prediction consistency. The model provides insights through visual analysis of accuracy comparison, confusion matrices, and feature importance distributions. Additionally, the system accepts live user input to recommend a learning path in domains like AI & ML, Data Science, Web Development, Cyber Security, and UI/UX Design. This approach demonstrates the potential of ML to revolutionize academic counseling by providing data-driven, objective, and personalized recommendations, bridging the gap between aptitude and career direction.
Abstract
Audio Deepfake Detection Using MFCC-SVM, CQCC-GMM, and Spectrogram-CNN: A Comparative Study
Abhiram Sri Paravastu, B. Harshat, H. Keerthi Lakshmi, Abhishek Harish, Vivek S Nair, Neelam Sanjeev Kumar
DOI: 10.17148/IJIREEICE.2025.131103
Abstract: The dawn of sophisticated voice synthesis methods gave rise to audio deepfakes, jeopardizing security and trust in digital communications. In this work, three complementary measures are employed for deepfake detection. First, Mel-Frequency Cepstral Coefficients (MFCCs) are extracted and subjected to Support Vector Machine (SVM) classification. Second, Constant-Q Cepstral Coefficients (CQCCs) are analysed by Gaussian Mixture Models (GMMs), operating with log-likelihood ratio decisions—a long-standing baseline in spoofing countermeasures. Third, Mel- spectrogram representations stand as inputs to a Convolutional Neural Network (CNN) for end-to-end learning. The experiments on the In-the-Wild dataset indicate that CNN enjoys state-of-the-art accuracy, while SVM, on the other hand, is able to provide computational efficiency, and CQCC-GMM holds onto a bit more robustness inherited from traditional anti-spoofing. Therefore, these results present the trade-off between the classical machine learning and the deep learning paradigms, provide insightful considerations when developing reliable deepfake detection methods in real-world conditions.
Keywords: Audio Deepfake Detection, Voice Spoofing, Support Vector Machine (SVM), Gaussian Mixture Model (GMM), Convolutional Neural Network (CNN), MFCC, CQCC, Spectrograms
Keywords: Audio Deepfake Detection, Voice Spoofing, Support Vector Machine (SVM), Gaussian Mixture Model (GMM), Convolutional Neural Network (CNN), MFCC, CQCC, Spectrograms
Abstract
CosmoScan – A Galaxy Type Identifier Using Computer Vision
Jaishree Baskaran, Kirthika Hariram, Charulatha.R.T
DOI: 10.17148/IJIREEICE.2025.131104
Abstract: This research presents CosmoScan, a computer vision-based system designed to identify and classify galaxies into their respective morphological types using real telescope images. The model leverages Convolutional Neural Networks (CNNs) alongside traditional image processing techniques such as HOG and ORB filters to extract visual features from galaxy images. By training on the Galaxy10 dataset, CosmoScan achieves approximately 91% classification accuracy, demonstrating its efficiency in automating the galaxy morphology classification process. The project bridges the gap between classical computer vision and modern deep learning, offering a scalable solution for astronomical image analysis and research.
Keywords: Computer Vision, Deep Learning, Galaxy Classification, CNN, Astronomy, Morphology.
Keywords: Computer Vision, Deep Learning, Galaxy Classification, CNN, Astronomy, Morphology.
Abstract
AI-Driven Communication Analysis for Biodiversity Conservation: A Fruit Bat Prototype
Ankit Sharma, Owais Kadri, Shourya Kapoor, Surya Amrit, Yogesh Choudhary, Neelam Sanjeev Kumar
DOI: 10.17148/IJIREEICE.2025.131105
Abstract: This research aims to explore how Artificial Intelligence (AI) could help us understand a lot better and interpret all the unique vocalizations/communications of various different kinds of species, with the goal of supporting Biodiversity Conservation (BD). The project will develop a prototype system using vocalization data from Egyptian fruit bats. By showing that AI can successfully decode communication in one species, this work lays the groundwork for expanding such models to more complex, multi-species ecosystems in the future.
The system uses a publicly available dataset of Egyptian fruit bat calls, which includes detailed annotations such as who made the call, the context, the intended recipient, and the outcome of the interaction. To prepare the data, the audio was segmented, cleaned of background noise, and converted into sound features like MFCCs and mel-spectrograms. We then tested several deep learning models like CNNs, LSTMs, and Transformers on four tasks: 1) Identifying the caller 2) Classifying the context 3) Recognizing the recipient. and 4) Predicting the interaction’s outcome. Model performance was measured using balanced accuracy, precision, recall, and F1-score, with the results being tested for statistical and numerical significance for p < 0.05.
Inside the fruit bat communication context prediction, our model achieved 63% accuracy (In “Isolation” context our model achieved 100%accuracy).
Keywords: Animal Communication (AC), Biodiversity Conservation (BD), Deep Learning (DL), Egyptian Fruit Bat (EFB), AI for Ecology (AFE), Species Communication Translation (SCE), Wildlife Monitoring (WM)
The system uses a publicly available dataset of Egyptian fruit bat calls, which includes detailed annotations such as who made the call, the context, the intended recipient, and the outcome of the interaction. To prepare the data, the audio was segmented, cleaned of background noise, and converted into sound features like MFCCs and mel-spectrograms. We then tested several deep learning models like CNNs, LSTMs, and Transformers on four tasks: 1) Identifying the caller 2) Classifying the context 3) Recognizing the recipient. and 4) Predicting the interaction’s outcome. Model performance was measured using balanced accuracy, precision, recall, and F1-score, with the results being tested for statistical and numerical significance for p < 0.05.
Inside the fruit bat communication context prediction, our model achieved 63% accuracy (In “Isolation” context our model achieved 100%accuracy).
Keywords: Animal Communication (AC), Biodiversity Conservation (BD), Deep Learning (DL), Egyptian Fruit Bat (EFB), AI for Ecology (AFE), Species Communication Translation (SCE), Wildlife Monitoring (WM)
Abstract
Intelligent Prioritisation of Tomato Leaf Disease Diagnosis Using Symptom Hierarchies and Computer Vision
Himanshu, Nishant Kumar, Ishaan Chandola, Neelam Sanjeev Kumar
DOI: 10.17148/IJIREEICE.2025.131106
Abstract: Tomato Leaf Diseases (TLD) pose a significant threat to crop productivity and fruit quality, as they can spread rapidly if not detected and treated at an early stage. Manual inspection of leaves is time-consuming and prone to human error, particularly because different diseases often exhibit similar visual symptoms. With the advancement of Computer Vision (CV), automatic detection of TLD has become possible; however, most existing methods rely solely on image-based classification and fail to consider the relationships among symptoms that agricultural experts typically use for accurate diagnosis.
This study introduces a Multi-Modal Diagnosis (MMD) system that integrates CV techniques with a Symptom Hierarchy (SH) to enhance both the accuracy and interpretability of TLD detection. The proposed system employs a pre-trained ResNet-18 (RN18) model to extract visual features from leaf images while simultaneously identifying relevant symptom tags. These symptoms are organized hierarchically—from general indicators such as spots and discoloration to more specific patterns—allowing the system to rank potential diseases based on their severity and likelihood.
Additionally, Data Augmentation (DA) and class balancing techniques are applied to improve model reliability and minimize bias. Experimental results demonstrate that the proposed model achieves an accuracy of 92.5% and an F1- score of 91.8%, outperforming single-modality approaches. By combining CV with hierarchical symptom reasoning, the system provides early, reliable, and interpretable disease detection, empowering farmers to make informed and timely management decisions.
Keywords: Tomato Leaf Disease (TLD), Symptom Hierarchy (SH), ResNet-18 (RN18), Multi-Modal Diagnosis (MMD), Data Augmentation (DA), Leaf Disease Detection(LDD).
This study introduces a Multi-Modal Diagnosis (MMD) system that integrates CV techniques with a Symptom Hierarchy (SH) to enhance both the accuracy and interpretability of TLD detection. The proposed system employs a pre-trained ResNet-18 (RN18) model to extract visual features from leaf images while simultaneously identifying relevant symptom tags. These symptoms are organized hierarchically—from general indicators such as spots and discoloration to more specific patterns—allowing the system to rank potential diseases based on their severity and likelihood.
Additionally, Data Augmentation (DA) and class balancing techniques are applied to improve model reliability and minimize bias. Experimental results demonstrate that the proposed model achieves an accuracy of 92.5% and an F1- score of 91.8%, outperforming single-modality approaches. By combining CV with hierarchical symptom reasoning, the system provides early, reliable, and interpretable disease detection, empowering farmers to make informed and timely management decisions.
Keywords: Tomato Leaf Disease (TLD), Symptom Hierarchy (SH), ResNet-18 (RN18), Multi-Modal Diagnosis (MMD), Data Augmentation (DA), Leaf Disease Detection(LDD).
