← Back to Archives
Keywords: Swarm Intelligence, starlings, Optimisation algorithm, bio-inspired, Environment dynamic.
VOLUME 15, ISSUE 8, AUGUST 2026
Publication in progress..
Secure AI Orchestration for Construction Collaboration Platforms
Mohammed Imran Ahmed
XAI-IDS: Interpretable Machine Learning for Real-Time Multi-Class Intrusion Detection in Internet of Things Environments
Amith G M, Yoga Lakshmi M, Shailaja G S, Rihan Ahamed, Mrs. Roopa Patrimath
AuraVoice: An End-to-End Multilingual Voice Cloning and Video Dubbing Pipeline Integrating OpenVoice V2, Whisper, and MeloTTS
Sagar M, Rakshith G C, Dr. Dharani N V
Smart SQL Injection Attack Detection Using Hybrid Deep Neural Network
Mrs. T Geetha M.E., MBA., (Ph. D)*, Manjuparkavi S, Mithunapriya M, Navitha G, Rajeshwari R
Mitro: Calibrated Multimodal Gesture, Voice, and LLM Control
Srijan Mani Tripathi, Vyom Pandey, Mohammad Rayyan Basha, Dr. Beenarani Manoj
CloudOpt-AI: A Predictive FinOps Framework for Intelligent Cloud Cost Optimization Using Machine Learning and Multi-Cloud Resource Scheduling
Prof. C HemaPrabha, Lakshmisha N, Rudraprasad S
BLIP-Driven Multi-modal Deep Learning Framework for Temporal Depression Prediction from Social Media Posts
Dr Bharathi M P, Amulya M, Bhumika U
Mathematical Modeling of Plasma and Cellular Transport in Pulsatile Arterial Blood Flow
Vishanu Mohan Sharma*, Dr. Dileep Singh
Energy-Efficient Federated Learning Framework for Personalized Human Activity Recognition Using Smartphone Sensor Data
Prof. Priyanka Mohan, Miss Yashaswini V, Miss Bhavyashree A
A Comprehensive Survey of an Agentic AI Powered Multimodal RAG Learning System
Mrs. Supriya, Likitha R, Madhan N, Raksha D O, Shylashree E
Neuro-EvoSwarm Optimizer (NESO): A Hybrid Deep Learning, Genetic Algorithm and Particle Swarm Optimization Framework for Seasonal Multicropping Strategy Optimization under Drip Irrigation
Mrs.N. Amirtha Gowri M.Sc., M.Phil., (Ph.D), Dr. R. Nandhakumar, MCA., M.Phil., MBA., Ph.D., SET
Enhancement of Vehicle Number Plate Recognition System
Manpreet Kaur, Jatinder Singh Saini, Sandeep Kaur Dhanda
Age-Aware Multimodal Web-Based Framework for Parkinson's Disease Detection Using Handwriting and Voice Analysis
Korupala Pavani, Dr G Suvarna Kumar
An Intelligent Digital Twin Framework with Biomarker-Aware XGBoost for Ovarian Cancer Prediction
Bhavana B R*, Sindhu A G and Akash Gowda S G
AI-Powered Resume Analyzer with Intelligent Job Recommendation and Skill Gap Analysis: A Survey of NLP and Machine Learning Approaches
M B Chethana, P Harshitha, R Manvitha, Mrs. Sougandika Narayan
A Comparative Study of Classical and Deep Learning Approaches for Non-Uniform Motion Blur Restoration
Sarthak Agarwal, Ms.Charulatha RT
AI-Based Disaster Relief Demand Prediction System
