← Back to Archives
VOLUME 14, ISSUE 4, APRIL 2026
AI-Based Multi-Cloud Autoscaler with Pricing and Risk-Aware Resource Optimization
Shaik Aqheel Pasha, Mohammed Imran Ahmed
Online Car Rental System
Tanmay Khogare, Krishna Biradar, Radhesh Munde, Ashok Khedakar, Ms. P. B. Borate
AN AUTOMATED SYSTEM FOR TIME SERIES DATA ANALYSIS USING CORRELATION MEASURES
Swathi S, Janarthanan S
DEEP LEARNING-BASED CLASSROOM ANOMALY DETECTION USING OBJECT-CENTRIC TEMPORAL MODELING
Anusree M, Revathi A
Smart Vision Inventory Advisor: An Integrated OCR-Based Intelligent Inventory Management Framework with Machine Learning Demand Forecasting for Small Retail Enterprises
Mohamed Athfan D, Maria Lavanya P, Selva Pujith T
Deep Neural Network Enhanced Compressive Sensing and Kalman Filtering for 5G Channel Estimation
Dr. G. Krishna Reddy, L. Akhila, S. Ankitha, M. Kavitha, K. Chandrika Tejaswini
A Review of AI-Driven Medical Assistant System and Intelligent Chatbots Platforms for Modern Healthcare
Chaithanya reddy, Rishindra Chowdary, Saikumar, Siddhardha
Eye Disease Detection using Convolutional Neural Network
REVATHI A, POOBESH KS, SRI BALAJI T
Load Flow Analysis of Grid Connected Wind Farm
Dr. N. Narmadhai, J. Lavanya
Interpretable vs Predictive: Comparing Statistical and ML Models of Walmart Black Friday Demographics in Consumer Behaviour Analysis
Harish I, Arun Kumar K
RESEARCH ON SMART PUBLIC ANNOUNCEMENT AERIAL DRONE WITH PERSON DETECTION
Ms. Tanaya P. Dudhe, Mr. Abhay B. Rathod, Ms. Ishwari P. Garode, Ms. Radhika R. Jaiswal, Mr. Premraj P. Chavhan, Ms. Shrushti P. Botule, Ms. Anisha A. Daf, Mr. Zishan A. Khan, Mr. Sahil S. Sonarkhan
AI-Powered Cloud Removal from Satellite Imagery
Manjula Devi K, Roohitha NS, C.Chithra Devi M.Sc., (Ph.D),
AI Driven Market Price Forecasting and Risk Analysis For Agricultural Commodities
Kalaiamuthan N, M. Saravanakumar
Smart Attendance System Using Deep Learning-Based Facial Recognition
MALA BHARUMATHI M, UDHAYA KUMAR R
E-Vehicle Recharge Hub: A Smart Location-Based Charging Solution
Saravana Kumar M, Vasanika S, Priya Dharshini M
SAFETRACK: Real-Time IoT-Based Vehicle Accident Detection, Severity Analysis and Emergency Alert System
Pavan Gharat, Swapnil Kadam, Gaurishankar Awale, Mr. U. S. Shirshetti
AN AI FRAMEWORK FOR UNSTRUCTURED DATA QUALITY ASSESSMENT WITH INTEGRATED PII DETECTION
Vanitha A, Jumana J
REAL-TIME TOMATO LEAF DISEASE CLASSIFICATION AND TREATMENT RECOMMENDATION USING DEEP LEARNING TECHNIQUES
Arockia Ajay J, Arun Kumar K
A Q-Learning Based Dynamic Pricing Model for Minimizing Food Waste in Supermarkets
D. Vimal Kumar, A. Hemalatha, S. Harish, M. Mohammed Yunush
DETECTION OF STAMMERING IN SPEECH USING MACHINE LEARNING WITH HUMAN EVALUATION
ARAVINDAGOKUL. P, ARUN KUMAR K
DATA-DRIVEN MANUFACTURING QUALITY TRACKING AND VISUALIZATION PLATFORM
Arockiya Anjugan M L, A S Krishna
Machine Learning-Based Predictive Analytics for Sustainable Resource Consumption Forecasting: A Comparative Study
Dr. D. VIMAL KUMAR, SHARMILA S, LAKSHYA SHREE R, HARISH I
AI-Powered Digital Marketing: Transforming the Future of Customer Engagement A Comprehensive Study on Artificial Intelligence Applications in Modern Marketing
Madhulekha E, Ms. A. Hemalatha
Medical Insurance Price Prediction Using Machine Learning
Chithra Devi C M.Sc (Ph.D), Abdul Rasheed M, Pravin Kumar V
Using Machine Learning and Natural Language Processing, A System Can Find Spam Emails
Ayan Husain, Khan Amir Alam, Khan Abis, Abdul Gaffar, Prof. Imran Shahid
AI-Powered Real-Time Firearm Detection and Alert System Using YOLOv12m Deep Learning Model
Savita Waghmode, Prof. Vidhate S.N.
Threat Intelligence System for Cyber Attacks
Kunal Naikade, Ritesh Patni, Prathmesh Jaiswal, Smita Chunamari
IOT BASED SMART HOME AUTOMATION SYSTEM
Mrs. Suryasevak Singh, Rafay Siddiqui, Janhvi Patil, Bhavya Chavri, Irfan Sarkar, Balwant Yedage
Comparative Analysis and Cross-Dataset Validation of Traditional Machine Learning Models for Fake News Detection
Pranjal L. Bhamre, Isha Y. Jain, Sunita N. Deore
The Effectiveness of CTV Advertising on Brand Equity and Consumer Engagement
Eshika Makwana, Hetang Mahajan and Hiren Harsora
Customer Churn Prediction In Subscription-Based Platforms Using Machine Learning
YUVA DHARSHINI M, MALA BHARUMATHI M
Career Guidance System For Personalised Career Pathways
Arti Jaibhai, Rushikesh Bembale, Arjun Gade, Jeevan Maske, Varun Baporikar
An AI-Powered Conversational System for Rapid Digital Forensic Analysis
Praveen R, Krishna AS
ELECTRICAL VEHICLE CHARGING WITH INTRELEAVED CONVERTER
Shruti M. Kavare, Pranav T. Chougule, Sakshi K. Powar, Pratiksha S. Rokade, Ms. Shubhangi Manjare
Training & Development- A Study on the Role of Training and Development in Enhancing Employee Performance and Organizational Growth
Lakshya Kothari, Avesh Khanjada, Dr. Sapana Chauhan
AI-DRIVEN DATA LIFECYCLE OPTIMIZATION SYSTEM
V TEJASREE, MANEESHA P A
Machine Learning in Healthcare: A Comprehensive Review
Shruti Gosavi, Pooja Gunjal, Sunita N. Deore
Animal Detection in Farms Using OpenCV
Sharmila S, M.Saravanakumar
AIR QUALITY DATA ANALYSIS AND DYNAMIC VISUALIZATION
Roshan Zameer M, Ms. Hemalatha A
AgroConsultant: Optimizing Fertilizer Recommendations in Precision Agriculture
S SRINITHI, M SARAVANAKUMAR
Intelligent Signal Processing for 5G/6G OFDM Systems Using Deep Learning
V.Amulya, V.Jyothsna, G.Yukthavi, P.Nakshathra
PREDICTIVE ANALYSIS MODEL FOR EARLY DETECTION OF ALZHEIMER’S DISEASE USING CLINICAL AND COGNITIVE DATA
Mr.A.Naga Srinivasa Rao¹,T.Sai Meghana²,A.Sumadhuravani³, P.Padma Gandha †, V.Venkata Kalyaniµ
Design And Development of Thermoelectric Generator For Waste Heat Energy Harvesting
Lokesh Ramawadh Chauhan, Nainiksha Anil Dhoke, Riya Manoj Kalamkar, Divya Shailendra Patil, Kashish Manoj Mahato, Prof. Rajendra Bhombe, Prof. Priyanka Rajput
An Advanced Explainable Deep Learning Approach with Grad-CAM and Post-Hoc Analysis for Secure QR Code Threat Detection
Y Roshni and Dr. Golda Dilip
CONFIDENCE-GUIDED MULTI-AGENT LLM FRAMEWORK FOR CLINICAL DECISION SUPPORT IN OPHTHALMOLOGY USING BIOMEDICAL RAG AND WEB INTELLIGENCE
Mrs.D. Archana¹, P. Tanmai Sai², S. Bhavana³, Y. Suhitha Priya´, G. Siri Saranyaµ
Barriers Against the Use of EV-Review
Prof.Akash A. Gophane, Sampada V. Chaure, Yashasvi G. Rohane, Kuldeepak B. Pise
SOLAR ENERGY MONITORING SYSTEM
Ankam Varshini, Ganta Praveen Sai, Goli Harika Lakshmi,Kunchanapalli Sri Anjhanee, Vasamsetti Lakshmi Sri Sai Devika, Dr. M. Srinivasa Rao
Perceived Role of AI-Based tools in enhancing employability and Job Readiness among MBA students in Bangalore.
Badri Narayanan KS, Dr. Mahalakshmi S
Dynamic Voltage Regulation using Solar Powered Synchro-Inverter
Chitra S, Archana Prasadini G S
SRMunch: A Data-Driven Institutional Food Operational Efficiency System
Shreya Sathapathi, Rakshitha S N, Yashvitha J, Dr. Durgadevi P
SENTINEL: Multimodal AI Framework for Contract Risk Analysis and Negotiation Strategy Generation
S Yashwant, Surya Sivakumar, J Joshua Haniel, Niranjana S
Abstract
AI-Based Multi-Cloud Autoscaler with Pricing and Risk-Aware Resource Optimization
Shaik Aqheel Pasha, Mohammed Imran Ahmed
DOI: 10.17148/IJIREEICE.2026.14401
Abstract: Finding the right balance between cloud costs and application performance has become much more challenging due to the quick adoption of multi-cloud architectures. The majority of autoscaling solutions are rule-driven, reactive, and intended for single-cloud scenarios. This restricts their applicability in intricate multi-cloud settings with disparate pricing schemes, data transfer fees, provisioning delays, and spot/preemptible instances. In order to reduce costs while maintaining service-level objective (SLO) limitations, this article presents MCCAS (Multi-Cloud Cost-Aware Autoscaler), an AI-driven autoscaling platform that intelligently optimizes resource allocation across several cloud providers. Three interrelated parts make up MCCAS: (i) a workload forecasting module that uses deep learning to predict short-term demand and uncertainty; (ii) a thorough cost and risk modelling layer that takes provider-specific pricing, migration overheads, and pre-emption risks into account; and (iii) a hierarchical reinforcement learning (HRL) decision engine that makes a distinction between short-term tactical scaling actions and long-term strategic placement decisions. MCCAS uses hierarchical control to make the decision space simpler. This makes it simple to adapt to changes in pricing and workload. With rewards that explicitly penalize expenses, SLO violations, and inefficient migrations, the suggested method sees autoscaling as a sequential decision-making issue with limitations. Using real workload traces and simulated multi-cloud pricing scenarios, a comprehensive experimental design is shown along with comparisons to rule-based, prediction-only, and single-cloud learning-based autoscalers. The outcomes should show that cost-conscious, AI-driven multi-cloud autoscaling may drastically save operating costs without sacrificing application performance guarantees. This study offers researchers, cloud architects, and FinOps teams a practical and repeatable method for implementing intelligent autoscaling solutions in actual multi-cloud systems.
Keywords: AI-powered autoscaling, resource provisioning, cloud economics, workload forecasting, finOps, multi-cloud computing, cloud cost optimization, reinforcement learning, and hierarchical reinforcement learning.