Abstract
AI Discharge & Readmission Prevention System
Ragul M, Derick Dilip, Ms. D. Kalpana
DOI: 10.17148/IJIREEICE.2025.131107
Abstract: Hospital readmission represents one of the most complex challenges in modern healthcare management, contributing significantly to increased operational costs, inefficient resource utilization, and patient dissatisfaction. This paper presents an Artificial Intelligence (AI)-driven Discharge and Readmission Prevention System designed to predict the probability of patient readmission using advanced machine learning models such as Random Forest and XGBoost. Unlike traditional methods that rely primarily on linear regression models and manual discharge procedures, the proposed system integrates real-time clinical data, patient medical history, and post-discharge behavioral indicators to generate accurate and explainable predictions. The framework emphasizes interpretability through explainable AI techniques, enabling clinicians to understand key contributing factors influencing readmission risk. By leveraging data-driven insights, the system aims to reduce preventable readmissions, enhance hospital workflow efficiency, and improve overall patient care quality. Experimental evaluations conducted on multiple healthcare datasets demonstrate the system’s superior predictive accuracy and scalability, showcasing its potential for real-world deployment in diverse hospital management environments.
Keywords: Artificial Intelligence, Machine Learning, Electronic Health Record, Healthcare Informatics, Predictive Analytics, Explainable AI
Keywords: Artificial Intelligence, Machine Learning, Electronic Health Record, Healthcare Informatics, Predictive Analytics, Explainable AI
Abstract
FINANCIAL FRAUD DETECTION
Sai Chandana Y, Rama Devi DP, Neethu Jimmy Joy, Neelam Sanjeev Kumar
DOI: 10.17148/IJIREEICE.2025.131108
Abstract: This project presents a machine learning-based financial fraud detection system designed to enhance the security and reliability of digital financial transactions. The model analyzes key transactional and behavioral features such as transaction amount, time, location, customer spending patterns, and account history to accurately detect fraudulent activities. Multiple classification algorithms were evaluated, including Logistic Regression, Support Vector Classification (SVC), Decision Tree, Random Forest, and Multilayer Perceptron (MLP). Among these, the Random Forest algorithm achieved the highest accuracy of 98.9%, demonstrating superior capability in handling imbalanced and complex financial datasets. The system was deployed as an interactive web application using Streamlit, enabling real- time fraud prediction and alert generation. This work highlights the potential of ensemble and deep learning approaches for secure, data-driven financial systems, offering an efficient and scalable solution to mitigate fraud risks and enhance transaction safety.
Keywords: machine learning, financial fraud detection, random forest, anomaly detection, decision tree, SVM, logistic regression, multilayer perceptron.
Keywords: machine learning, financial fraud detection, random forest, anomaly detection, decision tree, SVM, logistic regression, multilayer perceptron.
Abstract
Predictive Mental Health Risk Assessment in Undergraduate Students Using Personalized Adaptive Machine Learning
Saai Jaswant, Ahamed Saif, Aishwarya Balaji, Sree Mouthika, Neelam Sanjeev Kumar
DOI: 10.17148/IJIREEICE.2025.131109
Abstract: Mental health challenges among college students have reached epidemic proportions, with approximately 40% of students experiencing significant psychological distress during their university years. Traditional reactive approaches to student mental health support have proven inadequate in addressing the scale and complexity of this crisis affecting millions of students globally. This study presents an Enhanced Adaptive K-Nearest Neighbor (EAKNN) algorithm for early detection of psychological issues in college students through intelligent analysis of behavioral patterns and temporal changes. The proposed system learns individual baseline behaviors, adapts to personal patterns, and identifies concerning deviations that may indicate developing mental health concerns before they escalate. Experimental validation with 200 college students over six months demonstrated that EAKNN achieved 92.5% overall accuracy, 92.7% sensitivity, and 91.1% precision in identifying at-risk students an average of 3.7 weeks before traditional screening methods. The algorithm incorporates temporal weighting, adaptive feature importance, and explicit uncertainty quantification to provide personalized, interpretable assessments tailored to individual circumstances. Statistical analysis confirmed significant improvements over baseline methods including Random Forest (p < 0.001, Cohen's d = 0.82). This research demonstrates that personalized, adaptive machine learning approaches can transform mental health support from reactive crisis management to proactive early intervention, potentially improving outcomes for thousands of students while optimizing limited counseling resources.
Keywords: Mental Health Detection, Machine Learning, EAKNN Algorithm, College Students, Early Detection, Behavioral Analysis, Predictive Analytics, Student Wellbeing.
Keywords: Mental Health Detection, Machine Learning, EAKNN Algorithm, College Students, Early Detection, Behavioral Analysis, Predictive Analytics, Student Wellbeing.
Abstract
IoT-Based Smart Waste Management System for Smart Cities: A Synthesized Framework Addressing Security and Interoperability Challenges
Prof. Rana Afreen Sheikh, Rushikesh R. Gunjarkar, Snehal R. Nagrale
DOI: 10.17148/IJIREEICE.2025.131110
Abstract: The rapid urbanization of modern cities has exacerbated the challenges of efficient waste management, leading to environmental and public health concerns. The Internet of Things (IoT) presents a transformative opportunity to create smart, data-driven waste management systems. However, the deployment of such systems is hindered by significant challenges, including device interoperability, data security, and the need for real-time processing. This paper synthesizes a comprehensive framework for an IoT-based smart waste management system by integrating principles of interoperability, real-time data processing, and decentralized security, drawing upon established research in adjacent smart city domains. We propose a multi-layered architecture that leverages an interoperable IoT platform for device integration, a Kappa architecture for real-time and batch analytics, and Blockchain technology for ensuring data integrity and trust. The resulting framework offers a blueprint for developing scalable, secure, and efficient waste management solutions that are essential for the sustainable development of smart cities.
Keywords: Smart Cities, Internet of Things (IoT), Waste Management, Cybersecurity, Interoperability, Blockchain, Kappa Architecture.
Keywords: Smart Cities, Internet of Things (IoT), Waste Management, Cybersecurity, Interoperability, Blockchain, Kappa Architecture.
Abstract
ATS Resume Classification Project Analysis
Vibhu Sahu, Kunal Meshram, Anshita Tripathi
DOI: 10.17148/IJIREEICE.2025.131111
Abstract: The high volume of applications in modern recruitment processes presents a significant challenge to hiring managers, making manual resume screening inefficient and prone to bias. Traditional Applicant Tracking Systems (ATS) often rely on basic keyword matching, which lacks the nuance to accurately classify candidates into appropriate job domains. This paper introduces an intelligent ATS designed to automate and enhance the initial screening phase using a classic, yet effective, machine learning architecture. The system leverages a TF-IDF vectorizer to convert resume text into numerical features and a Multinomial Naive Bayes classifier for high-speed job domain categorization. By processing PDF and CSV resumes, the system cleans and standardizes textual data before feeding it into the ML pipeline for prediction. This paper provides a complete blueprint for the development and deployment of this application, from the data preprocessing and model training stages to the design of its user-facing frontend and backend API. Future work will focus on migrating from lexical-based models to modern contextual embeddings to improve semantic understanding and predictive accuracy.
Keywords: Applicant Tracking System (ATS), Natural Language Processing (NLP), Resume Classification, TF-IDF, Multinomial Naive Bayes, HR Technology, Machine Learning, Text Classification.
Keywords: Applicant Tracking System (ATS), Natural Language Processing (NLP), Resume Classification, TF-IDF, Multinomial Naive Bayes, HR Technology, Machine Learning, Text Classification.
Abstract
Comparative Predictive Modeling of Dry Eye Disease: An Integrated Approach Using Decision Tree and Random Forest Techniques
Mohammed Ihsan N, Sharan R, Dr. Paavai Anand
DOI: 10.17148/IJIREEICE.2025.131112
Abstract: Dry Eye Disease (DED) is a multifactorial ocular disorder characterized by tear film instability and ocular surface inflammation, manifesting as discomfort and visual disturbances. Traditional diagnostic methods rely on subjective clinical evaluation and costly procedures, limiting accessibility. This work proposes a machine learning-based, non-invasive approach for predicting DED risk using patient demographics, lifestyle, and reported symptoms. Both Decision Tree and Random Forest classifiers are compared: Random Forest achieves superior accuracy (72.8\%) and F1- score (0.76). Feature importance ranks symptomology and behavioural factors as key predictors, supporting practical early intervention strategies.
Keywords: dry eye disease, machine learning, decision tree, random forest, predictive modelling, feature importance
Keywords: dry eye disease, machine learning, decision tree, random forest, predictive modelling, feature importance
Abstract
Forest Fire Severity Prediction using Random Forest and Neural Network Stacking with SMOTE
Mani Shankar B, Jai Vignesh K, Arjun Ashkar N C, Dr G. Paavai Anand
DOI: 10.17148/IJIREEICE.2025.131113
Abstract: Forest fire prediction is essential for sustainable forest management, as wildfires cause significant ecological, environmental, and economic damage. This paper introduces a hybrid stacking model that improves the precision and resilience of forest fire risk prediction by merging Random Forest and Neural Network classifiers with Logistic Regression acting as a meta-learner. To solve the problem of class imbalance, new data samples were introduced and the SMOTE was used to add more high-risk fire events to the dataset, which was originally sourced from Kaggle. Additionally, feature engineering was used to create new variables that captured intricate connections between environmental and meteorological aspects. The suggested stacking framework achieves an accuracy of 87.2% on tests and an Area Under the Curve of 0.925 by combining the interpretability of RF with the nonlinear learning strength of NN. Strong dependability in differentiating between low-risk and high-risk fire incidents is demonstrated by these data. Overall, this approach provides an effective, data-driven foundation for intelligent wildfire monitoring and proactive forest management.