M. Sri Chaitanya Varma, Pyla Jyothi, N.J.V.D. Jagadeesh, N. Enosh Raju, K. Jagadeesh, G. Omkar Venkat
Explainable AI-Based Customer Churn Prediction Using Machine Learning and SHAP Analysis
Dr. Bharathi M P, Narendra C, Abhishekaraddi Maraddi
A Comparative Analysis of Starling Swarm Algorithm for Optimisation in Dynamic Environment
Karatu Musa Tanimu, Wei Pang, and George M. Coghill
Automated Business Card Digitization and Smart Contact Follow Up Using AI
Basavaraj R Palled, Shobha Rani B R
SECURE API COMMUNICATION IN CLOUD APPLICATIONS
Mr. Naveen J, Manjunatha H, Sojo T Johnson
Reducing Round-Trip Time in Cloud Environments Using Software-Defined Networking with Intelligent Prefetching and Caching: A Review
Chandu H M, Nagendra Swamy M S, Chandan Hegde
Smart Underground Cable Fault Detector
Mrs. Mamatha N, B Kiran Kumar, Charan K V, Bharath Gowda, Jnanesh R
Cloud Resource Allocation and Scheduling: A Comprehensive Study
Srinivasa G., KareemPasha, Chandan Hegde
Short-Term Price Movement Prediction Using Order Book Heatmap Vision
R.T.N Sura Reddy, Sharavana Ragav S, Ms. Charulatha R .T
FrameDeblur: NAFNet-Based Selective Video Frame Deblurring and Restoration Framework
S.Roshan Pranao, Harihara Balan S, Yelavarthi Sai Dheeraj, Ms. Charulatha R T
ARTIFICIAL INTELLIGENCE IN TRANSFORMING THE GIG ECONOMY
Chandrakala G S
PaperPilot AI: An AI-Powered Multimodal Research Paper Mentor for Intelligent Academic Content Understanding, Mathematical Equation Decoding, and Comparative Literature Analysis
George, Dr. Beenarani Manoj
The influence of blockchain technology and high- involvement human resource practices on supply chain resilience and organizational performance
Mohammad Ali Yousef Yamin, Abd Arahman Hussain Al Amri, Safar Said Alamri
A COMPREHENSIVE SURVEY OF AN AQUA CYCLE INTELLIGENCE: AI BASED SYSTEM FOR RO REJECT WATER OPTIMIZATION AND SUSTAINABILITY
Ankit, Ayush Gupta, Darshan K S, Dharmaraj, Mrs.Supriya
Atlas AI: Watching the planet evolve through AI-Powered Satellite Intelligence
Deekshith M, Dhruv Kaushik, Likhith Gowda A K, Varun M, Asha K H
Design and FPGA Implementation of a Low- Power Dynamic Clock-Gated Data Processing Unit
Macherla Sunitha, Dr B Ramprasad
DEEPFAKE IMAGE DETECTION USING DEEP LEARNING
Pavaman D.K, Yashodara.R, Prakash Naik, Puneeth.A, Divyashree.K, Ullas Reddy .G
A Comparative Study of Lightweight Machine Learning Models for Detecting Suspicious Network Traffic
Dadavali S. P.
Deep Learning-Based Image Classification and Real-Time Object Detection Using CNN and YOLOv8
Madan S, Prem Singh M, T Chandraiah, Poornima B H, Basavanna M
A Robust Multi-Stage Framework for Noise-Resilient Bone Tumor Detection, Segmentation, and Classification
V.Dineshkumar, Dr. N. Kamaraj
EXTENSIBLE STORAGE MANAGEMENT AND DATA AVAILABILITY TECHNIQUES
Anozie, E. L., Okoronkwo, M. C., Emesiani, J. U., Eze, R. C.