Keywords: AI-powered autoscaling, resource provisioning, cloud economics, workload forecasting, finOps, multi-cloud computing, cloud cost optimization, reinforcement learning, and hierarchical reinforcement learning.
Abstract
Online Car Rental System
Tanmay Khogare, Krishna Biradar, Radhesh Munde, Ashok Khedakar, Ms. P. B. Borate
DOI: 10.17148/IJIREEICE.2026.14402
Abstract: Currently, the people finds a car to rent via social network and make some call to car’s owner for a rent. To offer the benefits for both owner and rental, the hire car system was developed. This project was done to beat the matter of individuals to seek out a car to be easier and for the car’s owner they will manage booking made by rental through this technique. This technique includes three modules which are of rental, car’s owner and administrator. The title of this project is named “Online Car Rental System”. Car rental system primarily serve people who require a temporary vehicle, for example, those who do not own their own car, travellers who are out of town, or owners of damaged or destroyed vehicles who are awaiting repair or insurances compensation. Car rental agencies may also serve the self- moving industry needs, by renting vans or trucks or bikes, and in certain markets, other types of vehicles such as motorcycles or scooters may also be offered. The aim of the project is to create an automatic system for reserving Vehicles online and simply manage rental services by using the online based system. The project aims to create a web hire car system. We used five stages development life cycle including planning, analysis, design, implementation and use, which utilized programing language of PHP and MYSQL database.
Keywords: Online hire car, Luxury hire car, car rental Online, Car Near Me
Keywords: Online hire car, Luxury hire car, car rental Online, Car Near Me
Abstract
AN AUTOMATED SYSTEM FOR TIME SERIES DATA ANALYSIS USING CORRELATION MEASURES
Swathi S, Janarthanan S
DOI: 10.17148/IJIREEICE.2026.14403
Abstract: Looking at data over time helps spot trends and how things connect. Because machines produce tons of time- based records, quick ways to examine them matter a lot. Built into this work sits a tool that automatically studies such sequences through correlation math. Instead of skipping gaps, it fills holes in inputs before moving forward. Normalization adjusts scales so comparisons stay fair across different sources. Noise gets filtered out carefully to keep results clear and meaningful. Each step prepares the ground for trustworthy outcomes without extra effort later on. To check how closely two sets of data match and follow each other over time, Pearson and cross-correlation methods come into play. Instead of one fixed view, a moving window tracks how connections shift through different moments. Seeing the data unfold in charts makes it easier to spot repeating shapes or unusual gaps. Tests run on both artificial examples and actual recordings show the method catches shared movements, delays, and odd behaviors well. Because of this, people spend less time digging through signals by hand while getting more consistent outcomes.
Keywords: Time Series Analysis, Correlation Measures, Pearson Correlation, Sliding Window, Data Visualization, Python.
Keywords: Time Series Analysis, Correlation Measures, Pearson Correlation, Sliding Window, Data Visualization, Python.
Abstract
DEEP LEARNING-BASED CLASSROOM ANOMALY DETECTION USING OBJECT-CENTRIC TEMPORAL MODELING
Anusree M, Revathi A
DOI: 10.17148/IJIREEICE.2026.14404
Abstract: Detecting the unusual patterns in video footage plays a crucial role in ensuring safety by alerting authorities to potential risks. In many real-world scenarios, delayed identification of abnormal events such as fights or suspicious behavior can lead to serious consequences. This project focuses on developing an automated video anomaly detection system using deep learning and computer vision techniques. The system processes video footage by extracting frames and detecting foreground objects using the YOLOv8 algorithm. Relevant spatial features are extracted using a ResNet-based convolutional neural network, and temporal patterns are learned using a Bidirectional Long Short-Term Memory network. The extracted features are analyzed to identify abnormal activities using statistical methods such as Z-Score-based anomaly scoring. When an abnormal event is detected, the system generates real-time alerts by displaying warning messages, triggering an alarm sound, and sending an email notification to the user. The proposed system is implemented using Python and integrated with a Streamlit-based web interface for visualization. This approach improves real-time montoring, reduces response time, and enhances the effectiveness of surveillance systems.
Keywords: Video Anomaly Detection, Deep Learning, YOLOv8, BiLSTM, Temporal Analysis
Keywords: Video Anomaly Detection, Deep Learning, YOLOv8, BiLSTM, Temporal Analysis
Abstract
Smart Vision Inventory Advisor: An Integrated OCR-Based Intelligent Inventory Management Framework with Machine Learning Demand Forecasting for Small Retail Enterprises
Mohamed Athfan D, Maria Lavanya P, Selva Pujith T
DOI: 10.17148/IJIREEICE.2026.14405
Abstract: Effective inventory management remains a persistent operational challenge for small and medium-sized retail enterprises, many of which continue to depend on manual record-keeping and intuitive judgment rather than data-driven methodologies. These practices yield elevated error rates, inadequate demand forecasting, and reactive rather than proactive restocking decisions, culminating in stock shortages, excess inventory accumulation, and avoidable financial losses. This paper presents the Smart Vision Inventory Advisor, a unified intelligent inventory management framework integrating Optical Character Recognition (OCR)-based product recognition with fuzzy matching algorithms for high- accuracy product identification. The system employs Tesseract OCR for text extraction from product images, followed by a fuzzy matching approach using difflib's get_close_matches and a custom keyword-based scoring mechanism that achieves 90–95% recognition accuracy. The system incorporates Multiple Linear Regression (MLR) for demand forecasting, leveraging features such as day-of-week, month, weather conditions, festival indicators, and discount information. Empirical evaluation demonstrated product recognition accuracy of 91.4%, a 77.1% reduction in cataloging time, and forecasting MAPE of 16.2%, representing a 33.9% improvement over moving-average baselines. Implemented exclusively using open-source technologies on standard consumer hardware, the system demonstrates that advanced AI- driven inventory management is both economically feasible and practically beneficial for the small retail sector.
Keywords: Optical Character Recognition; Inventory Management; Demand Forecasting; Multiple Linear Regression; Fuzzy Matching; Predictive Analytics; Retail Intelligence; Gradio
Keywords: Optical Character Recognition; Inventory Management; Demand Forecasting; Multiple Linear Regression; Fuzzy Matching; Predictive Analytics; Retail Intelligence; Gradio
Abstract
Deep Neural Network Enhanced Compressive Sensing and Kalman Filtering for 5G Channel Estimation
Dr. G. Krishna Reddy, L. Akhila, S. Ankitha, M. Kavitha, K. Chandrika Tejaswini
DOI: 10.17148/IJIREEICE.2026.14406
Abstract: Massive MIMO systems operating at mmWave frequencies (mmWave) constitute one of the applications of 5G communication, which provides high data rate and spectral efficiency. Nevertheless, the channel estimation and tracking is still a problem because of high channel variations and sparse propagation. In this paper, I am going to suggest a joint architecture that would combine Compressive Sensing (CS), Long Short-Term Memory (LSTM) and Kalman Filtering (KF) to effectively estimate channels. CS is employed in order to recover sparsely-spread channel information with a reduced number of pilot signals. The LSTM corrects the errors and removes noise in the estimated channel. KF monitors the change in channel across the time, and maintains the accuracy. The given approach is more precise in estimations, decreases the rate of errors, and has a higher spectral efficiency than the traditional approaches.
Keywords: 5G Communication, mmWave Massive MIMO, Compressive Sensing, Deep Neural Network, Kalman Filter, Channel Estimation
Keywords: 5G Communication, mmWave Massive MIMO, Compressive Sensing, Deep Neural Network, Kalman Filter, Channel Estimation
Abstract
A Review of AI-Driven Medical Assistant System and Intelligent Chatbots Platforms for Modern Healthcare
Chaithanya reddy, Rishindra Chowdary, Saikumar, Siddhardha
DOI: 10.17148/IJIREEICE.2026.14407
Abstract: Healthcare is not made to work in a world where people look for help online before they talk to a doctor. It used to be simple with hospital websites and places to make appointments. Now it is complex with artificial intelligence helping to figure out what is wrong with people patients wanting to get information right away and healthcare providers having to manage appointments and keep patient information safe. As these systems get bigger it becomes a deal to make sure, they are reliable give the right medical information keep data private and are secure. In the ten years people have made many ideas for medical assistants that use artificial intelligence, systems that can figure out what is wrong with people and platforms that can schedule appointments. These systems can look at symptoms give people some idea of what might be wrong and help them figure out what condition they might have. At the time systems that let people make doctor appointments online have made healthcare better by reducing wait times letting people make appointments online and making it easier for patients and doctors to talk.
This review looks at assistant systems that use artificial intelligence and are connected to platforms that let people make doctor appointments. It looks at how these systems work what they are made of and how they keep data safe. By looking at what people have done and what technology is available this study shows what these systems can do now what they cannot do and where they are going to improve how patients are helped and how healthcare is made available, to people.
Keywords: AI Medical Assistant; Intelligent Healthcare Systems; Medical Decision Support Systems; Medical Chatbots; Healthcare Data Privacy; AI-Driven Healthcare Services
This review looks at assistant systems that use artificial intelligence and are connected to platforms that let people make doctor appointments. It looks at how these systems work what they are made of and how they keep data safe. By looking at what people have done and what technology is available this study shows what these systems can do now what they cannot do and where they are going to improve how patients are helped and how healthcare is made available, to people.
Keywords: AI Medical Assistant; Intelligent Healthcare Systems; Medical Decision Support Systems; Medical Chatbots; Healthcare Data Privacy; AI-Driven Healthcare Services
Abstract
Eye Disease Detection using Convolutional Neural Network
REVATHI A, POOBESH KS, SRI BALAJI T
DOI: 10.17148/IJIREEICE.2026.14408
Abstract: Diabetic Retinopathy (DR) is one of the leading causes of vision impairment worldwide and is directly associated with prolonged diabetes. The disease progresses silently and often remains undetected in its early stages, which makes timely diagnosis critical. Early identification of DR can significantly reduce the risk of severe vision loss and improve patient outcomes. This research proposes an automated deep learning-based framework for detecting and classifying Diabetic Retinopathy using retinal fundus images. The system leverages Transfer Learning on pre- trained Convolutional Neural Network architectures including VGG16, ResNet50 V2, and EfficientNet B0. These models are fine-tuned to extract meaningful features from retinal images and classify them into different severity levels. Extensive experiments are conducted to evaluate the performance of the models using metrics such as accuracy, precision, recall, and F1-score. Visualization techniques including confusion matrix and ROC curves are also considered to analyze model behavior. The results indicate that VGG16 achieves superior performance in comparison to other models. The findings demon- strate that Transfer Learning can be effectively used to develop reliable and efficient diagnostic systems for medical imaging applications.
Keywords: Digital Eye Strain, Ocular Micro-Motion, Cognitive Load, Multi-Device Monitoring, Computer Vision.
Keywords: Digital Eye Strain, Ocular Micro-Motion, Cognitive Load, Multi-Device Monitoring, Computer Vision.
Abstract
Load Flow Analysis of Grid Connected Wind Farm
Dr. N. Narmadhai, J. Lavanya
DOI: 10.17148/IJIREEICE.2026.14409
Abstract: Nowadays, the need for electrical energy is increasing day by day, which has made renewable energy sources more important. Among them, wind energy is widely used because it is clean and easily available. In this work, a grid- connected wind farm system is considered and its load flow behaviour is studied. The main aim is to observe how power is distributed in the system and to check voltage levels and losses under normal conditions. The modelling and simulation are carried out using Dig SILENT Power Factory software. The Newton-Raphson method is selected for analysis since it gives faster and reliable results. From the simulation, it can be observed that the system operates within acceptable limits. This kind of study helps in understanding the performance of wind farms and supports better integration with the power grid.