Keywords: Forest Fire Prediction, Random Forest, Neural Network, SMOTE, Machine Learning, Stacking, Feature Engineering.
Keywords: Forest Fire Prediction, Random Forest, Neural Network, SMOTE, Machine Learning, Stacking, Feature Engineering.
Abstract
Machine Learning Approaches to Passenger Survival Prediction: A Titanic Dataset Analysis
ESWARAMUTHU M, M KIRITHIKA, ABHISHEK R, LAVANYA S, MOHANA KRISHNAN B, VAISHNAVI S, Dr. M. ULAGAMMAI
DOI: 10.17148/IJIREEICE.2025.131114
Abstract: The Spaceship Titanic competition on Kaggle offers a unique machine learning challenge inspired by the classic Titanic disaster, reimagined in a futuristic space setting. The goal of this research is to create a predictive model that decides if passengers on the Spaceship Titanic were transported to another dimension after the ship collided with a spacetime anomaly. This study examines different data preprocessing, feature engineering, and classification techniques to improve predictive performance on the dataset.
The dataset provides details on passenger demographics, travel itineraries, and onboard service usage, offering rich information for building classification models. Our approach involves meticulous data cleaning, handling missing values through imputation, normalizing numeric features, and encoding categorical variables. We evaluated several supervised learning algorithms—Logistic Regression, Random Forest, Gradient Boosting, and XGBoost—using metrics like accuracy, precision, recall, and F1-score. The top-performing model demonstrated strong generalization on the test set, highlighting the value of analyzing feature interactions and fine-tuning hyperparameters. This research underscores the power of data-driven techniques in predictive analytics and illustrates how machine learning can effectively tackle complex, hypothetical problems with incomplete and noisy data. Our findings contribute to the broader understanding of structured data prediction and highlight the effectiveness of ensemble methods for classification in both practical and simulated scenarios.
The dataset provides details on passenger demographics, travel itineraries, and onboard service usage, offering rich information for building classification models. Our approach involves meticulous data cleaning, handling missing values through imputation, normalizing numeric features, and encoding categorical variables. We evaluated several supervised learning algorithms—Logistic Regression, Random Forest, Gradient Boosting, and XGBoost—using metrics like accuracy, precision, recall, and F1-score. The top-performing model demonstrated strong generalization on the test set, highlighting the value of analyzing feature interactions and fine-tuning hyperparameters. This research underscores the power of data-driven techniques in predictive analytics and illustrates how machine learning can effectively tackle complex, hypothetical problems with incomplete and noisy data. Our findings contribute to the broader understanding of structured data prediction and highlight the effectiveness of ensemble methods for classification in both practical and simulated scenarios.
Abstract
A Hybrid Machine Learning Approach for Apple (AAPL) Stock Price Prediction Using Ensemble Methods and Anomaly Detection
Gautam R, Shreya R, Dr G. Paavai Anand
DOI: 10.17148/IJIREEICE.2025.131115
Abstract: This paper presents a rigorous comparative analysis of ensemble tree-based models and deep sequential networks for financial time-series prediction, focusing on Apple Inc. stock data. The methodology employs a sophisticated feature engineering pipeline incorporating technical indicators (Relative Strength Index, Moving Average Convergence Divergence, Bollinger Bands), volatility metrics, and lagged returns. We integrate an anomaly detection component using Isolation Forest for both data cleaning and market surveillance. The study targets two primary objectives: regression (forecasting daily Close Price) and classification (predicting 5-day directional movement, H=5. Our results show that the CNN-LSTM regression model achieves an R2 score of 0.6695, demonstrating a strong statistical fit for continuous value prediction21. However, the Ensemble Classification approach, specifically the Stacked Ensemble, offers a superior and more actionable directional signal, achieving 80.54% accuracy after optimization via threshold tuning on a validation set. This is supplemented by a parallel GARCH(1,1) volatility analysis, which provides a robust framework for forecasting risk. The analysis confirms the critical role of Isolation Forest in identifying and mitigating the impact of outliers. The discussion highlights the crucial trade-off between the high interpretability and efficiency of tree-based models and the potential temporal dependency capture of deep learning architecture. Practical deployment recommendations favour tree-based models for high-volume, real-time trading signals, reserving the resource-intensive sequential models for strategic, offline risk analysis.
Keywords: Ensemble Methods, Stock Price Prediction, Anomaly Detection, Isolation Forest, CNN-LSTM, GARCH
Keywords: Ensemble Methods, Stock Price Prediction, Anomaly Detection, Isolation Forest, CNN-LSTM, GARCH
Abstract
Comparative Analysis of User-Based and Item-Based Collaborative Filtering Using the MovieLens Dataset
M. V. Karthikeya, Vishnu Vardhan, D.Nanda Kishore, Subhrajit Panda, Sai Tejas, A. Narendrasai B Sharath Reddy, Dr. M. ULAGAMMAI
DOI: 10.17148/IJIREEICE.2025.131116
Abstract: In today's Now, recommender systems in the present time become an essential tool to screen out relevant content for users. This paper constitutes a comparative study of user-based and item-based collaborative filtering (CF), using MovieLens dataset. Methodology, the details of implementation, and evaluation on metrics RMSE, Precision@K, Recall@K, and Coverage are described. A working web prototype of the recommendation system is presented, with screenshots demonstrating its functioning. Results showed that Item-Based CF has more stability and accuracy for large recommendation tasks while User-Based CF still captures dynamic user similarities.
Keywords: Recommender Systems; Collaborative Filtering; MovieLens; User Based CF; Item Based CF; Evaluation Metrics.
Keywords: Recommender Systems; Collaborative Filtering; MovieLens; User Based CF; Item Based CF; Evaluation Metrics.
Abstract
Song Skip Prediction Using XGBoost: A Machine Learning Approach for Music Recommendation Systems
Surya Sivakumar, Koduri Pranav, Tharun kumaar SD, Dr G.Paavai Anand
DOI: 10.17148/IJIREEICE.2025.131117
Abstract: With the rapid evolution of digital music platforms like Spotify and Apple Music, understanding user behavior has become crucial for enhancing music recommendation systems. One key behavioral indicator is song skipping, which reflects user preferences and engagement levels. This study introduces a Song Skip Prediction System using the XGBoost (Extreme Gradient Boosting) algorithm to predict whether a user will skip a song based on playback and contextual features. The dataset, comprising 149,860 interaction records, was preprocessed through encoding, scaling, and feature selection to ensure data consistency and accuracy.
The model achieved an impressive accuracy of 97.27% and an ROC-AUC score of 0.9770, outperforming traditional ensemble methods. Evaluation metrics such as the confusion matrix and ROC curve confirmed its strong discriminative performance. To prevent overfitting and data leakage, techniques like cross-validation and regularization were employed. The trained model was deployed using a Flask backend with a React-based frontend, allowing real-time skip predictions through a user-friendly interface. Overall, this work demonstrates how XGBoost can effectively model user listening behavior, offering a scalable foundation for intelligent and personalized music recommendation systems.
Keywords: Music recommendation, user behavior, song skip prediction, XGBoost, machine learning, user engagement, Flask, React, real-time prediction, personalization
The model achieved an impressive accuracy of 97.27% and an ROC-AUC score of 0.9770, outperforming traditional ensemble methods. Evaluation metrics such as the confusion matrix and ROC curve confirmed its strong discriminative performance. To prevent overfitting and data leakage, techniques like cross-validation and regularization were employed. The trained model was deployed using a Flask backend with a React-based frontend, allowing real-time skip predictions through a user-friendly interface. Overall, this work demonstrates how XGBoost can effectively model user listening behavior, offering a scalable foundation for intelligent and personalized music recommendation systems.
Keywords: Music recommendation, user behavior, song skip prediction, XGBoost, machine learning, user engagement, Flask, React, real-time prediction, personalization
Abstract
Sentiment Analysis on Customer Reviews Using Fine-Tuned DistilBERT Transformer Model
Caroline Vineeta S L, Vedika Singh, Asritha D, G. Paavai Anand
DOI: 10.17148/IJIREEICE.2025.131118
Abstract: Customer feedback has grown in importance as a source of information on user satisfaction, requirements, and expectations due to the rise of digital communication and the growth of online platforms. But sifting through thousands of assessments by hand is slow and frequently wrong, which emphasizes the need for automated alternatives. A refined DistilBERT transformer model that can automatically categorize app reviews into positive, negative, or neutral sentiments is shown in this study. Even in complex language that include slang, acronyms, sarcasm, or emoticons, the model is able to identify emotions and tone because to transformer-based contextual embeddings. The process begins with gathering and cleaning review data, removing superfluous symbols, tokenizing text, and then using the DistilBERT tokenizer to transform it into numerical form.The model is then fine-tuned on a labeled dataset to capture sentiment patterns and contextual relationships accurately. To assess its effectiveness, performance metrics such as accuracy, precision, recall, and F1-score are used, ensuring dependable results.In summary, this system provides an intelligent, efficient, and scalable approach for businesses to automatically analyze customer feedback, track sentiment trends, and make data-informed decisions that improve overall user experience and satisfaction.