Reskilling the Banking Workforce in the Age of Artificial Intelligence: Emerging HRM Priorities and Future Skill Requirements in the Indian BFSI Sector
Dr. Avadhesh Vyas, Mr. Virendra Choudhary
Historical Handwritten Document Age Identification using DCT-based Frequency Analysis
Dr. Pushpalata Gonasagi
Abstract
Secure AI Orchestration for Construction Collaboration Platforms
Mohammed Imran Ahmed
DOI: 10.17148/IJARCCE.2026.15801
Abstract: AI agents are being used in construction collaboration platforms increasingly to perform tasks related to planning, safety, subcontracting, document validation, and inspection among others. However, highly sensitive data such as contractual details, finances, personnel information, and design layouts are processed through collaboration platforms. In order to facilitate coordinated functions among different AI agents without compromising their integrity, security, and regulatory compliance, this paper proposes a Secure AI Orchestration Framework (SAIOF) that would support collaboration platforms development. The SAIOF framework encompasses role-based access controls, multi-agent coordination through policies, agent orchestration federations, and encrypted communication among AI agents. The proposed model will allow inspectors, subcontractors, and project managers to collaborate securely without unnecessarily exposing sensitive or confidential data. As can be seen from the simulation experiment of a construction collaboration workflow, there is less risk of data leaks, greater efficiency, and greater workflow transparency with this orchestration approach. From the analysis of experimental results, it follows that secure orchestration improves the accuracy of coordination actions, limits malicious attempts to gain unauthorized access to sensitive information, and increases scalability in the deployment of AI systems in smart buildings.
Keywords: AI orchestration, intelligent coordination, safe multi-agent systems, federated AI, construction cybersecurity, access control, and privacy-preserving AI
Keywords: AI orchestration, intelligent coordination, safe multi-agent systems, federated AI, construction cybersecurity, access control, and privacy-preserving AI
Abstract
XAI-IDS: Interpretable Machine Learning for Real-Time Multi-Class Intrusion Detection in Internet of Things Environments
Amith G M, Yoga Lakshmi M, Shailaja G S, Rihan Ahamed, Mrs. Roopa Patrimath
DOI: 10.17148/IJARCCE.2026.15802
Abstract: Existing IoT Intrusion Detection Systems (IDS) achieve competitive accuracy but are opaque black-box models, relying on single imbalanced datasets, centralising raw traffic, and incompatible with edge hardware. This paper presents XAI-IDS, a seven-layer federated explainable IDS resolving all four limitations simultaneously. The system trains LightGBM and a CNN-LSTM hybrid on CICIOT2023 (80/10/10 split, 5-fold CV) and evaluates without retraining on ToN-IoT and NSL-KDD across 33 attack types. LightGBM achieves macro-F1=0.9412/0.9138/0.9267 on CICIOT2023/ToN-IoT/NSL-KDD; CNN-LSTM achieves 0.9534. FPR=3.12% vs 8.92% (Extra Trees [4], p<0.001, - 5.80 p.p.). SMOTE-ENN recovers normal-class F1 from 0.6312 to 0.9134. TreeSHAP provides exact attribution in 4.3Β±0.4 ms; LIME cross-validates at 83.2% top-5 agreement; feature ablation confirms faithfulness (ΞAcc=19.3Β±1.7 p.p.). An LLM explanation layer raises non-technical comprehension from 31.2% to 84.3% (t(19)=14.23, p<0.001, n=20, ethics-approved). FedProx ΞΌ=0.01 achieves F1 within 1.32% of centralised baseline in 7 rounds under DP-SGD Ξ΅=5. ADWIN recovers from 93.7% of drift events in 47.3Β±4.8 s. The ONNX INT8 LightGBM (14.2 MB) achieves 1.83 ms inference and +0.4 W power draw on Raspberry Pi 4. All improvements: p<0.001, Bonferroni-corrected. Complete hyperparameters (Table IX), 78-feature schema (Table X), and edge metrics (Table XI) provided for full reproducibility.
Keywords: IoT Security, IDS, XAI, SHAP, LIME, LightGBM, CNN-LSTM, FedProx, Differential Privacy, SMOTE- ENN, ADWIN, Concept Drift, ONNX, LLM.
Keywords: IoT Security, IDS, XAI, SHAP, LIME, LightGBM, CNN-LSTM, FedProx, Differential Privacy, SMOTE- ENN, ADWIN, Concept Drift, ONNX, LLM.