Keywords: Wind Farm, Load Flow Analysis, Power Factory, Newton-Raphson Method, Renewable Energy.
Keywords: Wind Farm, Load Flow Analysis, Power Factory, Newton-Raphson Method, Renewable Energy.
Abstract
Interpretable vs Predictive: Comparing Statistical and ML Models of Walmart Black Friday Demographics in Consumer Behaviour Analysis
Harish I, Arun Kumar K
DOI: 10.17148/IJIREEICE.2026.14410
Abstract: Walmart’s Black Friday, falling the day after Thanksgiving in the U.S., marks the launch of holiday shopping and drives massive retail sales nationwide. Retailers face the challenge of balancing interpretable statistical insights with predictive machine learning to analyze consumer behavior. This research proposes a unified framework that emphasizes clarity, actionable insights, and predictive strength to guide retail decision-making. By integrating statistical methods with machine learning, the study enhances both retail insights and forecasting power. Looking ahead, retail analytics is poised to evolve through deep learning, real-time processing, and scalable data platforms, enabling faster adaptation to shifting consumer trends and competitive dynamics.
Keywords: Black Friday - Consumer Behavior - Machine Learning - statistics -Unified Framework
Keywords: Black Friday - Consumer Behavior - Machine Learning - statistics -Unified Framework
Abstract
RESEARCH ON SMART PUBLIC ANNOUNCEMENT AERIAL DRONE WITH PERSON DETECTION
Ms. Tanaya P. Dudhe, Mr. Abhay B. Rathod, Ms. Ishwari P. Garode, Ms. Radhika R. Jaiswal, Mr. Premraj P. Chavhan, Ms. Shrushti P. Botule, Ms. Anisha A. Daf, Mr. Zishan A. Khan, Mr. Sahil S. Sonarkhan
DOI: 10.17148/IJIREEICE.2026.14411
Abstract: Flood disasters often result in large-scale human displacement, making timely detection and rescue operations critical. Conventional ground-based approaches for locating stranded individuals are often time-consuming, labor- intensive, and limited by accessibility challenges. To overcome this limitation, the project combines an ESP32-CAM module with a drone system to provide lightweight aerial monitoring and real-time detection of individuals during flood emergencies. By deploying a lightweight version of the YOLO (You Only Look Once) object detection model, the ESP32-CAM processes live video streams and identifies individuals from a top-view perspective. The captured feed is transmitted via Wi-Fi and monitored through a laptop browser, providing rescuers with an efficient tool for rapid assessment. The system can detect people entering or exiting the frame and automatically count the number of individuals, offering valuable situational awareness. Additionally, communication between operators and the drone is supported via a walkie-talkie system to enhance coordination during rescue missions. This solution emphasizes portability, affordability, and real-time processing, making it suitable for disaster-prone regions with limited infrastructure. In summary, the proposed system seeks to connect drone-based monitoring with AI-powered disaster management by delivering a dependable, cost-effective, and fast-response solution for flood rescue operations.
Keywords: ESP32-CAM, YOLO, Flood Disaster, Drone, Person Detection, People Counting, Aerial Surveillance, Real- Time Monitoring.
Keywords: ESP32-CAM, YOLO, Flood Disaster, Drone, Person Detection, People Counting, Aerial Surveillance, Real- Time Monitoring.
Abstract
AI-Powered Cloud Removal from Satellite Imagery
Manjula Devi K, Roohitha NS, C.Chithra Devi M.Sc., (Ph.D),
DOI: 10.17148/IJIREEICE.2026.14412
Abstract: Satellite images have kind of become part of everyday decision-making now. Farmers look at them to check crop health, governments use them during floods, and urban planners rely on them to see how cities are expanding.They’re powerful tools, no doubt. But let’s be real — they’re not perfect. One issue that keeps getting in the way? Clouds. Even a thin layer can block important surface details, and once that happens, the image isn’t nearly as useful as it should be. You’re left guessing what’s underneath, which defeats the whole purpose That’s what pushed me toward this project. The goal wasn’t to reinvent satellite imaging or anything dramatic. It was more practical than that: create a web-based system that can reduce cloud interference using AI. The idea is straightforward. A user uploads a satellite image through a simple web interface, and the system takes care of the rest.
The image is sent to the backend, where an AI-based process analyzes the cloudy regions and tries to reduce their impact without disturbing the actual land features I wanted the system to feel usable, not complicated. No heavy software installation, no confusing steps. Just upload, process, compare, and download.
The before-and-after comparison is actually one of my favorite parts because you can immediately see what changed. It makes the improvement feel real, not theoretical.At the end of the day, the purpose is pretty clear: make satellite images easier to work with. If analysts, researchers, or planners can see the ground more clearly, they can make better decisions. Sometimes, solving something as simple as cloud obstruction can quietly improve a lot of larger processes behind the scenes.
Keywords: satellite image processing, Cloud Removal, Web-based Application, Cloud Detection, Image enhancement, AI - Based Image Processing, React Web Application
The image is sent to the backend, where an AI-based process analyzes the cloudy regions and tries to reduce their impact without disturbing the actual land features I wanted the system to feel usable, not complicated. No heavy software installation, no confusing steps. Just upload, process, compare, and download.
The before-and-after comparison is actually one of my favorite parts because you can immediately see what changed. It makes the improvement feel real, not theoretical.At the end of the day, the purpose is pretty clear: make satellite images easier to work with. If analysts, researchers, or planners can see the ground more clearly, they can make better decisions. Sometimes, solving something as simple as cloud obstruction can quietly improve a lot of larger processes behind the scenes.
Keywords: satellite image processing, Cloud Removal, Web-based Application, Cloud Detection, Image enhancement, AI - Based Image Processing, React Web Application
Abstract
AI Driven Market Price Forecasting and Risk Analysis For Agricultural Commodities
Kalaiamuthan N, M. Saravanakumar
DOI: 10.17148/IJIREEICE.2026.14413
Abstract: Agricultural commodity markets are highly volatile due to the combined influence of climatic variability, seasonal production cycles, supply–demand imbalances, transportation constraints, and policy interventions. In developing economies such as India, unpredictable market prices significantly affect farmers income stability, food security, and national economic planning. Traditional statistical forecasting approaches often fail to capture the complex, nonlinear, and temporal dependencies inherent in agricultural price data. This paper presents an AI-driven framework for market price forecasting and risk analysis of agricultural commodities using machine learning and deep learning techniques. Historical price data of major crops such as rice, wheat, maize, and tomato are analyzed to predict future price trends and assess market risks. Models including Autoregressive Integrated Moving Average (ARIMA), Random Forest Regression, and Long Short-Term Memory (LSTM) neural networks are implemented and evaluated. The proposed system integrates data preprocessing, feature engineering, predictive modeling, and risk assessment modules to provide actionable insights for farmers, traders, and policymakers. Experimental results demonstrate that deep learning models outperform traditional methods in forecasting accuracy, while the risk analysis module effectively quantifies price volatility and potential losses. The proposed framework supports informed decision-making and contributes to the development of intelligent agricultural market systems.
Keywords: Artificial Intelligence, Machine Learning, Agricultural Commodities, Price Forecasting, Risk Analysis, Time-Series Analysis
Keywords: Artificial Intelligence, Machine Learning, Agricultural Commodities, Price Forecasting, Risk Analysis, Time-Series Analysis
Abstract
Smart Attendance System Using Deep Learning-Based Facial Recognition
MALA BHARUMATHI M, UDHAYA KUMAR R
DOI: 10.17148/IJIREEICE.2026.14414
Abstract: Traditional attendance management methods in academic and organizational settings are labour-intensive, error-prone, and vulnerable to proxy attendance. This paper presents a Smart Attendance System (SAS) that leverages deep learning-based facial recognition to automate and secure the attendance process. The proposed system employs a Residual Convolutional Network (RCN) for feature extraction, combined with FaceNet embeddings and a Support Vector Machine (SVM) classifier for identity verification. Real-time face detection is achieved using Haar Cascade and Multi-task Cascaded Convolutional Networks (MTCNN), while OpenCV handles video frame acquisition. The system is engineered to capture attendance within a bounded temporal window (25–30 minutes) that corresponds to a single class session, thereby eliminating duplicate entries. Experimental evaluation on a dataset of 200 subjects yields a mean recognition accuracy of 97.4%, a false acceptance rate of 0.8%, and an average processing latency of 1.2 seconds per frame. The system stores records in a relational database and provides administrators with export and reporting capabilities. Results demonstrate that the proposed architecture outperforms conventional LBPH and Eigenface baselines by a statistically significant margin, offering a scalable, contactless, and cost-effective solution for modern smart institutions.
Keywords: facial recognition; deep learning; convolutional neural network; FaceNet; attendance automation; MTCNN; OpenCV; residual network; smart classroom; biometric system.
Keywords: facial recognition; deep learning; convolutional neural network; FaceNet; attendance automation; MTCNN; OpenCV; residual network; smart classroom; biometric system.
Abstract
E-Vehicle Recharge Hub: A Smart Location-Based Charging Solution
Saravana Kumar M, Vasanika S, Priya Dharshini M
DOI: 10.17148/IJIREEICE.2026.14415
Abstract: Electric vehicles are getting really popular these days. People like them because they are better for the environment and fuel prices keep going up. Plus theres this big push everywhere to cut down on carbon emissions. But even with all that, EV owners still have a hard time finding places to charge up. Especially if youre driving somewhere new, its frustrating not knowing where the stations are. That leads to delays and this thing called range anxiety, which I think just makes people worry about running out of battery.
The app I came up with is called E-Vehicle Recharge Hub.The whole point of this app is to make it simpler to find charging stations. You know, with info thats updated in real time so you dont waste time looking around. It pulls in your location using GPS, figures out where you are right now. Then it just pops up the nearest spots on the map. That seems pretty handy, I guess, but sometimes the real time part might lag a bit. Anyway, it shows them quick enough to get you going. You get details like how far they are and if theyre available. That seems like it would help a lot when youre on the road.
One part I focused on is letting users book a slot ahead of time. So you dont have to wait around if its busy. Then theres an admin side where people in charge can update the station info or handle bookings. Once you book something, you get a notification to confirm. It feels like that makes the whole process smoother, though im not totally sure how it handles super crowded areas.
For building it, I used Android Studio. The front end is XML and Java, and SQLite for storing data in the back. It processes things in real time and tries to be easy to use. I tested it some and it does make things more convenient, like optimizing how stations are used. Accessibility improves too, I guess.
Overall this app supports the EV world growing. Combining location stuff with good data handling and thinking about users. It might encourage more people to switch to electric, but theres still room to add more features maybe.
Keywords: Electric Vehicles, Location-Based Services, Android Application, Charging Stations, Booking System
The app I came up with is called E-Vehicle Recharge Hub.The whole point of this app is to make it simpler to find charging stations. You know, with info thats updated in real time so you dont waste time looking around. It pulls in your location using GPS, figures out where you are right now. Then it just pops up the nearest spots on the map. That seems pretty handy, I guess, but sometimes the real time part might lag a bit. Anyway, it shows them quick enough to get you going. You get details like how far they are and if theyre available. That seems like it would help a lot when youre on the road.
One part I focused on is letting users book a slot ahead of time. So you dont have to wait around if its busy. Then theres an admin side where people in charge can update the station info or handle bookings. Once you book something, you get a notification to confirm. It feels like that makes the whole process smoother, though im not totally sure how it handles super crowded areas.
For building it, I used Android Studio. The front end is XML and Java, and SQLite for storing data in the back. It processes things in real time and tries to be easy to use. I tested it some and it does make things more convenient, like optimizing how stations are used. Accessibility improves too, I guess.