Keywords: App Store Reviews, Emotion analysis, Sentiment classification, Sentiment features, Machine learning, Natural Language Processing (NLP).
Keywords: App Store Reviews, Emotion analysis, Sentiment classification, Sentiment features, Machine learning, Natural Language Processing (NLP).
Abstract
INNOVATIVE MPPT TRACKING TECHNIQUES FOR INTEGRATED WIND AND PHOTOVOLTAIC ENERGY SYSTEMS
M. Divakar, U. Lakshmi
DOI: 10.17148/IJIREEICE.2025.131119
Abstract: Traditional techniques employed to monitor the maximum power point in wind energy conversion systems (WECS) encounter numerous challenges. One of the most prevalent methods is the perturb and observe (P&O) algorithm, which tracks and logs the highest achievable power point. Nevertheless, a significant drawback of this algorithm is the challenge of determining an optimal step size. To tackle this problem, this research introduces an innovative approach that combines fuzzy logic control with the trapezoidal rule. The suggested method is evaluated against two existing techniques: the trapezoidal rule-based P&O (TRPO) algorithm and the standard P&O method. MATLAB/Simulink simulations are conducted to assess the performance of all three algorithms under randomly fluctuating wind speeds. The findings indicate that the proposed approach markedly decreases power oscillations while improving DC output current, voltage, and power. Furthermore, this study expands the methodology by integrating a wind-solar hybrid system with an artificial neural network (ANN) model, thereby enhancing the efficiency and stability of renewable energy generation.
Abstract
Password Strength Prediction using Regression-Based Machine Learning Models with Entropy and Caesar Cipher–Driven Synthetic Dataset
Dr G. Paavai Anand, Roshni Y
DOI: 10.17148/IJIREEICE.2025.131120
Abstract: In today’s digital era, weak and predictable passwords remain a major cause of cybersecurity breaches. This paper presents a novel machine learning–based password strength prediction model using regression techniques on a synthetically generated dataset enhanced with Caesar cipher transformations, incorporating features such as length, character composition, whitespace inclusion, and entropy. Linear, Random Forest, and Gaussian Process Regression models are compared, with Gaussian Regression achieving the highest accuracy (R² = 0.9998), providing a scalable and interpretable framework for real-time password strength evaluation.
Keywords: Password Strength Evaluation, Machine Learning, Regression Analysis, Shannon Entropy, Password Security, Caesar Cipher, Random Forest Regression, Gaussian Process Regression, Synthetic Dataset, Cybersecurity, Feature Engineering.
Keywords: Password Strength Evaluation, Machine Learning, Regression Analysis, Shannon Entropy, Password Security, Caesar Cipher, Random Forest Regression, Gaussian Process Regression, Synthetic Dataset, Cybersecurity, Feature Engineering.
Abstract
A Machine Learning Approach for E-Commerce Counterfeit Product Detection Using Transactional and Behavioral Data
Nilesh J, Ashwin C, Anoop Mahesh, Dr G. Paavai Anand
DOI: 10.17148/IJIREEICE.2025.131121
Abstract: Identifying fraudulent transactions has become a crucial task for maintaining digital security and customer confidence due to the quick growth of e-commerce platforms. Based on customer, payment, and behavioral characteristics, this study introduces a Counterfeit Transaction Detection System that uses the Random Forest algorithm to identify transactions as either authentic or fraudulent. To increase the accuracy and dependability of the model, the dataset was preprocessed using techniques like feature engineering, encoding, scaling, and data cleaning. The suggested model performed well on precision, recall, and F1-score metrics, achieving a high classification accuracy of 96.85%. Cross-validation methods were used to improve generalization and reduce overfitting. A Streamlit-based interface was used to deploy the trained model, allowing users to upload transaction data and get predictions about authenticity in real time. All things considered, this study demonstrates how well machine learning works to prevent online fraud and improve transaction security in e-commerce platforms.
Keywords: Counterfeit Detection, E-Commerce, Machine Learning, Random Forest, Fraud Analytics, Feature Engineering
Keywords: Counterfeit Detection, E-Commerce, Machine Learning, Random Forest, Fraud Analytics, Feature Engineering
Abstract
Industrial pollution Monitoring System
E.Sajjan, M.Shivaleela, T.Priyanka, L.Praveen, S.Sanjay
DOI: 10.17148/IJIREEICE.2025.131122
Abstract: Industrial pollution is a significant environmental concern, affecting air, water, and soil quality. This paper presents a real-time Industrial Pollution Monitoring System (IPMS) using Internet of Things (IoT) sensors and cloud computing. The proposed system integrates various IoT sensors to monitor parameters such as particulate matter (PM), nitrogen dioxide (NO2), ozone (O3), and carbon monoxide (CO). The sensor data is transmitted to a cloud-based platform using wireless communication protocols. The cloud platform analyzes the data and provides real-time insights on pollution levels, enabling industries to take corrective measures. The system also includes a mobile app for alerts and notifications. The proposed IPMS is a low-cost, scalable, and efficient solution for monitoring industrial pollution, enabling a cleaner and healthier environment.
Abstract
Automatic Vehicle Speed Regulation System
E.Sajjan, A.Shruthi, Ch.Sai Teja, G.Aakash, J.SaiKiran
DOI: 10.17148/IJIREEICE.2025.131123
Abstract: An automatic vehicle speed Regulation System is a technology designed to automatically Control a vehicle's speed based on pre-set parameters, typically utilizing motor driver Controllers to detect the current speed & adjust engine power accordingly, aiming to maintain Consistent speed enforce Speed excessive speed. The automatic vehicle Speed regulation system often referred to as a speed limit contour Cruise Control. The speed control system for a vehicle describes a method for regulating the velocity of a vehicle based on its current speed and external factors such as road conditions, implemented in school zones to reduce the number of accidents. This automated efficiently. The proposed speed control system aims to reduce accidents, improve fuel efficiency, and provide a comfortable ride for passengers. The abstract may also discuss the potential applications of the speed control system in different types of vehicles, such as cars, buses, and trains, and its impact on transportation systems and the environment. Here the Arduino is programmed in such a way that, the prescribed speed limit was corporate in the transmitter unit which transmits the signals, and it was received by the receiver in the vehicle using Zigbee wireless communication and the speed of the vehicle. was automatically controlled by the input signals by the receiver, with the help of speed encoder sensor. Once this technique was implemented the accidents will be reduced on a larger rate, and also reduce the nuisance by some.
Abstract
Toxic Comment Classification Using Ensemble Machine Learning Techniques
AATHITYA.A, KRISHITH TP, SARVESH S, KEVIN BENJAMIN SAMUEL, PRANAV M, VIGNESH D, Dr. M. ULAGAMMAI
DOI: 10.17148/IJIREEICE.2025.131124
Abstract: The exponential growth of social media and online discussion forums increases the difficulty of maintaining healthy digital spaces with growing volumes of toxic comments. This study designs an efficient machine learning model that can classify toxic content by employing techniques in Natural Language Processing and ensemble learning. The approach mixes the models Logistic Regression, Random Forest, and XGBoost into a framework based on Voting Ensemble, boosting predictive accuracy. By using TF-IDF for feature extraction, along with a soft voting mechanism, the proposed ensemble outperforms the stand-alone classifiers in both ROC-AUC and precision. The system proposed here will provide a robust, efficient, and scalable way to identify and manage toxicity online.
Keywords: Toxic comments, Natural Language Processing, TF-IDF, Ensemble Learning, Voting Classifier, XGBoost, Logistic Regression, Random Forest.
Keywords: Toxic comments, Natural Language Processing, TF-IDF, Ensemble Learning, Voting Classifier, XGBoost, Logistic Regression, Random Forest.
Abstract
Optimized Ensemble Learning Integrated with Anomaly Detection for Road Accident Severity Prediction
Rohith M, Sharan S, Suryaprakasam B, Dr. G. Paavai Anand
DOI: 10.17148/IJIREEICE.2025.131125
Abstract: This research introduces a hybrid framework combining ensemble-based learning and anomaly detection for the prediction of road accident severity. Conventional predictive systems often fail to manage noisy and imbalanced accident datasets effectively. To address this limitation, the proposed design integrates Decision Tree and Random Forest classifiers with clustering methods—KMeans and DBSCAN—for simultaneous classification and hotspot detection.
Contextual factors such as weather patterns, road conditions, traffic intensity, and casualty ratios are incorporated through tailored feature engineering. The Random Forest model achieved an accuracy of 83.6%, surpassing baseline methods. By fusing anomaly detection with ensemble classification, the framework not only enhances prediction accuracy but also provides interpretable insights for preventive traffic management and policy- making.