Abstract
AuraVoice: An End-to-End Multilingual Voice Cloning and Video Dubbing Pipeline Integrating OpenVoice V2, Whisper, and MeloTTS
Sagar M, Rakshith G C, Dr. Dharani N V
DOI: 10.17148/IJARCCE.2026.15803
Abstract: Creating multilingual voice content today typically requires re-recording narration in every target language or stitching together separate tools for transcription, translation, and speech synthesis. This paper presents AuraVoice, an end-to-end web-based platform that unifies voice cloning, cross-lingual speech translation, and automated video dubbing in a single pipeline. The system combines OpenVoice V2 for zero-shot tone-color conversion, faster-whisper for automatic speech recognition, MeloTTS for multilingual base speech synthesis, and FFmpeg for audio/video processing, orchestrated through a FastAPI backend with a lightweight browser-based frontend. Given a short reference audio sample, AuraVoice extracts speaker tone characteristics and reproduces them in synthesized speech across six languages while preserving the original speaker's identity. The video dubbing module extends this pipeline to full videos by extracting the source audio, translating and re-synthesizing it in the target language, and re-muxing it with the original visual track. We describe the system architecture, module-level implementation, and lazy-loading/caching strategy used to manage GPU memory across multiple deep learning models, and discuss the functional testing performed on each module. The platform demonstrates that recent advances in zero-shot voice cloning and open automatic speech recognition can be composed into a practical, integrated tool for multilingual content creation, reducing the manual effort required compared to disjoint single-purpose tools.
Keywords: Voice Cloning, Tone-Color Conversion, OpenVoice, Text-to-Speech, Speech Translation, Video Dubbing, Whisper, FastAPI, Speech Synthesis
Keywords: Voice Cloning, Tone-Color Conversion, OpenVoice, Text-to-Speech, Speech Translation, Video Dubbing, Whisper, FastAPI, Speech Synthesis
Abstract
Smart SQL Injection Attack Detection Using Hybrid Deep Neural Network
Mrs. T Geetha M.E., MBA., (Ph. D)*, Manjuparkavi S, Mithunapriya M, Navitha G, Rajeshwari R
DOI: 10.17148/IJARCCE.2026.15804
Abstract: SQL Injection attacks are one of the most critical and frequently occurring security threats in modern web applications. These attacks exploit vulnerabilities in input handling to inject malicious SQL statements, enabling attackers to bypass authentication, retrieve confidential data, modify database contents, or even gain complete control over backend systems. Traditional defense mechanisms such as manual input validation, rule-based filtering, and signature-based intrusion detection systems are limited in their ability to detect complex, obfuscated, and previously unseen attack patterns. This project proposes a Smart SQL Injection Attack Detection System using a Hybrid Deep Neural Network (CNN + LSTM) to effectively identify both known and unknown SQL injection attacks. Incoming SQL queries are preprocessed through tokenization and encoding, converted to numerical form using TF-IDF, and refined with Chi-Square feature selection. CNN layers capture local attack signatures while LSTM layers model sequential and contextual relationships among query tokens, jointly improving learning capability and detection accuracy. The model is trained and evaluated on benchmark and real-world SQL injection datasets using accuracy, precision, recall, and F1-score. Experimental results show detection accuracy consistently above 95%, with lower false positive rates than traditional and single-model detection techniques, and the system is suitable for real-time deployment in existing web application security infrastructures.
Keywords: SQL Injection, Cybersecurity, Deep Learning, Hybrid Neural Network, CNN, LSTM, Web Security, Attack Detection.
Keywords: SQL Injection, Cybersecurity, Deep Learning, Hybrid Neural Network, CNN, LSTM, Web Security, Attack Detection.