Overall this app supports the EV world growing. Combining location stuff with good data handling and thinking about users. It might encourage more people to switch to electric, but theres still room to add more features maybe.
Keywords: Electric Vehicles, Location-Based Services, Android Application, Charging Stations, Booking System
Abstract
SAFETRACK: Real-Time IoT-Based Vehicle Accident Detection, Severity Analysis and Emergency Alert System
Pavan Gharat, Swapnil Kadam, Gaurishankar Awale, Mr. U. S. Shirshetti
DOI: 10.17148/IJIREEICE.2026.14416
Abstract: Road accidents are a major cause of fatalities worldwide, often worsened by delayed emergency response, drunk driving, and lack of real-time monitoring systems. This paper presents SAFETRACK, a real-time IoT-based vehicle safety system that integrates accident detection, accident severity analysis, alcohol detection, smoke detection, and automated emergency alerting into a single embedded platform. The system employs an accelerometer (MPU6050) to detect sudden vehicle impacts, an MQ-3 alcohol sensor to monitor driver sobriety, an MQ-2 smoke sensor to detect in- cabin smoke or gas, a GPS module (NEO-6M) for real-time location tracking, and a GSM module (SIM800L) for emergency SMS dispatch. Upon detection of any critical event, the microcontroller (ESP32/Arduino) processes sensor data, determines severity, and transmits alerts with GPS coordinates to pre-registered emergency contacts and uploads data to an IoT cloud dashboard. The system also disables the vehicle ignition upon alcohol detection. Experimental results demonstrate reliable detection, low response latency, and effective integration of cloud monitoring using ThingSpeak and Google Maps API.
Keywords: IoT, vehicle safety, accident detection, ESP32, MPU6050, GPS, GSM, alcohol detection, smoke detection, emergency alert, severity analysis
Keywords: IoT, vehicle safety, accident detection, ESP32, MPU6050, GPS, GSM, alcohol detection, smoke detection, emergency alert, severity analysis
Abstract
AN AI FRAMEWORK FOR UNSTRUCTURED DATA QUALITY ASSESSMENT WITH INTEGRATED PII DETECTION
Vanitha A, Jumana J
DOI: 10.17148/IJIREEICE.2026.14417
Abstract: Unstructured data, which includes everything from text, documents, and emails to scanned images and log files, is the dominant type of data in many industries, including the realm of enterprise systems and digital communications. Despite the immense potential this type of data holds for analytics and decision-making processes, the usefulness of this data is hindered by the quality issues it faces, including duplication, inconsistency, incomplete data, and unclear data. Making the situation even worse is the presence of personal data, which creates privacy and compliance issues. The current data quality frameworks, which were initially designed to work with structured data, are inadequate to deal with the challenges posed by unstructured data. As such, this project seeks to address the limitations of the current data quality frameworks by developing an innovative and exhaustive data quality assessment framework. This framework is designed to automatically assess the quality of the data while protecting the privacy of the data. It incorporates anomaly detection techniques for log data, cleaning and normalization techniques for text data, and OCR techniques for image data. Additionally, the framework incorporates transformer-based techniques to automatically identify and mask PII. Data quality is assessed based on different parameters, including completeness, consistency, duplication, semantic correctness, and privacy. Beyond reporting, the system produces a cleaned, privacy preserved dataset that is ready for safe use in analytics and machine learning pipelines. By combining AI driven quality assessment with automated privacy safeguards, this project bridges a critical gap between data reliability and regulatory compliance, offering organizations a scalable solution for managing unstructured data with confidence.
Keywords: Unstructured Data, Data Quality Framework, Automation, PII Detection, Text Analytics, Image Quality, Log Analysis, AI-driven Data Cleaning
Keywords: Unstructured Data, Data Quality Framework, Automation, PII Detection, Text Analytics, Image Quality, Log Analysis, AI-driven Data Cleaning
Abstract
REAL-TIME TOMATO LEAF DISEASE CLASSIFICATION AND TREATMENT RECOMMENDATION USING DEEP LEARNING TECHNIQUES
Arockia Ajay J, Arun Kumar K
DOI: 10.17148/IJIREEICE.2026.14418
Abstract: Tomato plants are vulnerable to numerous fungal, bacterial, and viral diseases that severely impact crop yield and farm productivity. Early and accurate disease identification is essential to prevent spread and reduce economic losses. Traditional manual inspection methods are time-consuming and impractical at scale, highlighting the need for automated solutions. This research presents a real-time tomato leaf disease classification and treatment recommendation system using deep learning. Four architectures are evaluated comparatively: CNN, MobileNetV2, ResNet50, and EfficientNetB0. EfficientNetB0 achieves the highest classification accuracy and is deployed for live webcam-based inference. A rule-based recommendation module delivers disease-specific guidance covering chemical, organic, and preventive management strategies. The integrated system provides farmers with a reliable, end-to-end decision-support tool for early diagnosis and effective crop management.
Keywords: Tomato Leaf Disease Detection, Deep Learning, Transfer Learning, Convolutional Neural Networks (CNN), MobileNetV2, ResNet50, EfficientNetB0, Real-Time Disease Classification, Treatment Recommendation System, Precision Agriculture.
Keywords: Tomato Leaf Disease Detection, Deep Learning, Transfer Learning, Convolutional Neural Networks (CNN), MobileNetV2, ResNet50, EfficientNetB0, Real-Time Disease Classification, Treatment Recommendation System, Precision Agriculture.
Abstract
A Q-Learning Based Dynamic Pricing Model for Minimizing Food Waste in Supermarkets
D. Vimal Kumar, A. Hemalatha, S. Harish, M. Mohammed Yunush
DOI: 10.17148/IJIREEICE.2026.14419
Abstract: Food waste in supermarkets is a big problem for the economy and the environment, especially when it comes to perishable goods that don't last long. Traditional static pricing strategies don't take into account changes in demand or the freshness of products, which can lead to unsold inventory and more waste. This study proposes a Q-learning-based dynamic pricing model that modifies product prices as they near expiration to tackle this issue.The pricing issue is framed as a Markov Decision Process (MDP), wherein the system acquires optimal pricing strategies via ongoing engagement with the environment. Important things like how long the item will last, how much demand there is for it, and how much stock there is are all taken into account when making decisions. The model's goal is to make as much money as possible while keeping unsold stock and food waste to a minimum.Experimental results show that the suggested method works better than traditional pricing methods to cut down on food waste and increase overall profits. This study shows that reinforcement learning methods can be used to create smart and long-lasting pricing systems for stores.
Keywords: Dynamic pricing, Q-learning, reinforcement learning, reducing food waste from perishable goods, the Markov decision process (MDP), and managing inventory
Keywords: Dynamic pricing, Q-learning, reinforcement learning, reducing food waste from perishable goods, the Markov decision process (MDP), and managing inventory
Abstract
DETECTION OF STAMMERING IN SPEECH USING MACHINE LEARNING WITH HUMAN EVALUATION
ARAVINDAGOKUL. P, ARUN KUMAR K
DOI: 10.17148/IJIREEICE.2026.14420
Abstract: Stammering, also known as stuttering, is a speech fluency disorder characterized by involuntary repetitions, prolongations, and blocks in speech production. This paper presents a comprehensive framework for automatic detection of stammering using machine learning, augmented with structured human evaluation. Feature extraction uses MFCC, pitch contour, zero-crossing rate, and nergy-based features. Multiple classifiers including SVM, Random Forest, LSTM, and CNN are trained and evaluated. A human evaluation protocol validated model predictions against speech-language pathologists (SLPs). The proposed hybrid LSTM+RF approach achieves 94.7% accuracy with an F1-score of 0.943, outperforming existing standalone methods. Human-Model Agreement of 91.5% with Cohen's Kappa k=0.83 confirms clinical reliability.
Keywords: Stammering Detection, Speech Processing, Machine Learning, SVM, CNN, Human Evaluation, Speech- Language Pathology.
Keywords: Stammering Detection, Speech Processing, Machine Learning, SVM, CNN, Human Evaluation, Speech- Language Pathology.
Abstract
DATA-DRIVEN MANUFACTURING QUALITY TRACKING AND VISUALIZATION PLATFORM
Arockiya Anjugan M L, A S Krishna
DOI: 10.17148/IJIREEICE.2026.14421
Abstract: The Manufacturing Quality Control Tracker is an advanced data-driven platform designed to enhance quality assurance in manufacturing through real-time monitoring, statistical analysis, and predictive machine learning. Built using Python, Streamlit, Plotly, and Scikit-learn, the system enables automated defect detection, process monitoring, and trend analysis while adhering to ISO 9001 standards. This paper presents the system architecture, mathematical models, experimental results, and a comparative evaluation against existing approaches. Results demonstrate significant reductions in defect rates and improvements in production efficiency.
Keywords: Manufacturing Quality Control, Defect Detection, Streamlit, Machine Learning, Statistical Process Control, Real-Time Monitoring, ISO 9001.
Keywords: Manufacturing Quality Control, Defect Detection, Streamlit, Machine Learning, Statistical Process Control, Real-Time Monitoring, ISO 9001.
Abstract
Machine Learning-Based Predictive Analytics for Sustainable Resource Consumption Forecasting: A Comparative Study
Dr. D. VIMAL KUMAR, SHARMILA S, LAKSHYA SHREE R, HARISH I
DOI: 10.17148/IJIREEICE.2026.14422
Abstract: Accurate prediction of resource consumption has emerged as a fundamental prerequisite for achieving global sustainability targets. Traditional statistical models, while useful, face inherent limitations in capturing the non-linear and multi-dimensional nature of resource usage dynamics. This study investigates the comparative predictive performance of five widely adopted supervised machine learning algorithms—Linear Regression, Decision Tree, Random Forest, Support Vector Machine (SVM), and Gradient Boosting—applied to a structured sustainability dataset comprising population density, industrial activity, energy consumption, water usage, rainfall, and recycling rate variables. Models were trained on an 80:20 stratified data split with standardized feature scaling, and evaluated using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and the Coefficient of Determination (R²). Empirical results reveal that Linear Regression achieved the highest R² score of 0.9757 with the lowest MAE of 3.7376, followed by Gradient Boosting (R² = 0.9486) and Random Forest (R² = 0.9203). Feature importance analysis confirmed that industrial activity index and energy consumption exert the dominant influence on resource demand predictions. The findings provide data-driven guidelines for policymakers and planners seeking to adopt machine learning-based forecasting frameworks for improved sustainable resource management.
Keywords: Machine Learning; Sustainable Resource Management; Predictive Analytics; Gradient Boosting; Random Forest; Feature Engineering; Comparative Evaluation
Keywords: Machine Learning; Sustainable Resource Management; Predictive Analytics; Gradient Boosting; Random Forest; Feature Engineering; Comparative Evaluation
Abstract
AI-Powered Digital Marketing: Transforming the Future of Customer Engagement A Comprehensive Study on Artificial Intelligence Applications in Modern Marketing
Madhulekha E, Ms. A. Hemalatha
DOI: 10.17148/IJIREEICE.2026.14423
Abstract: The integration of Artificial Intelligence (AI) in digital marketing has fundamentally transformed how businesses interact with customers, optimize campaigns, and achieve measurable results. This study explores AI-powered digital marketing, examining its core technologies, practical applications, implementation strategies, and future implications. AI technologies — including machine learning, natural language processing, and predictive analytics — are revolutionizing personalization, automation, customer service, and campaign optimization. Businesses adopting AI experience 40–60% reduction in content creation time and 15–25% uplift in engagement rates, providing valuable insights for marketing professionals and students in the evolving digital marketing ecosystem.