Keywords: Road Accident Severity · Anomaly Detection · Ensemble Models · Random Forest · Traffic Safety · Machine Learning.
Contextual factors such as weather patterns, road conditions, traffic intensity, and casualty ratios are incorporated through tailored feature engineering. The Random Forest model achieved an accuracy of 83.6%, surpassing baseline methods. By fusing anomaly detection with ensemble classification, the framework not only enhances prediction accuracy but also provides interpretable insights for preventive traffic management and policy- making.
Keywords: Road Accident Severity · Anomaly Detection · Ensemble Models · Random Forest · Traffic Safety · Machine Learning.
Abstract
INSURANCE CLAIM STATUS PREDICTION
Venkatapathy R, Senthilnathan M, Rohith G, Dr. G Paavai Anand
DOI: 10.17148/IJIREEICE.2025.131126
Abstract: Modern insurance industries increasingly rely on data-driven solutions to forecast the likelihood of claims, aiming to cut financial losses and streamline workflows. Traditional claim reviews are often inefficient and subjective, fueling the need for smart, automated prediction tools. This project develops a machine learning-based system to estimate which policyholders are most likely to file claims, leveraging detailed customer, vehicle, and policy data. Models like Logistic Regression, Random Forest, and Gradient Boosting are assessed on a dataset containing more than 58,000 samples and 41 variables. Advanced preprocessing and careful feature selection improve model stability. Results indicate that ensemble models outperform classic techniques in both accuracy and reliability, supporting better fraud detection, risk assessment, and premium calculation for insurers
Abstract
A Bibliographic Survey on Solar Policies and Regulations for the state of Himachal Pradesh
Jasmine Kaur
DOI: 10.17148/IJIREEICE.2025.131127
Abstract: This bibliographical survey presents a comprehensive review of rooftop solar installations in Himachal Pradesh up to 2025, analysing deployment data, policy incentives, regulatory reforms, and sectoral challenges. Himachal Pradesh, renowned for hydropower, is strategically shifting toward distributed solar generation, buoyed by state and central government initiatives such as the PM Surya Ghar Free Electricity Scheme and Swaran Jayanti Energy Policy 2021. By mid-2025, approximately 4,000 rooftop installations, totalling 14 MW, mark significant progress, although overall capacity remains modest at under 1 MW commissioned according to official sources. Substantial financial incentive including central and state subsidies totalling up to ₹85,800 per system have accelerated adoption across residential, institutional, and government buildings, notably in Shimla, Mandi, and Una. Regulatory reforms in 2024-25 have streamlined approval processes, introduced net metering, and clarified tariff and subsidy delivery mechanisms. Despite technical feasibility established by several institutional success stories, widespread adoption faces hurdles: procedural delays, grid-connection bottlenecks, suboptimal net metering frameworks, and administrative inefficiencies persist. Geographic constraints such as hilly terrain and variable solar irradiation complicate rooftop suitability and maintenance. This paper synthesizes academic recommendations advocating for high-resolution remote sensing methods to better target optimal installation sites in urban centers. Recent regulatory improvements signal optimism, but the paper posits that acceleration toward policy targets will require enhanced vendor accountability, greater data-driven outreach, flexible energy credit mechanisms, and robust execution of decentralized programs. Concluding, the survey highlights that while Himachal Pradesh’s rooftop solar sector demonstrates technical promise and progressive policy support, major barriers must be addressed to scale adoption for meaningful contributions to India’s renewable energy goals by 2030. The findings of this work shall serve as a valuable resource for stakeholders, providing a strategic outlook for sustainable deployment in mountainous regions.
Keywords: Rooftop solar, Renewable energy policy, Net metering, Regulatory framework, Solar reforms
Keywords: Rooftop solar, Renewable energy policy, Net metering, Regulatory framework, Solar reforms
Abstract
OPTIMUM CHARGING METHOD FOR E-VEHICLES
Ananya C, Dr. E. Latha Mercy
DOI: 10.17148/IJIREEICE.2025.131128
Abstract: The concept of a smart charging regulator, a novel technology designed to enhance battery efficiency and extend lifespan through intelligent control mechanism. Traditional charging techniques frequently fail to adapt to changing conditions, resulting in overcharging / undercharging or exposure to unfavourable working situations, all of which can harm battery health. The smart charging regulator is a potential improvement in battery management technology, providing a holistic approach to optimizing efficiency and lifespan. Uncoordinated charging of Electric Vehicles (EVs) can cause voltage fluctuations, power outages, and incremental overloads, all of which are taxing and damaging to the distribution networks. In this paper, a new method to minimize the cost of charging electric vehicles based on the Day-Ahead Electricity Price (DAEP) is proposed on taking battery degradation cost, taking into account the EVs' maximum power charger, state of charge (SOC) limits. The full charging of the EVs' batteries at the end of the charging period, and the distribution feeder subscribed power is analysed. Additionally, the EVs' initial SOC uncertainties are evaluated using their daily mileage. Lastly, a single-phase Low Voltage (LV) distribution network with an EV penetration rate of 50% and 100% has been installed in a residential area to demonstrate the effectiveness of the suggested strategy. In this paper, the linear programming approach is used to address the optimization problem. The findings demonstrate that, in comparison to uncoordinated EV charging, the suggested strategy can lower EV charging costs by 50% and 38% for 100% and 50% of EV penetration rate, respectively.
Keywords: battery efficiency, lifespan, volage fluctuations, cost of charging, linear approach.
Keywords: battery efficiency, lifespan, volage fluctuations, cost of charging, linear approach.
Abstract
Intelligent Switching System for Seamless Solar Inverter Transition Between Grid-Tied and Off-Grid Modes
Rajeswari Ramachandran, Sureshkumar Raghavan
DOI: 10.17148/IJIREEICE.2025.131129
Abstract: In conventional grid-tied solar power systems, a major limitation is their inability to supply power during grid outages, despite the availability of solar energy. This is due to the inverter’s dependency on a live grid signal to operate safely and prevent back-feeding. As a result, during blackouts, available solar power remains untapped, compromising the reliability of the system particularly for critical applications. To overcome this challenge, a "Modification System" was developed and tested using SimuRelay software. This enhanced system introduces an Intelligent Switching Mechanism capable of real-time monitoring and dynamic load management. During a grid outage, the system instantly detects the failure, isolates non-critical loads, and ensures uninterrupted power supply to critical loads by maintaining inverter operation independent of the grid. Simulation results validated this functionality, showcasing seamless transitions and sustained power delivery using solar energy. In normal grid conditions (On-Grid Mode), the system optimizes energy use by prioritizing utility power while placing solar and backup systems on standby. This intelligent control ensures energy efficiency, cost-effectiveness, and reduced strain on backup sources. Safety remains a core feature, with integrated protections such as MCCBs (Molded Case Circuit Breakers) for grid and critical circuits, and secure relay-based switching. These protections were effectively validated during simulation, confirming the systems robustness against faults and unsafe conditions. In conclusion, the Modification System significantly enhances the resilience, efficiency, and safety of solar installations. Through SimuRelay simulation, it has demonstrated its ability to ensure continuous power for essential loads, maximize solar utilization, and provide reliable performance in both on-grid and off-grid scenarios, making it a practical solution for modern energy needs.
Keywords: Intelligent Switching Mechanism, Critical Load Management and Seamless Grid-to-Backup Transition.
Keywords: Intelligent Switching Mechanism, Critical Load Management and Seamless Grid-to-Backup Transition.
Abstract
The Use of Interactive Whiteboards in Education: A Quantitative Analysis of Teachers’ Perceptions in Greece
Dr. Konstantinos N. Domouchtsis, Dr. Apostolos Ch. Klonis
DOI: 10.17148/IJIREEICE.2025.131130
Abstract: This study investigates Greek teachers’ perceptions and attitudes toward the use of Interactive Whiteboards (IWBs) in classroom instruction. A structured questionnaire was administered to 45 educators across different age groups and teaching specializations. The results reveal that the majority of teachers recognize the pedagogical benefits of IWBs in enhancing interactivity, improving student engagement, and facilitating differentiated instruction. However, actual usage levels vary significantly due to barriers such as insufficient training, limited infrastructure, and technical challenges. Younger educators and those with prior ICT training demonstrated higher rates of IWB adoption and more positive perceptions. The findings highlight the importance of professional development and institutional support to maximize the educational impact of interactive technologies in Greek schools.
Keywords: Interactive Whiteboard, Educational Technology, ICT in Education, Teacher Perceptions, Digital Transformation.
Keywords: Interactive Whiteboard, Educational Technology, ICT in Education, Teacher Perceptions, Digital Transformation.
Abstract
AI-Driven Plant Disease Prediction Using Deep Learning and Web-Based Deployment
Sujana Mallavaram, Pranesh S, Chandrika Banerjee, Shalini. M, Dr. Ulagammai. M
DOI: 10.17148/IJIREEICE.2025.131131
Abstract: Plant diseases seriously affect agricultural productivity and cause economic loss and food insecurity. Herein, this paper presents an AI-based plant disease prediction system that uses Convolutional Neural Networks for image- based diagnosis. The model was trained based on the PlantVillage dataset containing over 80,000 labeled images from 38 classes of diseases. The CNN model was able to classify crop leaf diseases with more than 95% accuracy. In addition, the system provides a Streamlit-based web interface that facilitates real-time disease prediction by enabling users to upload leaf images and get instant diagnostic feedback. Experimental results showed that the deployed system is scalable, low latency, and of high precision, and hence can be used for practical early disease detection and smart agriculture. Future enhancements include mobile deployment, analysis using real-time cameras, and integration with IoT-enabled sensors for precision farming.
Keywords: Plant Disease Detection, Convolutional Neural Network (CNN), Deep Learning, Streamlit, Image Classification, Smart Agriculture, Computer Vision, AI in Farming
Keywords: Plant Disease Detection, Convolutional Neural Network (CNN), Deep Learning, Streamlit, Image Classification, Smart Agriculture, Computer Vision, AI in Farming
Abstract
Predicting Road Accident Risk
Sushanth. A, Moses. C M, Ashmith. S, Febi Andrew. R, Pranav Sakthi. S, Keshiv Raajh. SK, Joel Sam. S R, M. Ulagammai
DOI: 10.17148/IJIREEICE.2025.131132
Abstract: Road accidents constitute one of the leading causes of fatalities and economic losses worldwide. Predicting the likelihood of such incidents using data-driven approaches can significantly enhance road safety management and resource allocation. This paper presents an ensemble-based machine learning framework for predicting road accident risk, integrating multiple gradient boosting models—XGBoost, LightGBM, and CatBoost. The proposed ensemble combines the predictive strengths of each model through weighted averaging to minimize Root Mean Square Error (RMSE) and improve generalization across diverse driving conditions. Extensive experiments were conducted on the Kaggle Playground Series (Season 5, Episode 10) dataset, which contains multi-dimensional traffic, environmental, and temporal attributes. The ensemble achieved an RMSE of 0.1346, outperforming individual learners and demonstrating the effectiveness of hybrid boosting in accident risk assessment. The study provides valuable insights into the influence of key features such as speed limit, road surface condition, and weather index, offering a scalable model for intelligent transport and safety analytics.
Keywords: Road Accident Risk, Machine Learning, Ensemble Learning, Gradient Boosting, Traffic Prediction.
Keywords: Road Accident Risk, Machine Learning, Ensemble Learning, Gradient Boosting, Traffic Prediction.
Abstract
Recognition of Handwritten Digit using Convolutional Neural Network in Python
Chandhan Sayee.R, Mohul Raj.T, Merlin.R. S, Sanjith Kumar.K, Nishanth.G. S, Monesh.V, M. Ulagammai
DOI: 10.17148/IJIREEICE.2025.131133
Abstract: In recent years, deep learning has become a very important part of artificial intelligence. With the growth of Artificial Neural Networks (ANN), deep learning has made machines much smarter and more capable. It is used in many areas such as health care, security, sports, robotics, and drones because it can solve many real-life problems. Among deep learning methods, the Convolutional Neural Network (CNN) is one of the most successful. It combines the ideas of ANN with modern deep learning methods. CNNs are widely used for pattern recognition, speech and face recognition, text and document classification, scene detection, and handwritten digit recognition.
Keywords: Handwritten Digit Recognition, Convolutional Neural Network,MNIST, Batch Normalization, Dropout, Data Augmentation, Learning Rate Scheduling, Deep Learning.
Keywords: Handwritten Digit Recognition, Convolutional Neural Network,MNIST, Batch Normalization, Dropout, Data Augmentation, Learning Rate Scheduling, Deep Learning.
Abstract
SMART GARBAGE MONITORING SYSTEM USING IoT
B. Sandeep Kumar, P. Vyshnavi, Sana, Y. Rahul Kumar, N. Pranay
DOI: 10.17148/IJIREEICE.2025.131134
Abstract: IoT based smart Garbage monitoring system is a solution designed to automate waste management using modern technology. The system uses a ESP8266 microcontroller along with an ultrasonic sensor, GPS module and GSM module to monitor garbage levels and send alerts The ultrasonic sensor detects the fill level of the garbage bin, and the GPS module tracks its location once the bin is full, the GSM module sends an SMS notification to the waste Collection authorities, allowing them to pick up the trash Promptly. The system helps in reducing unnecessary garbage collection trips, improving efficiency and Promoting a cleaner environment by ensuring timely waste disposal. This IoT-based system not only improves the efficiency of waste management but also promotes sustainability by minimizing unnecessary trips and optimizing resource utilization. Furthermore, it helps reduce environmental pollution, carbon emissions, and operational costs associated with waste collection.
Keywords: ESP8266 Microcontroller, GSM, GPS, Smart Garbage Monitoring, SMS Notification, Ultrasonic Sensor, Waste Management
Keywords: ESP8266 Microcontroller, GSM, GPS, Smart Garbage Monitoring, SMS Notification, Ultrasonic Sensor, Waste Management
Abstract
IoT Based Surveillance Robot
D Jagan, M Archana, D Varsha, B Rahul, R Pavan Kumar
DOI: 10.17148/IJIREEICE.2025.131135
Abstract: The main objective is to develop a surveillance robot to perform surveillance activities in industrial areas, militarized war zones or radioactive field areas with the objective of analysing, governing and protecting the areas from unwanted threats. The use of robots and their role in our day-to-day life has been rapidly increasing since the day they were introduced to the world, further reducing the errors and life risk to humans.
The objective is to design and develop an Internet of Things (IoT) based surveillance robot at a low cost that will roam around freely and give live updates about their surroundings by broadcasting video and information. The sensor collects the data from the surroundings and send it to the Arduino microcontroller which can be seen by the user any time. This technology is controlled by the user remotely through any device such as mobile phone, tablet or laptop with the help of IoT based services. The entire project is built and monitored by wireless platform to minimize the use of wire and help it work smoothly in remote places.
The objective is to design and develop an Internet of Things (IoT) based surveillance robot at a low cost that will roam around freely and give live updates about their surroundings by broadcasting video and information. The sensor collects the data from the surroundings and send it to the Arduino microcontroller which can be seen by the user any time. This technology is controlled by the user remotely through any device such as mobile phone, tablet or laptop with the help of IoT based services. The entire project is built and monitored by wireless platform to minimize the use of wire and help it work smoothly in remote places.
Abstract
Microcontroller Based Automatic Engine Locking System for Drunken Drivers
B. Sandeep kumar, P. Udayasri, P. Shivashankar, E. Saikumar, G. Vishal
DOI: 10.17148/IJIREEICE.2025.131136
Abstract: This project presents a novel approach to prevent accidents caused by drunk driving. The proposed system is sufficient, reliable and can be integrated with existing vehicle system. This innovation has the potential to significantly reduce the risk of accidents caused by drunk driving, promoting road safely and responsible driving habits. The system consists of a microcontroller, breathalyzer sensor, relay module, LCD display, motor driver and buzzer. The micro controller processes the sensor data and controls the engine locking system. The system provides a safe and efficient solution to prevent accidents caused by drunken driving. Its accuracy, reliability, and ease of use make it a viable solution for implementation in vehicles.
Keywords: Power supply, Arduino Uno, L293D, LCD, BO motor, Buzzer, MQ3â… .
Keywords: Power supply, Arduino Uno, L293D, LCD, BO motor, Buzzer, MQ3â… .
Abstract
SMART FLOOR CLEANING ROBOT BY IoT
B. Hanumanthu, B. Prasanna, A. Ramya sree, K. Achala, A. Raghavendra
DOI: 10.17148/IJIREEICE.2025.131137
Abstract: The floor cleaning robot aims to create an autonomous robot that can clean floors without human intervention. Households of today are becoming smarter and more automated cleaning of floor is very important role in our health and this robot reduces man power requirement. This project is used for domestic purpose to clean the surface automatically. Cleaning the dust from the floor is one of the daily tasks that must be completed. This is a common practice not only at home, but also at companies and shopping malls. Due to the fact that dust cleaning operations take a long time, other activities are sometimes disregarded. To eradicate this problem, we came up with this project based on cleaning mechanism, our robot can reach out to places where human access is not possible. Taking the advantage of advancements achieved in mechanical technology innovation have made human life much easier and more pleasant.
Keywords: Smart Cleaning Robot, IoT-based Floor Cleaner, Home Automation, Smart Home Robotics, IoT for Domestic Automation.
Keywords: Smart Cleaning Robot, IoT-based Floor Cleaner, Home Automation, Smart Home Robotics, IoT for Domestic Automation.