Keywords: Digital Marketing, Digital Marketing Tools, Semrush, AI Chatbot, N8N, Word AI
Keywords: Digital Marketing, Digital Marketing Tools, Semrush, AI Chatbot, N8N, Word AI
Abstract
Medical Insurance Price Prediction Using Machine Learning
Chithra Devi C M.Sc (Ph.D), Abdul Rasheed M, Pravin Kumar V
DOI: 10.17148/IJIREEICE.2026.14424
Abstract: In recent years, the rising cost of healthcare has made medical insurance an essential component of financial planning. However, accurately estimating insurance charges remains a challenging task due to the influence of multiple factors such as age, gender, body mass index (BMI), lifestyle habits, and medical history. Traditional methods used by insurance companies often rely on manual calculations and generalized assumptions, which may lead to inaccurate pricing and lack of transparency.
This paper presents a machine learning-based approach for predicting medical insurance costs using historical data. The proposed system analyzes key features including age, BMI, number of dependents, smoking status, and region to identify patterns that influence insurance charges. Various machine learning algorithms such as Linear Regression, Decision Tree, Random Forest, and Gradient Boosting are implemented and compared to determine the most accurate predictive model.
The dataset is preprocessed through data cleaning, feature encoding, and normalization to improve model performance. The models are trained and evaluated using appropriate performance metrics such as Mean Absolute Error (MAE), Mean Squared Error (MSE), and R-squared score. Among the models tested, ensemble techniques like Random Forest and Gradient Boosting demonstrate superior prediction accuracy due to their ability to handle complex, non-linear relationships in the data.
The results show that machine learning can significantly improve the accuracy and efficiency of insurance cost prediction compared to traditional methods. This system can assist insurance companies in fair pricing strategies and help individuals estimate their medical expenses more effectively.
In conclusion, the proposed model highlights the potential of machine learning in transforming the healthcare insurance sector by providing data-driven, transparent, and reliable cost predictions.
This paper presents a machine learning-based approach for predicting medical insurance costs using historical data. The proposed system analyzes key features including age, BMI, number of dependents, smoking status, and region to identify patterns that influence insurance charges. Various machine learning algorithms such as Linear Regression, Decision Tree, Random Forest, and Gradient Boosting are implemented and compared to determine the most accurate predictive model.
The dataset is preprocessed through data cleaning, feature encoding, and normalization to improve model performance. The models are trained and evaluated using appropriate performance metrics such as Mean Absolute Error (MAE), Mean Squared Error (MSE), and R-squared score. Among the models tested, ensemble techniques like Random Forest and Gradient Boosting demonstrate superior prediction accuracy due to their ability to handle complex, non-linear relationships in the data.
The results show that machine learning can significantly improve the accuracy and efficiency of insurance cost prediction compared to traditional methods. This system can assist insurance companies in fair pricing strategies and help individuals estimate their medical expenses more effectively.
In conclusion, the proposed model highlights the potential of machine learning in transforming the healthcare insurance sector by providing data-driven, transparent, and reliable cost predictions.
Abstract
Using Machine Learning and Natural Language Processing, A System Can Find Spam Emails
Ayan Husain, Khan Amir Alam, Khan Abis, Abdul Gaffar, Prof. Imran Shahid
DOI: 10.17148/IJIREEICE.2026.14425
Abstract: Spam emails aren’t just an annoyance—they signal that your information’s probably out there, somewhere, and there’s always a scam lurking behind the next “Congratulations!” subject line. You can keep hitting delete, but let’s be honest, the junk keeps coming. Spammers always find new tricks, and those dusty old filters? They’re not up for the challenge.
But you don’t have to keep fighting a losing battle. Machine learning and natural language processing can actually do the heavy lifting. The goal: catch all the bad stuff, save what matters, and move past filters that don’t really get the job done anymore.
Here’s how it works. The system grabs incoming emails and strips out all the mess—HTML tags, weird symbols, filler words. Basically, anything that clouds the real message gets cleared away. Then, it turns the cleaned-up text into numbers using TF-IDF, which helps home in on what’s actually being said instead of the usual noise.
Once that’s done, the machine learning models take over. We throw a few at the problem—Support Vector Machine, Logistic Regression, and Random Forest—all taking their best shot at flagging spam. And we don’t just check for accuracy; we look at precision, recall, and F1-score. We want the filter to spot spam, but not at the cost of real emails slipping through the cracks.
Plus, it’s not all tucked away behind the scenes. Up front, there’s a Streamlit app where you can try out a single email or toss in a whole batch. Need to go bigger? There’s an API ready to plug into larger systems or future upgrades.
So, does it actually work? Yeah, it does. Tests show it nails the spam, and your legit emails aren’t collateral damage. Whether you’re using this solo or rolling it out for a group, it can keep up.
At its core, this isn’t just buzzwords and promises. It’s fast, flexible, and ready for whatever comes next—maybe more advanced models or smarter features down the road. No nonsense. Just a better way to keep your inbox from turning into a junkyard.
Keywords: Spam Detection, Machine Learning, Natural Language Processing, TF-IDF, Email Classification, SVM.
But you don’t have to keep fighting a losing battle. Machine learning and natural language processing can actually do the heavy lifting. The goal: catch all the bad stuff, save what matters, and move past filters that don’t really get the job done anymore.
Here’s how it works. The system grabs incoming emails and strips out all the mess—HTML tags, weird symbols, filler words. Basically, anything that clouds the real message gets cleared away. Then, it turns the cleaned-up text into numbers using TF-IDF, which helps home in on what’s actually being said instead of the usual noise.
Once that’s done, the machine learning models take over. We throw a few at the problem—Support Vector Machine, Logistic Regression, and Random Forest—all taking their best shot at flagging spam. And we don’t just check for accuracy; we look at precision, recall, and F1-score. We want the filter to spot spam, but not at the cost of real emails slipping through the cracks.
Plus, it’s not all tucked away behind the scenes. Up front, there’s a Streamlit app where you can try out a single email or toss in a whole batch. Need to go bigger? There’s an API ready to plug into larger systems or future upgrades.
So, does it actually work? Yeah, it does. Tests show it nails the spam, and your legit emails aren’t collateral damage. Whether you’re using this solo or rolling it out for a group, it can keep up.
At its core, this isn’t just buzzwords and promises. It’s fast, flexible, and ready for whatever comes next—maybe more advanced models or smarter features down the road. No nonsense. Just a better way to keep your inbox from turning into a junkyard.
Keywords: Spam Detection, Machine Learning, Natural Language Processing, TF-IDF, Email Classification, SVM.
Abstract
AI-Powered Real-Time Firearm Detection and Alert System Using YOLOv12m Deep Learning Model
Savita Waghmode, Prof. Vidhate S.N.
DOI: 10.17148/IJIREEICE.2026.14426
Abstract: The increasing frequency of firearm-related incidents in public spaces demands intelligent, automated surveillance systems capable of real-time threat detection. Conventional closed-circuit television (CCTV) systems depend entirely on human operators, making them susceptible to fatigue-induced delays and missed detections. This paper proposes an AI- powered real-time firearm detection and alert system leveraging the YOLOv12m (You Only Look Once, version 12, medium variant) deep learning model. YOLOv12m is selected over its nano counterpart due to its significantly larger parameter count, deeper feature extraction capability, and superior mean Average Precision (mAP), making it more suitable for detecting small and partially occluded firearms in complex surveillance environments. The system captures live video from CCTV cameras or webcams using OpenCV, preprocesses each frame to a YOLO-compatible resolution, and performs single-pass inference to detect pistols and rifles with bounding boxes and confidence scores. Upon detection, the system triggers both visual overlays and audio alerts to immediately notify security personnel. The model is trained on a curated firearm dataset sourced from Roboflow and Kaggle using transfer learning on pre-trained COCO weights. Experimental evaluation yields a Precision of 83.5%, Recall of 80%, mAP@0.5 of 87.1%, and a False Positive Rate of 0.25%, demonstrating the system’s reliability for deployment in smart cities, airports, educational campuses, and government facilities.
Keywords: YOLOv12m, firearm detection, deep learning, real-time surveillance, object detection, OpenCV, transfer learning, convolutional neural network, alert system
Keywords: YOLOv12m, firearm detection, deep learning, real-time surveillance, object detection, OpenCV, transfer learning, convolutional neural network, alert system
Abstract
Threat Intelligence System for Cyber Attacks
Kunal Naikade, Ritesh Patni, Prathmesh Jaiswal, Smita Chunamari
DOI: 10.17148/IJIREEICE.2026.14427
Abstract: Cybersecurity threats have become increasingly sophisticated, with phishing attacks remaining one of the most prevalent and damaging forms of cybercrime. This project focuses on developing a threat intelligence system designed to detect and mitigate phishing links in real time. By utilizing advanced machine learning algorithms and natural language processing techniques, the system analyzes URLs, email content, and website characteristics to identify malicious patterns indicative of phishing attempts. The model is trained on large datasets of legitimate and fraudulent links to maximize detection accuracy and reduce false positives. Additionally, the system integrates threat intelligence feeds to enhance adaptability against evolving attack strategies. The ultimate goal of this project is to provide a proactive cybersecurity solution that identifies phishing threats before they compromise user data or organizational networks. If effectively implemented, the system can strengthen online security, prevent financial losses, and support the broader effort toward safer digital ecosystems.
Abstract
IOT BASED SMART HOME AUTOMATION SYSTEM
Mrs. Suryasevak Singh, Rafay Siddiqui, Janhvi Patil, Bhavya Chavri, Irfan Sarkar, Balwant Yedage
DOI: 10.17148/IJIREEICE.2026.14428
Abstract: The Internet of Things (IOT) based Smart Home Automation System is a modern technological solution developed to enhance comfort, convenience, security, and energy efficiency in residential environments. With the rapid growth of IOT technology, everyday household devices can now be connected to the internet, enabling users to monitor and control them remotely through smartphones, tablets, or computers. This system integrates micro-controllers, sensors, actuators, and wireless communication modules such as Wi-Fi to create an intelligent and interconnected home environment. The proposed system allows users to control electrical appliances including lights, fans, air conditioners, televisions, and other devices from anywhere using a mobile application or web interface. It also supports automation features where devices can operate automatically based on sensor inputs such as temperature, humidity, light intensity, gas leakage, or motion detection. For example, lights can turn on automatically when motion is detected, and fans can adjust speed according to room temperature.
Keywords: IoT, Smart Home Automation, ESP32, Sensors, Wi-Fi, Remote Monitoring, Home Security, Energy Efficiency, Automation System
Keywords: IoT, Smart Home Automation, ESP32, Sensors, Wi-Fi, Remote Monitoring, Home Security, Energy Efficiency, Automation System
Abstract
Comparative Analysis and Cross-Dataset Validation of Traditional Machine Learning Models for Fake News Detection
Pranjal L. Bhamre, Isha Y. Jain, Sunita N. Deore
DOI: 10.17148/IJIREEICE.2026.14429
Abstract: The viral spread of digital misinformation, often called an "infodemic," has moved beyond a technical nuisance to become a genuine threat to public health and social stability. While modern research favors Deep Learning models such as BERT, these architectures often demand hardware resources that are not practical for real-time, decentralized deployment. This study shifts the focus to computational efficiency by evaluating four traditional machine learning classifiers—Logistic Regression, Support Vector Machine (SVM), Multinomial Naïve Bayes, and the Passive- Aggressive Classifier (PAC)—on a corpus of 39,103 news articles. By combining a Regex-based preprocessing pipeline with optimized TF-IDF vectorization, the proposed framework achieved a peak in-domain accuracy of 0.995 using PAC. However, cross-dataset validation on the LIAR benchmark dataset revealed a performance decline to 0.474, primarily due to contextual sparsity in short-form political statements. These findings suggest that while traditional models are highly effective for long-form news classification, they require semantic enhancement to handle sparse social media content. Overall, this work supports a sustainable "Green AI" perspective that emphasizes computational efficiency while acknowledging cross-domain limitations.