Abstract
Vehicle-to-Vehicle (V2V) Communication: Design, Components, Benefits, Challenges, and Implementation Using Embedded Systems
G. Jagan, Afreen, D. Navya, D. Venu, M. Manoj
DOI: 10.17148/IJIREEICE.2025.131138
Abstract: Vehicle-to-Vehicle (V2V) communication is a key technology in Intelligent Transportation Systems (ITS), enabling vehicles to share real-time information such as speed, braking, and hazard warnings. Unlike conventional sensors limited by line-of-sight, V2V enhances cooperative awareness to reduce accidents, improve traffic flow, and optimize fuel efficiency. This paper presents the design of a low-cost prototype using Arduino Uno, ultrasonic sensors, tilt sensors, and HC-12 wireless modules, supported by an LCD, buzzer, and regulated power supply for driver alerts. The system detects obstacles, monitors driver attentiveness, and transmits warnings within a 1 km range. While the prototype shows feasibility for affordable deployment, challenges such as limited range, security, and standardization must be addressed. Future integration with 5G, C-V2X, and AI-based traffic management can further enhance safety and scalability.
Keywords: Vehicle-to-Vehicle (V2V), Intelligent Transportation Systems (ITS), Embedded Systems, Arduino, Ultrasonic Sensor, Tilt Sensor, HC-12 Wireless Module, Road Safety, Collision Avoidance, Cooperative Driving.
Keywords: Vehicle-to-Vehicle (V2V), Intelligent Transportation Systems (ITS), Embedded Systems, Arduino, Ultrasonic Sensor, Tilt Sensor, HC-12 Wireless Module, Road Safety, Collision Avoidance, Cooperative Driving.
Abstract
Enhancing Railway Accident Prevention Using Deep Learning, Machine Learning, And GPS Tracking: A Historical And Knowledge-Based Analysis
VIKAS CHANDRA GIRI, PARINEETA JHA
DOI: 10.17148/IJIREEICE.2025.131139
Abstract: This research extends the foundational study “Enhancing Railway Accident Prevention Using Deep Learning, Machine Learning, and GPS Tracking” (Vikas Chandra Giri & Parineeta Jha, 2025). The present work introduces an adaptive framework that integrates Deep Learning (DL), Internet of Things (IoT) sensor networks, and real-time GPS analytics to advance predictive railway safety. With results Sets in these volume.The architecture enhances detection accuracy, reduces response latency, and improves contextual awareness using hybrid CNN–LSTM–Random Forest fusion models with a multi-layer IoT sensing infrastructure. The model achieves 97.8% accuracy and maintains average GPS latency under 4 seconds, marking a significant improvement over earlier implementations.Through continuous learning and edge-cloud synchronization, the system advances toward fully autonomous accident prevention, predictive maintenance, and real-time operational intelligence across railway networks.
Keywords: Deep Learning · IoT · LSTM · GPS Tracking · Predictive Analytics · Accident Prevention. ML, MQTT
Keywords: Deep Learning · IoT · LSTM · GPS Tracking · Predictive Analytics · Accident Prevention. ML, MQTT
Abstract
The Use of Educational Robotics in Greek Education: Trends, Challenges, and Future Prospects
Dr. Konstantinos N. Domouchtsis, Dr. Apostolos Ch. Klonis
DOI: 10.17148/IJIREEICE.2025.131140
Abstract: This study provides a qualitative and analytical overview of the current status and implementation challenges of Educational Robotics (E.R.) in the Greek education system. Drawing upon a synthesis of national policy reports, academic research, and statistics from leading Greek educational robotics organizations, the analysis highlights E.R.'s critical role in promoting STEM competencies and Computational Thinking. Findings confirm that E.R. integration is strongly perceived as beneficial for increasing student engagement and developing 21st-century skills. However, the effective transition from pilot projects to systematic classroom practice is hindered by significant barriers, primarily insufficient, targeted professional development for educators and the lack of coherent curriculum integration. This paper emphasizes the necessity of strategic planning and sustained governmental support to maximize the pedagogical impact of E.R. across all educational levels in Greece.
Keywords: Interactive Whiteboard, Educational Technology, ICT in Education, Teacher Perceptions, Digital Transformation
Keywords: Interactive Whiteboard, Educational Technology, ICT in Education, Teacher Perceptions, Digital Transformation
Abstract
Development of an Laser-Based Border Security System
K. Chiranjeevi, J Srujana, B Shailaja, M Mohanasrija, S Sairam
DOI: 10.17148/IJIREEICE.2025.131141
Abstract: This paper presents a design and implementation of an advanced laser-based border security system aimed at detecting intrusions and enhancing surveillance capabilities along sensitive border areas. The system utilizes laser beams as invisible security boundaries, coupled with Light Dependent Resistor (LDR) sensors or photodetectors to sense interruptions caused by intruders. Upon detection, the system triggers alarms and alerts control rooms for rapid response. Integration with IoT enables real-time remote monitoring and automated alarm management. The system provides a cost-effective, scalable, and reliable method for border security with potential for operation in diverse environmental conditions.
Securing national borders against unauthorized entry is a critical challenge requiring sophisticated technology solutions. Conventional systems based on radar, ultrasonic sensors, or human patrols have limitations in cost, range, reliability, or manpower requirements. Laser based security systems offer advantages due to their ability to create narrow, invisible light beams that can cover long distances without scattering, enabling the establishment of defined security perimeters. This paper proposes a laser-based border security system that detects
The laser security system is implemented using Arduino, with sensors calibrated to distinguish between normal environmental light changes and true intrusions. IoT connectivity is established via Wi-Fi modules to enable remote access. The system also supports integration with cameras and other sensors such as ultrasonic or vibration sensors for multi-modal detection if needed
The invisible but direct laser beam ensures discreet monitoring. The system can operate continuously in various weather conditions and is scalable for large border areas. Integration with IoT allows timely alarms and remote monitoring thereby reducing manpower and enhancing security effectiveness. An additional layer of functionality is added through the GSM module, which allows for remote communication and control. The GSM module enables the system to send SMS notifications to predefined mobile numbers, informing users about door access events, unauthorized attempts, or system status updates. Moreover, it allows users to interact with the system remotely by sending specific SMS or GPRS commands to lock or unlock the door from a distance, adding a high level of convenience and control.
Securing national borders against unauthorized entry is a critical challenge requiring sophisticated technology solutions. Conventional systems based on radar, ultrasonic sensors, or human patrols have limitations in cost, range, reliability, or manpower requirements. Laser based security systems offer advantages due to their ability to create narrow, invisible light beams that can cover long distances without scattering, enabling the establishment of defined security perimeters. This paper proposes a laser-based border security system that detects
The laser security system is implemented using Arduino, with sensors calibrated to distinguish between normal environmental light changes and true intrusions. IoT connectivity is established via Wi-Fi modules to enable remote access. The system also supports integration with cameras and other sensors such as ultrasonic or vibration sensors for multi-modal detection if needed
The invisible but direct laser beam ensures discreet monitoring. The system can operate continuously in various weather conditions and is scalable for large border areas. Integration with IoT allows timely alarms and remote monitoring thereby reducing manpower and enhancing security effectiveness. An additional layer of functionality is added through the GSM module, which allows for remote communication and control. The GSM module enables the system to send SMS notifications to predefined mobile numbers, informing users about door access events, unauthorized attempts, or system status updates. Moreover, it allows users to interact with the system remotely by sending specific SMS or GPRS commands to lock or unlock the door from a distance, adding a high level of convenience and control.
Abstract
A Comparative Analysis of DTW Variants for Kinect-Based Motor Skill Assessment in Children with Developmental Coordination Disorder (DCD)
Jackie Ting Tiew Wei, Ting Huong Yong, Tiong Meng Chung
DOI: 10.17148/IJIREEICE.2025.131142
Abstract: Developmental Coordination Disorder (DCD) assessment traditionally relies on subjective, time-consuming tests. This paper proposes an objective, Kinect-based framework to analyze motor skills in children with DCD. Building on our previous work using a lossless Dominant Range of Movement Index (D-RoMI) representation, this study conducts a comparative analysis of three advanced Dynamic Time Warping (DTW) algorithms: Constrained DTW (cDTW), Derivative DTW (DDTW), and Soft-DTW. We evaluated their ability to classify motor patterns from DCD and typically developing children against "gold-standard" templates. Our findings show that DDTW, which compares movement velocity rather than position, achieves the highest classification accuracy. This suggests that analyzing movement dynamics is more informative for identifying DCD's characteristic motor patterns than spatial analysis alone. This framework offers a more reliable and efficient pathway for DCD assessment.
Keywords: Developmental Coordination Disorder (DCD), Derivative DTW (DDTW), Motion Analysis, Kinect.
Keywords: Developmental Coordination Disorder (DCD), Derivative DTW (DDTW), Motion Analysis, Kinect.
Abstract
Physico-Chemical Characterization of Soil in the Vicinity of Banki Dam and Ghaghi Waterfall, Surguja District, Chhattisgarh, India
Mr. Govind Prasad, Dr. M.K. Maurya, Mr. Pramod Tigga, Miss Deepa Singh, Miss Khileshwari Rajwade
DOI: 10.17148/IJIREEICE.2025.131143
Abstract: This study presents a detailed physico-chemical characterization of soils collected from the vicinity of Banki Dam and Ghaghi Waterfall in the Surguja district of Chhattisgarh, India, with the aim of assessing their agricultural potential and fertility status. Soil samples from multiple locations were analyzed for key parameters including texture (sand, silt, clay), pH, electrical conductivity (EC), organic carbon content, major nutrients (N, P, K), micronutrients, cation exchange capacity (CEC), and bulk density. The results revealed notable variations in soil texture and structure, affecting water retention and drainage properties. Soil pH was found to range from slightly acidic to neutral, generally favorable for a wide range of crops, while EC values indicated low to moderate salinity in certain areas. Organic carbon levels were sufficient to maintain soil fertility, though some deficiencies in nitrogen and phosphorus were observed. The findings underscore the need for site-specific soil management practices, including targeted fertilizer application and organic amendments, to optimize crop productivity and maintain soil health. These insights provide valuable guidance for farmers, researchers, and land managers to support sustainable agriculture and long-term soil conservation in the region.