Keywords: Fake News Detection; Traditional Machine Learning; Passive-Aggressive Classifier; TF- IDF; Cross- Dataset Validation; Domain Shift; Text Classification
Keywords: Fake News Detection; Traditional Machine Learning; Passive-Aggressive Classifier; TF- IDF; Cross- Dataset Validation; Domain Shift; Text Classification
Abstract
The Effectiveness of CTV Advertising on Brand Equity and Consumer Engagement
Eshika Makwana, Hetang Mahajan and Hiren Harsora
DOI: 10.17148/IJIREEICE.2026.14430
Abstract: Connected TV (CTV) advertising has become a major factor in influencing consumer behaviour due to the rapid growth of digital streaming platforms, especially among digitally active audiences. However, limited research exists on how CTV advertising impacts brand equity and consumer engagement, particularly in the Indian context. This study examines consumer responses and analyses how key aspects of CTV advertising such as personalization, content relevance, and targeting efficiency affect brand perception and engagement levels. A structured questionnaire was used to gather data from 100 active digital media users in India as part of a quantitative research design. Descriptive statistics, correlation analysis, and basic analytical tools were used to interpret the data. The results show that while CTV advertising has a moderate effect on building brand trust, it has a strong and significant impact on consumer engagement and brand recall. Additionally, the role of Amazon DSP in enabling precise audience targeting enhances the effectiveness of CTV campaigns. The findings also indicate that consumers respond positively to personalized and non-intrusive advertisements delivered through CTV platforms. By highlighting the direct impact of CTV advertising on engagement and brand outcomes, this study contributes to the growing literature on digital and programmatic advertising.
Keywords: Connected TV (CTV), Amazon DSP, Consumer Engagement, Brand Equity, Digital Advertising, Personalization, Targeting Efficiency
Keywords: Connected TV (CTV), Amazon DSP, Consumer Engagement, Brand Equity, Digital Advertising, Personalization, Targeting Efficiency
Abstract
Customer Churn Prediction In Subscription-Based Platforms Using Machine Learning
YUVA DHARSHINI M, MALA BHARUMATHI M
DOI: 10.17148/IJIREEICE.2026.14431
Abstract: Customer churn is a major challenge in subscription-based platforms, where customers discontinue their services, leading to revenue loss. This project aims to predict customer churn and understand the key factors influencing customer behavior so that businesses can take actions to improve customer retention. In this project, a subscription-based customer dataset will be used, which includes features such as subscription type, usage patterns, login frequency, and payment details. The data will be preprocessed by handling missing values and encoding categorical variables. Exploratory Data Analysis (EDA) will be performed to identify patterns and trends related to customer churn. Machine learning algorithms such as Logistic Regression, Decision Tree, Random Forest, and XGBoost will be implemented to build predictive models. A comparative analysis will be carried out using evaluation metrics such as accuracy, precision, recall, and F1-score to identify the best-performing model. XGBoost is included as an advanced algorithm because it improves prediction accuracy by combining multiple weak models and handling complex data patterns effectively. By performing this analysis, the project will identify high-risk customers and the key factors leading to churn, enabling businesses to design effective retention strategies such as personalized offers and improved customer engagement. As an enhancement, an interactive dashboard will be developed to visualize churn patterns and monitor customer risk levels for better decision-making.
Keywords: Customer Churn Prediction, Machine Learning, XGBoost, Model Comparison, Subscription-Based Platforms
Keywords: Customer Churn Prediction, Machine Learning, XGBoost, Model Comparison, Subscription-Based Platforms
Abstract
Career Guidance System For Personalised Career Pathways
Arti Jaibhai, Rushikesh Bembale, Arjun Gade, Jeevan Maske, Varun Baporikar
DOI: 10.17148/IJIREEICE.2026.14432
Abstract: The transition after 12th standard is one of the most critical phases in a student’s academic journey, where choosing an appropriate career path often becomes challenging due to a lack of proper guidance and awareness. Traditional counseling methods are generally limited, generalized, and may be influenced by human bias. To overcome these limitations, this research proposes a Career Guidance System for Personalized Career Pathways that provides data- driven career recommendations based on student aptitude, academic performance, personal interests, and psychometric assessment results. The system utilizes Cosine Similarity and Decision Tree algorithms to analyze multiple parameters and generate personalized suggestions across domains such as Engineering, Medical, Commerce, Management, Arts, Design, and Government Services. Experimental evaluation on sample student profiles shows that the system delivers relevant and unbiased recommendations with improved decision accuracy compared to conventional counseling methods. The proposed system is scalable, cost effective, and user-friendly, making it a practical solution for students seeking informed career decisions.
Keywords: Career Guidance System, 12th Standard Students, Cosine Similarity, Decision Tree, Career Selection, Stream Selection (Science, Commerce, Arts), Personalized Career Recommendations, Data Analysis, Psychometric Assessment, Aptitude Analysis, Digital Platform, Accessibility, Career Counseling.
Keywords: Career Guidance System, 12th Standard Students, Cosine Similarity, Decision Tree, Career Selection, Stream Selection (Science, Commerce, Arts), Personalized Career Recommendations, Data Analysis, Psychometric Assessment, Aptitude Analysis, Digital Platform, Accessibility, Career Counseling.
Abstract
An AI-Powered Conversational System for Rapid Digital Forensic Analysis
Praveen R, Krishna AS
DOI: 10.17148/IJIREEICE.2026.14433
Abstract: In the field of digital forensics, the escalating volume and complexity of evidence data constitutes a critical bottleneck for investigators. Traditional manual extraction and analysis workflows are time-intensive and error-prone, particularly when dealing with large-scale datasets from seized mobile and computing devices. This paper presents an innovative AI-driven conversational system designed to substantially accelerate forensic analysis pipelines. Leveraging a Retrieval-Augmented Generation (RAG) architecture, the system ingests and indexes data encoded in the Universal Forensic Data Representation (UFDR) format, enabling forensic analysts to interrogate complex digital evidence through natural language queries. Our proposed architecture integrates a transformer-based embedding model, a vector similarity retrieval engine, and a large language model (LLM) generation layer to deliver contextually accurate, evidence-grounded responses. Experimental evaluations on a curated UFDR dataset demonstrate that our system reduces mean query-response time by 67% compared to conventional keyword-based tools, achieving a retrieval precision of 0.91 and an answer faithfulness score of 0.88. These results validate the efficacy of RAG-based conversational interfaces for investigative workflows and signal a paradigm shift in digital forensic methodology.
Keywords: Digital forensics, retrieval-augmented generation, large language models, UFDR, natural language processing, conversational AI, evidence retrieval, transformer models.
Keywords: Digital forensics, retrieval-augmented generation, large language models, UFDR, natural language processing, conversational AI, evidence retrieval, transformer models.
Abstract
ELECTRICAL VEHICLE CHARGING WITH INTRELEAVED CONVERTER
Shruti M. Kavare, Pranav T. Chougule, Sakshi K. Powar, Pratiksha S. Rokade, Ms. Shubhangi Manjare
DOI: 10.17148/IJIREEICE.2026.14434
Abstract: The rapid growth of electric vehicles (EVs) has increased the demand for efficient and reliable charging systems. Conventional chargers often suffer from high current ripple, poor power quality, and thermal stress. To overcome these challenges, this work proposes an EV charging system using an Interleaved DC–DC Converter. Multiple converter phases operate with phase-shifted control signals to share load current, reducing current ripple, switching losses, and thermal stress. The system includes a front-end AC–DC rectifier with power factor correction followed by the interleaved converter for regulated battery charging. Simulation results show improved efficiency, lower total harmonic distortion (THD), and better dynamic performance compared to conventional converters.
Keywords: Electric Vehicle (EV), Interleaved Converter, DC–DC Conversion, Power Factor Correction (PFC), Ripple Reduction, Battery Charging System.
Keywords: Electric Vehicle (EV), Interleaved Converter, DC–DC Conversion, Power Factor Correction (PFC), Ripple Reduction, Battery Charging System.
Abstract
Training & Development- A Study on the Role of Training and Development in Enhancing Employee Performance and Organizational Growth
Lakshya Kothari, Avesh Khanjada, Dr. Sapana Chauhan
DOI: 10.17148/IJIREEICE.2026.14435
Abstract: Training and Development (T&D) is one of the most practical ways to improve employee performance and drive business growth. In today’s fast-paced market, companies need a skilled and adaptable workforce, especially in strict, highly regulated industries like pharmaceutical manufacturing. This study evaluates how effective the T&D practices are at Gopaldas Visram & Company Limited (GVCL) in Halol, and looks at how these programs actually impact employee output and the company's success. We collected our data by surveying 100 employees using a structured questionnaire. To make sense of the responses, we used descriptive statistics, correlation analysis, and linear regression. The results showed a very strong positive link ($r = 0.85$) between good training and highly motivated employees. Additionally, our regression model proved that for every one-unit improvement in training quality, employee performance goes up by 0.83 units.
Ultimately, this study confirms that well-planned training programs do more than just teach skills—they directly boost performance, reduce mistakes, and help the organization grow.
Keywords: Training and Development, Employee Performance, Organizational Growth, Pharmaceutical Industry, HR Practices.
Ultimately, this study confirms that well-planned training programs do more than just teach skills—they directly boost performance, reduce mistakes, and help the organization grow.
Keywords: Training and Development, Employee Performance, Organizational Growth, Pharmaceutical Industry, HR Practices.
Abstract
AI-DRIVEN DATA LIFECYCLE OPTIMIZATION SYSTEM
V TEJASREE, MANEESHA P A
DOI: 10.17148/IJIREEICE.2026.14436
Abstract: Today, most organizations store huge amounts of data without knowing what is still useful and what is no longer needed. They often keep everything, which wastes storage space and increases unnecessary costs. The main issue is the lack of a proper system to automatically decide when data should be retained or removed, as existing methods rely on fixed rules like deleting data after a certain number of years, which is not effective for all types of data. To address this, this paper presents an AI-Driven Data Lifecycle Optimization System that scans data, evaluates its usefulness, and decides whether it should be kept or deleted. Instead of fixed rules, it uses machine learning to classify data based on importance and survival analysis to predict when data is no longer needed. The system is built using Python for data processing, R for lifecycle prediction, SQL for managing retention schedules, and Power BI for visualization. The results show that this approach reduces storage costs, saves time, and improves data management compared to traditional rule- based methods, providing a smarter and more efficient solution for managing data lifecycle.
Keywords: Data Lifecycle, AI-Driven System, Machine Learning, Survival Analysis, Data Retention, Power BI, Data Optimization.
Keywords: Data Lifecycle, AI-Driven System, Machine Learning, Survival Analysis, Data Retention, Power BI, Data Optimization.