Keywords: Soil texture, pH, electrical conductivity (EC), organic carbon, nutrient status, Banki Dam, Ghaghi Waterfall, Surguja district, soil fertility, sustainable agriculture.
Keywords: Soil texture, pH, electrical conductivity (EC), organic carbon, nutrient status, Banki Dam, Ghaghi Waterfall, Surguja district, soil fertility, sustainable agriculture.
Abstract
MQTT BASED COAL MINE MONITORING SYSTEM USING – IOT
K.Amarender, V Srikanth, P Pavitra, B Supriya, M Navya
DOI: 10.17148/IJIREEICE.2025.131144
Abstract: The coal mining industry faces numerous safety and operational challenges, including hazardous environmental conditions, equipment failures, and worker safety. To address these issues, a monitoring system based on the Internet of Things (IoT) and MQTT (Message Queuing Telemetry Transport) is proposed for real-time data collection and analysis in coal mines. The system integrates various sensors for monitoring environmental parameters such as gas levels (e.g., methane), temperature, humidity, dust, and equipment performance. The IoT network collects sensor data and transmits it via MQTT, a lightweight and efficient messaging protocol, to a centralized cloud-based system for real-time analysis. The use of MQTT ensures reliable communication in low-bandwidth environments typical of coal mines. Alerts and notifications are generated if critical thresholds are exceeded, enabling proactive safety measures and equipment maintenance. This system aims to improve the safety of miners, reduce operational downtime, and optimize resource management in coal mining operations. Additionally, the proposed solution facilitates data-driven decision- making by providing mine operators with real-time insights and predictive analytics.
Abstract
Integrating Flow Theory in Designing Educational Virtual Worlds in the Metaverse
Dimitrios Magetos, Prof. Sarandis Mitropoulos, Prof. Christos Douligeris
DOI: 10.17148/IJIREEICE.2025.131145
Abstract: The paper presents a study that explores the interconnection of Flow Theory with the acceptance of educational virtual worlds at the Metaverse, making use of the Unified Acceptance and Use of Technology (UTAUT2) model. The study relied on synthetic data from 120 high school students, which were generated by artificial intelligence with the purpose of testing a theoretical framework without collecting empirical data. The model under consideration measures the interconnection of flow experience with the acceptance factors of performance expectancy, effort, social influence, facilitating conditions, enjoyment, habit, and intention to use. The findings reveal that flow experience acts as a mediator that increases cognitive engagement, intention to continue using, and acceptance of the Metaverse as an educational setting. The study concludes with design guidelines that connect flow theory with the principles of experiential and affective learning in the Metaverse.
Keywords: Flow Theory, UTAUT2, Metaverse, Virtual Worlds, VR/AR.
Keywords: Flow Theory, UTAUT2, Metaverse, Virtual Worlds, VR/AR.
Abstract
Anti-sleep Safety System for the Vehicle Drivers
Tapan Kumar Mohanta, Jaykrushna Mohanta, Dhriti Rudrapal
DOI: 10.17148/IJIREEICE.2025.131146
Abstract: To improve road safety by lowering accidents brought on by driver drowsiness, this initiative introduces an anti-sleep alert for drivers. The gadget uses real-time monitoring to detect early signs of fatigue and promptly alerts the driver using a combination of sound and vibration signals. Advanced sensors are used by the anti-sleep alert system to track behavioural and physical signs of driver drowsiness, including head movement, blink rate, and facial expressions. The system evaluates these factors in real-time using machine learning algorithms to accurately identify tiredness. The alarm instantly sounds when it detects indicators of weariness, keeping the driver awake before crucial concentration lapses take place.
Keywords: anti-sleep alert, blink rate, driver drowsiness, blink rate, drivers.
Keywords: anti-sleep alert, blink rate, driver drowsiness, blink rate, drivers.
Abstract
IDENTIFICATION AND PREVENTION OF ACCIDENTS USING SMART HELMET AND GPS SYSTEM
P Thirupathi, A Aishwarya, J Vaishnavi, B Navya, P Sheshu Kumar
DOI: 10.17148/IJIREEICE.2025.131147
Abstract: In each one hour 17 people are dying in India because of street mishaps. According to the administration report practically 1.5 lacks people are passed on street mishaps in 2017 Most of mishaps is bikes in view of person not worn helmet and consuming alcohol. So in this proposed framework if the rider isn't worn helmet or consumes any alcoholic substance is distinguished, the bike will not start. In addition, it has a smart feature to identifying accidents and sends SMS to rescue vehicle, police headquarters and family members with location by using GSM and GPS module, thus aiding ambulance to reach the correct location. We want to integrate all the sensors inside the helmet, which will send the all the data to the receiver connected in motorcycle wirelessly. This brilliant head. protector framework comprises of two modules, one is head protector(transmitter) and another one is bike (receiver) Alcohol sensor, IR sensor and ultrasonic sensor are associated inside the helmet unit and vibration sensor, GPS and GSM are connected in vehicle unit. The transmitting and receiving unit communicate wirelessly using RF transmitter and receiver, using Arduino uno.
Abstract
AI-Based Fraud Detection Using Deep Learning on Transaction Data
Abdul Hasham, Mubashir Ali Ahmed
DOI: 10.17148/IJIREEICE.2025.131148
Abstract: Since fraud is now much more common and sophisticated due to digital payments and online financial services, financial institutions need to be able to identify it promptly and accurately. In the context of real-world transaction streams, standard rule-based and classical machine learning techniques typically encounter difficulties with complicated temporal linkages, unbalanced transaction data, and fraud patterns that change over time. This paper provides a deep learning-based fraud detection system that models transactional activity and detects anomalous behaviors with high recall and precision in order to overcome these limitations.
The suggested method integrates representation learning and deep sequence learning to find both short-term and long- term patterns in transaction data. To comprehend how user activities correlate over time, recurrent and attention-based neural architectures are used to simulate transaction sequences. Compact representations of valid transactions are learned by an autoencoder-based anomaly detection module, which then identifies any deviations that point to fraud. Reconstruction mistakes, engineering behavioral data, and supervised classification scores are combined in a fusion layer to generate a high fraud risk score for every transaction. When there are numerous classes that differ significantly from one another, this hybrid design facilitates the detection of fraud as well as the discovery of novel fraud tactics.
Using benchmark transaction datasets that replicated real attacks, we tested the system in fake fraud situations. Based on the F1-score and the area under the precision-recall curve, experimental results demonstrate that the suggested model consistently performs better than both stand-alone deep learning models and conventional machine learning baselines. Additionally, it may be deployed immediately due to its short inference latency. By elucidating concepts and supporting decision-making, feature attribution methodologies can facilitate model comprehension. The findings demonstrate the efficacy of deep learning-based fraud detection systems and provide crucial criteria for developing scalable, accurate, and reliable AI solutions in financial contexts.
Keywords: Financial transaction analysis, deep learning, class imbalance, sequence modelling, autoencoders, explainable artificial intelligence (XAI), financial cybersecurity, anomaly detection, fraud detection, and real-time fraud analytics.
The suggested method integrates representation learning and deep sequence learning to find both short-term and long- term patterns in transaction data. To comprehend how user activities correlate over time, recurrent and attention-based neural architectures are used to simulate transaction sequences. Compact representations of valid transactions are learned by an autoencoder-based anomaly detection module, which then identifies any deviations that point to fraud. Reconstruction mistakes, engineering behavioral data, and supervised classification scores are combined in a fusion layer to generate a high fraud risk score for every transaction. When there are numerous classes that differ significantly from one another, this hybrid design facilitates the detection of fraud as well as the discovery of novel fraud tactics.
Using benchmark transaction datasets that replicated real attacks, we tested the system in fake fraud situations. Based on the F1-score and the area under the precision-recall curve, experimental results demonstrate that the suggested model consistently performs better than both stand-alone deep learning models and conventional machine learning baselines. Additionally, it may be deployed immediately due to its short inference latency. By elucidating concepts and supporting decision-making, feature attribution methodologies can facilitate model comprehension. The findings demonstrate the efficacy of deep learning-based fraud detection systems and provide crucial criteria for developing scalable, accurate, and reliable AI solutions in financial contexts.
Keywords: Financial transaction analysis, deep learning, class imbalance, sequence modelling, autoencoders, explainable artificial intelligence (XAI), financial cybersecurity, anomaly detection, fraud detection, and real-time fraud analytics.