Abstract
Machine Learning in Healthcare: A Comprehensive Review
Shruti Gosavi, Pooja Gunjal, Sunita N. Deore
DOI: 10.17148/IJIREEICE.2026.14437
Abstract: Machine learning (ML) has emerged as a transformative force in the healthcare sector, enabling advanced data analysis, improved diagnostic accuracy, and efficient clinical decision-making. The rapid digitization of healthcare systems has resulted in the generation of vast amounts of structured and unstructured data from sources such as electronic health records, medical imaging technologies, wearable devices, and laboratory systems. Traditional analytical approaches often struggle to manage and interpret such complex datasets effectively. In this context, machine learning provides powerful computational techniques that can identify hidden patterns, predict outcomes, and support medical professionals in delivering high-quality care. This paper presents a comprehensive review of machine learning applications in healthcare, covering areas such as disease diagnosis, predictive analytics, medical imaging, and healthcare operations. It also describes a structured methodology for developing machine learning models, including data preparation, preprocessing, model selection, and evaluation. Experimental results based on a synthetic dataset are discussed to illustrate model performance. Furthermore, the study highlights key challenges such as data privacy, model interpretability, and implementation barriers. The paper concludes by discussing future research directions aimed at enhancing the reliability and adoption of machine learning in healthcare systems.
Keywords: Machine Learning, Healthcare Analytics, Artificial Intelligence, Predictive Modeling, Medical Imaging, Natural Language Processing, Clinical Decision Support
Keywords: Machine Learning, Healthcare Analytics, Artificial Intelligence, Predictive Modeling, Medical Imaging, Natural Language Processing, Clinical Decision Support
Abstract
Animal Detection in Farms Using OpenCV
Sharmila S, M.Saravanakumar
DOI: 10.17148/IJIREEICE.2026.14438
Abstract: Animal intrusion in farms causes huge losses in agricultural revenue which a farmer cannot bear. Computer Vision are being increasingly applied in agricultural field for higher productivity by automating tasks. We propose an AI based system which monitors the field using cameras for any intrusion by the animals and alerts the farmer or can even take certain actions on its own.. Real time object detection as the name suggests is an art of detecting various objects at that particular time. Object detection has always been a daring task. For this purpose, faster computation power is required in the identification of an object. However, any system working in actual time generates data which is unlabeled and which has a requirement of huge set of labeled data for potent training purposes. In this work,there is a presentation of a developed application for detecting specific objects (i.e. animals) based on OpenCV libraries.
Keywords: Animal Detection, OpenCV, Real-Time Detection, Farm Surveillance, Deep Learning, Python.
Keywords: Animal Detection, OpenCV, Real-Time Detection, Farm Surveillance, Deep Learning, Python.
Abstract
AIR QUALITY DATA ANALYSIS AND DYNAMIC VISUALIZATION
Roshan Zameer M, Ms. Hemalatha A
DOI: 10.17148/IJIREEICE.2026.14439
Abstract: Air Quality Analysis is a project aimed at monitoring and evaluating air pollution levels using key parameters such as PM2.5, PM10, CO, SO₂, NO₂, and O₃. The system collects and analyses real-time or historical data to identify patterns and trends in air quality. Using data analysis and visualization techniques, the project provides insights into pollution levels and their impact on health and the environment. The results help in predicting air quality changes and support effective decision-making for pollution control and environmental sustainability.
Keywords: Air Quality Analysis, Air Pollution, PM2.5, PM10, Data Analytics, Environmental Monitoring, Pollution Prediction, Machine Learning
Keywords: Air Quality Analysis, Air Pollution, PM2.5, PM10, Data Analytics, Environmental Monitoring, Pollution Prediction, Machine Learning
Abstract
AgroConsultant: Optimizing Fertilizer Recommendations in Precision Agriculture
S SRINITHI, M SARAVANAKUMAR
DOI: 10.17148/IJIREEICE.2026.14440
Abstract: Agriculture remains the foundational pillar of the Indian economy, ensuring both food security and rural livelihood for millions; however, the sector faces significant hurdles as farmers often lack the data-driven guidance necessary for optimal crop selection. This disconnect frequently results in suboptimal yields, resource wastage, and long- term soil degradation. To address these challenges, this paper presents the Smart Crop Recommendation System (SCRS), a comprehensive precision agriculture framework designed to modernize traditional decision-making. At the heart of the SCRS is a robust Random Forest classification algorithm, chosen for its superior ability to handle the non-linear complexities of agricultural datasets. The model is trained on a high-dimensional feature set comprising critical soil parameters—nitrogen (N), phosphorus (P), potassium (K), and pH levels—alongside environmental variables such as temperature, humidity, and rainfall. A key innovation of this system is its departure from static historical models through the integration of the OpenWeatherMap API. This allows the SCRS to fetch real-time meteorological data, enabling the system to provide dynamic, hyper-localized recommendations tailored to the specific micro-climates of districts in Tamil Nadu, including Coimbatore, Erode, and Salem. To bridge the gap between complex machine learning and field application, the system is deployed via a Flask-based web dashboard. This user interface provides an intuitive experience where farmers and agronomists can access not only crop suitability predictions but also targeted fertilizer advice and historical weather visualizations. Experimental evaluations demonstrate that the Random Forest model achieves high classification accuracy and F1-scores, validating its reliability for real-world deployment. Ultimately, this research offers a scalable, intelligent decision-support tool that empowers stakeholders to transition toward resource-efficient, high-yield cultivation practices.
Keywords: Crop Recommendation, Random Forest, Machine Learning, Precision Agriculture, Weather Integration, Flask, Tamil Nadu, Soil Analysis, Fertilizer Recommendation
Keywords: Crop Recommendation, Random Forest, Machine Learning, Precision Agriculture, Weather Integration, Flask, Tamil Nadu, Soil Analysis, Fertilizer Recommendation
Abstract
Intelligent Signal Processing for 5G/6G OFDM Systems Using Deep Learning
V.Amulya, V.Jyothsna, G.Yukthavi, P.Nakshathra
DOI: 10.17148/IJIREEICE.2026.14441
Abstract: The fast development of wireless communication technologies has placed a great pressure on the requirements of high data rates, low latency, and reliable transmission, especially in innovative systems, like 5G and future 6G networks. Orthogonal Frequency Division Multiplexing (OFDM) has proven to be one of the most important modulation methods since it can effectively absorb multipath fading and effectively use the spectrum. Nonetheless, the functioning of the OFDM systems is very sensitive to the proper channel estimation and signal detection that becomes difficult in the dynamic wireless environment due to noise, interference and fading. Conventional methods like Least Squares (LS) and Minimum Mean Square Error (MMSE) are based on the priori mathematical models and statistical conditions, which may be ineffective in reflecting the actual variations of channels. In order to address these drawbacks, this paper suggests a smart signal processing solution based on deep learning, namely Long Short-Term Memory (LSTM) networks, to estimate the channel jointly and detect symbols in the OFDM system. The model proposed is capable of learning the complex channel behaviors directly through the data without having prior channel knowledge. MATLAB simulations are used to implement and assess the system under different conditions of the channel. The findings indicate that it has great enhancements in Bit Error Rate (BER) performance, robustness, and adaptability over traditional techniques. This research paper emphasizes how deep learning methods can be applied in wireless communication systems to develop intelligent, adaptive, and efficient receivers in the future high-speed network.
Keywords: OFDM, Deep Learning, LSTM, Channel Estimation, Signal Detection,5G ,6G ,BER Wireless Communication, Neural Networks, Intelligent Signal Processing, MATLAB Simulation.
Keywords: OFDM, Deep Learning, LSTM, Channel Estimation, Signal Detection,5G ,6G ,BER Wireless Communication, Neural Networks, Intelligent Signal Processing, MATLAB Simulation.
Abstract
PREDICTIVE ANALYSIS MODEL FOR EARLY DETECTION OF ALZHEIMER’S DISEASE USING CLINICAL AND COGNITIVE DATA
Mr.A.Naga Srinivasa Rao¹,T.Sai Meghana²,A.Sumadhuravani³, P.Padma Gandha †, V.Venkata Kalyaniµ
DOI: 10.17148/IJIREEICE.2026.14443
Abstract: Alzheimer’s disease is a progressive neurological disorder that impairs memory, cognitive functions, and daily activities, making early detection essential for effective treatment and management. This paper presents a predictive analysis model for the early detection of Alzheimer’s disease using clinical and cognitive data. The proposed system employs machine learning techniques to analyze input parameters such as age, body mass index, Mini-Mental State Examination (MMSE) score, memory score, physical activity, sleep quality, and family history. A classification- based approach is used to categorize individuals into four stages: No Alzheimer’s, Cognitively Normal, Mild Cognitive Impairment, and Alzheimer’s Disease. The model is trained and evaluated on a structured dataset, and its performance is assessed using standard metrics including accuracy, precision, recall, and F1-score. In addition, a user-friendly interface is developed to facilitate real-time prediction and improve accessibility. Experimental results demonstrate that the proposed model achieves high accuracy and reliability in early-stage detection. The system can assist healthcare professionals in decision-making and contribute to improved awareness and timely intervention. The study highlights the potential of machine learning techniques in enhancing traditional diagnostic processes and provides a scalable approach for early Alzheimer’s disease prediction.
Abstract
Design And Development of Thermoelectric Generator For Waste Heat Energy Harvesting
Lokesh Ramawadh Chauhan, Nainiksha Anil Dhoke, Riya Manoj Kalamkar, Divya Shailendra Patil, Kashish Manoj Mahato, Prof. Rajendra Bhombe, Prof. Priyanka Rajput
DOI: 10.17148/IJIREEICE.2026.14444
Abstract: This paper presents the design and development of a thermoelectric generator (TEG) prototype for waste heat energy harvesting. The system operates on the principle of the Seebeck Effect, where a voltage is generated due to the temperature difference across two surfaces. A thermoelectric module is placed between a hot surface and a cooled surface to create a temperature gradient, enabling direct conversion of heat energy into electrical energy.
The proposed system consists of a heat source, TEG module, heat sink for cooling, DC-DC boost converter, and an energy storage unit. Since the generated voltage is low and unstable, a boost converter is used to increase and regulate the output voltage for practical applications. The performance of the system is evaluated under different temperature conditions by measuring output voltage and power.
Experimental results indicate that the output voltage and power increase with an increase in temperature difference, and efficient cooling enhances overall system performance. The developed prototype provides a simple, reliable, and eco- friendly solution for harvesting waste heat energy. The system can be applied in low-power applications such as IoT devices and remote sensors.
The proposed system consists of a heat source, TEG module, heat sink for cooling, DC-DC boost converter, and an energy storage unit. Since the generated voltage is low and unstable, a boost converter is used to increase and regulate the output voltage for practical applications. The performance of the system is evaluated under different temperature conditions by measuring output voltage and power.
Experimental results indicate that the output voltage and power increase with an increase in temperature difference, and efficient cooling enhances overall system performance. The developed prototype provides a simple, reliable, and eco- friendly solution for harvesting waste heat energy. The system can be applied in low-power applications such as IoT devices and remote sensors.
Abstract
An Advanced Explainable Deep Learning Approach with Grad-CAM and Post-Hoc Analysis for Secure QR Code Threat Detection
Y Roshni and Dr. Golda Dilip
DOI: 10.17148/IJIREEICE.2026.14445
Abstract: The rapid adoption of QR codes in digital payments, authentication, and information sharing has increased cybersecurity risks, particularly QR-code-based phishing attacks known as quishing. Traditional machine learning methods rely on handcrafted features and often lack interpretability, limiting their effectiveness against evolving threats. This paper proposes an explainable deep learning–based framework for malicious QR code detection using a convolutional neural network (CNN) to classify QR code images as benign or malicious. To improve transparency, Grad- CAM is applied to highlight important regions influencing model decisions, while a post-hoc URL analysis module examines protocol usage, domain age, and suspicious lexical patterns to validate predictions. Experimental results demonstrate high detection accuracy along with meaningful visual and analytical explanations, making the proposed approach suitable for real-world cybersecurity applications.
Keywords: QR Code Security, Malicious QR Detection, Deep Learning, CNN, Grad-CAM, Cybersecurity, URL Analysis, Image Classification
Keywords: QR Code Security, Malicious QR Detection, Deep Learning, CNN, Grad-CAM, Cybersecurity, URL Analysis, Image Classification
Abstract
CONFIDENCE-GUIDED MULTI-AGENT LLM FRAMEWORK FOR CLINICAL DECISION SUPPORT IN OPHTHALMOLOGY USING BIOMEDICAL RAG AND WEB INTELLIGENCE
Mrs.D. Archana¹, P. Tanmai Sai², S. Bhavana³, Y. Suhitha Priya´, G. Siri Saranyaµ
DOI: 10.17148/IJIREEICE.2026.14446
Abstract: Eye diseases such as conjunctivitis, glaucoma, cataract, diabetic retinopathy, and optic neuritis are becoming increasingly common worldwide. If not detected at an early stage, these conditions can lead to serious vision impairment or even permanent blindness. In many parts of the world, especially in rural and remote areas, people face significant difficulty in accessing a qualified eye specialist on time. This paper presents a Confidence-Guided Multi-Agent Large Language Model (LLM) Framework developed for clinical decision support in ophthalmology. The system accepts symptom descriptions from patients written in plain English and processes them through a pipeline of five intelligent agents. Each agent performs a specific role, including symptom analysis, disease retrieval using Retrieval-Augmented Generation (RAG) with FAISS, web-based validation using DuckDuckGo, and final decision synthesis. A unique confidence scoring mechanism is incorporated, which combines the RAG similarity score and the web trust score to give users a reliable percentage-based prediction. The system is built using Python and Streamlit, making it accessible through any standard web browser without requiring specialized software. Experimental results demonstrate that the system correctly identifies common eye diseases with confidence scores above 80% for valid inputs, and appropriately rejects unrelated inputs. This system is intended for academic demonstration and can serve as a supportive tool for telemedicine platforms and rural healthcare centers.
Abstract
Barriers Against the Use of EV-Review
Prof.Akash A. Gophane, Sampada V. Chaure, Yashasvi G. Rohane, Kuldeepak B. Pise
DOI: 10.17148/IJIREEICE.2026.14447
Abstract: The transition toward electric vehicles (EVs) is widely recognized as a crucial step in reducing greenhouse gas emissions and promoting sustainable transportation. Despite continuous advancements in technology and increasing policy support, the adoption rate of EVs remains slower than expected, particularly in developing economies. This paper presents a systematic review of existing literature to identify and analyze the major barriers limiting EV adoption.The study synthesizes findings from empirical, analytical, and modeling-based research and categorizes the barriers into five key dimensions: economic, technical, infrastructural, policy and regulatory, and behavioral. The review highlights that high initial costs, limited charging infrastructure, range anxiety, policy inconsistencies, and lack of consumer awareness are among the most critical challenges. Furthermore, these barriers are highly interconnected and vary across regions. By providing a consolidated and structured understanding of EV adoption challenges, this study offers valuable insights for policymakers, researchers, and industry stakeholders to design effective strategies that can accelerate the transition toward sustainable electric mobility.
Keywords: Electric Vehicles, EV Adoption, Charging Infrastructure, Sustainable Transportation, Systematic Review, Adoption Barriers
Keywords: Electric Vehicles, EV Adoption, Charging Infrastructure, Sustainable Transportation, Systematic Review, Adoption Barriers
Abstract
Electrical Vehicle Battery Health and Management: A Review for Helpful Research
Dr. G. Naveen Kumar
DOI: 10.17148/IJIREEICE.2026.14448
Abstract: The growing demands for sustainable transportation and continuous improvements in battery technologies have accelerated the adoption of electric vehicles (EVs). A key component within EVs is the Battery Management System (BMS), which is responsible for maintaining the safety, efficiency, and dependability of battery operation. This report provides a detailed survey of existing research on BMS design, including its architecture, core functionalities, state estimation methods, thermal regulation, and recent technological advancements. Additionally, it identifies major challenges and outlines potential directions for future research in this field. New technologies are not introduced but scope for research in this direction is discussed.
Keywords: Battery Management System, Battery Health, Electric Vehicles, Thermal Control.
Keywords: Battery Management System, Battery Health, Electric Vehicles, Thermal Control.
Abstract
SOLAR ENERGY MONITORING SYSTEM
Ankam Varshini, Ganta Praveen Sai, Goli Harika Lakshmi,Kunchanapalli Sri Anjhanee, Vasamsetti Lakshmi Sri Sai Devika, Dr. M. Srinivasa Rao
DOI: 10.17148/IJIREEICE.2026.14449
Abstract: The rapid growth of renewable energy systems has increased the need for efficient monitoring and performance evaluation of solar photovoltaic installations. Traditional solar monitoring techniques rely on manual data collection and limited analysis, leading to inefficiencies, delayed fault detection, and reduced system performance. This paper presents Solar Energy Monitoring System, a web-based solar energy monitoring and performance analytics system designed to analyze, visualize, and manage solar power generation data. The system is developed using Python with the Flask framework for backend processing, HTML, CSS, and JavaScript for frontend design, and MySQL for Database management. Solar Energy Monitoring System enables secure user authentication, centralized data storage, real-time performance visualization, and automated report generation. The proposed system improves accuracy, efficiency, and scalability in solar energy monitoring and provides a foundation for future real-time IoT integration.
Keywords: Solar Energy Monitoring, Photovoltaic Systems, Web Application, Python Flask, MySQL, Renewable Energy Analytics
Keywords: Solar Energy Monitoring, Photovoltaic Systems, Web Application, Python Flask, MySQL, Renewable Energy Analytics
Abstract
Perceived Role of AI-Based tools in enhancing employability and Job Readiness among MBA students in Bangalore.
Badri Narayanan KS, Dr. Mahalakshmi S
DOI: 10.17148/IJIREEICE.2026.14450
Abstract: The growing use of Artificial Intelligence (AI) is changing how MBA students build skills and prepare for jobs. This study looks at how students in Bangalore perceive the role of AI-based tools in improving their employability and job readiness. A quantitative approach was used, with data collected from 186 students through a structured questionnaire. The study focuses on how students use AI tools, how helpful they find them, and how easy these tools are to use. The findings show that AI tools support the development of important skills such as analytical thinking, communication, and problem-solving. At the same time, their impact on overall job readiness is limited, as practical experience and real-world exposure still play a major role. The study also highlights concerns such as over-reliance on AI and lack of proper training. Overall, AI tools are useful, but they need to be used carefully and with proper guidance in management education.
Abstract
Dynamic Voltage Regulation using Solar Powered Synchro-Inverter
Chitra S, Archana Prasadini G S
DOI: 10.17148/IJIREEICE.2026.14451
Abstract: Voltage variations are a common issue in modern electrical power systems due to fluctuating load demand and increasing penetration of renewable energy sources. These variations can lead to voltage sag, swell, and power quality disturbances, which may affect the performance of sensitive electrical equipment. Maintaining a stable voltage level at the load terminal is therefore essential for reliable and efficient system operation.
This work presents a dynamic voltage regulation method using a solar powered Synchro-Inverter. The proposed system utilizes solar photovoltaic energy as the primary power source and incorporates a Maximum Power Point Tracking (MPPT) controller to ensure maximum power extraction from the PV array. A DC–DC boost converter is employed to regulate the DC link voltage, while a voltage-controlled Synchro-Inverter converts the DC output into AC power.
The inverter operates similar to a synchronous generator and dynamically adjusts its output voltage based on system conditions. A PID controller is implemented to maintain a constant load voltage by compensating voltage disturbances in real time. The inverter is connected to the distribution line through an injection transformer which enables effective voltage compensation.
The proposed system is modelled and simulated using MATLAB/Simulink. Simulation results demonstrate that the system effectively mitigates voltage fluctuations, improves voltage stability, and enhances the power quality of the distribution network.
Keywords: Dynamic Voltage Regulation, Solar PV, Synchro-Inverter, MPPT, PID Controller, Power Quality.
This work presents a dynamic voltage regulation method using a solar powered Synchro-Inverter. The proposed system utilizes solar photovoltaic energy as the primary power source and incorporates a Maximum Power Point Tracking (MPPT) controller to ensure maximum power extraction from the PV array. A DC–DC boost converter is employed to regulate the DC link voltage, while a voltage-controlled Synchro-Inverter converts the DC output into AC power.
The inverter operates similar to a synchronous generator and dynamically adjusts its output voltage based on system conditions. A PID controller is implemented to maintain a constant load voltage by compensating voltage disturbances in real time. The inverter is connected to the distribution line through an injection transformer which enables effective voltage compensation.
The proposed system is modelled and simulated using MATLAB/Simulink. Simulation results demonstrate that the system effectively mitigates voltage fluctuations, improves voltage stability, and enhances the power quality of the distribution network.
Keywords: Dynamic Voltage Regulation, Solar PV, Synchro-Inverter, MPPT, PID Controller, Power Quality.
Abstract
SRMunch: A Data-Driven Institutional Food Operational Efficiency System
Shreya Sathapathi, Rakshitha S N, Yashvitha J, Dr. Durgadevi P
DOI: 10.17148/IJIREEICE.2026.14452
Abstract: Campus canteens at large institutions routinely grapple with unpredictable demand patterns, excess food production, and a near-total absence of cohesive digital management infrastructure. To address these challenges, this paper introduces SRMunch, a web-based operational management platform developed specifically for SRM Institute of Science and Technology. The system is constructed using the Django web framework backed by a MySQL database, and extends access across multiple stakeholder categories including students, faculty, non-teaching staff, food vendors, and system administrators. SRMunch consolidates several capabilities that have historically remained fragmented: contactless digital ordering, live inventory surveillance, food waste quantification, pattern-driven analytics, and a customer loyalty mechanism - all within a single deployable application. The platform eliminates dependence on manual queue-based workflows and places actionable operational data at administrators' fingertips to curb overproduction and end-of-day surplus. An examination of twenty closely related publications establishes the novelty of SRMunch's integrated scope. Experimental validation across five completed development milestones confirms complete end-to-end functional coverage.
Keywords: food management system, canteen digitization, Django, inventory tracking, food waste reduction, demand forecasting, data-driven decision support, institutional catering, multi-role web application.
Keywords: food management system, canteen digitization, Django, inventory tracking, food waste reduction, demand forecasting, data-driven decision support, institutional catering, multi-role web application.
Abstract
SENTINEL: Multimodal AI Framework for Contract Risk Analysis and Negotiation Strategy Generation
S Yashwant, Surya Sivakumar, J Joshua Haniel, Niranjana S
DOI: 10.17148/IJIREEICE.2026.14453
Abstract: Manual examination of contractual documents demands extensive human effort and often leads to inconsistent identification of risk-bearing clauses. This work introduces SENTINEL, a multimodal analytical framework designed to automate risk assessment and generate context-aware negotiation recommendations from legal agreements. The system integrates a hybrid OCR pipeline combining CRNN and rule-based recognition for text extraction, followed by clause segmentation and domain-adapted LegalBERT classification for risk prediction. A retrieval-augmented mechanism further enhances the system by leveraging semantically similar historical clauses to generate negotiation suggestions using large language models. Evaluation conducted on 847 real-world contracts demonstrates 87.3% classification accuracy and strong agreement with expert assessments (r = 0.81). The results indicate that combining structured document understanding with retrieval-driven generation significantly improves efficiency and supports informed decision-making in legal workflows.
Keywords: Contract Analysis, Legal AI, Risk Classification, Retrieval-Augmented Generation, Natural Language Processing, LegalBERT, Document Intelligence
Keywords: Contract Analysis, Legal AI, Risk Classification, Retrieval-Augmented Generation, Natural Language Processing, LegalBERT, Document Intelligence