Abstract
Mitro: Calibrated Multimodal Gesture, Voice, and LLM Control
Srijan Mani Tripathi, Vyom Pandey, Mohammad Rayyan Basha, Dr. Beenarani Manoj
DOI: 10.17148/IJARCCE.2026.15805
Abstract: This paper presents Mitro, a desktop human-computer interaction (HCI) system that fuses three input modalities into a single interaction pipeline: (1) computer-vision-based hand-gesture recognition for direct manipulation of the operating system (cursor control, clicks, scroll, volume, and brightness); (2) speech-based natural-language commanding through automatic speech recognition (ASR) and text-to-speech (TTS); and (3) a large language model (LLM) reasoning layer that replaces brittle keyword-matched command dispatch with open-domain natural-language understanding and tool/function calling. Mitro is built as an upgrade to an existing rule-based gesture-and-voice controller that suffered from three concrete limitations: order-sensitive substring-matched command routing, a single hardcoded audio energy threshold that did not generalize across acoustic environments, and a single hardcoded gesture-classification threshold with a fragile single-frame-reset debounce scheme. This work addresses all three limitations with a unified calibrate-once-then-lock design pattern applied consistently across the audio and vision pipelines, and generalizes command handling from rule-based matching to LLM-mediated tool dispatch while preserving deterministic, network- independent fast paths for safety-critical control flow. Gesture-state changes, voice transcripts, and LLM responses are surfaced into a single chronological interaction log, providing an observable, demonstrable trace of the system's multimodal behavior. Section IV outlines the evaluation methodology used to measure calibration benefit, gesture- smoothing benefit, command-routing robustness, and end-to-end latency, together with the qualitative findings obtained during development; the corresponding quantitative results are to be reported once the described experiments are executed on the deployed system.
Keywords: Multimodal Human-Computer Interaction, Gesture Recognition, Speech Recognition, Large Language Models, Tool Calling, Adaptive Calibration, Temporal Smoothing, Cross-Modal Fusion, MediaPipe, Groq
Keywords: Multimodal Human-Computer Interaction, Gesture Recognition, Speech Recognition, Large Language Models, Tool Calling, Adaptive Calibration, Temporal Smoothing, Cross-Modal Fusion, MediaPipe, Groq
Abstract
CloudOpt-AI: A Predictive FinOps Framework for Intelligent Cloud Cost Optimization Using Machine Learning and Multi-Cloud Resource Scheduling
Prof. C HemaPrabha, Lakshmisha N, Rudraprasad S
DOI: 10.17148/IJARCCE.2026.15806
Abstract: Organizations increasingly rely on cloud platforms to store data, run applications, and manage IT infrastructure, drawn by the flexibility and on-demand scalability these platforms offer. That flexibility comes at a price, however: keeping cloud spending under control is one of the more persistent operational headaches for teams running workloads at scale. Resources sit over-provisioned or idle far more often than they should, and the resulting waste translates directly into avoidable expense. Bringing smarter, more automated cost control into everyday cloud operations is no longer optional as usage keeps climbing. This paper introduces CloudOpt-AI, a predictive FinOps framework that pairs machine learning with multi-cloud resource scheduling to tackle exactly that problem. The framework studies past usage patterns to project near-term resource needs and then allocates capacity accordingly. It also coordinates workload placement across several cloud providers at once, so that organizations can pick cost-effective resources without sacrificing performance or reliability. Demand forecasting, automated allocation, and FinOps discipline are woven together so that CloudOpt-AI supports faster, better-informed decisions and squeezes more value out of existing cloud capacity. The overarching goal is threefold: cut operational spend, raise resource efficiency, and make cloud cost governance genuinely manageable rather than a manual chore. Taken together, the results point to how predictive analytics and intelligent scheduling can help organizations strike a workable balance between cost and performance in today's cloud environments.
Keywords: Cloud Computing, Cloud Cost Optimization, FinOps, Machine Learning, Multi-Cloud Computing, Resource Allocation, Resource Scheduling, Demand Forecasting, Cost-Aware Computing, Workload Management.
Keywords: Cloud Computing, Cloud Cost Optimization, FinOps, Machine Learning, Multi-Cloud Computing, Resource Allocation, Resource Scheduling, Demand Forecasting, Cost-Aware Computing, Workload Management.
Abstract
BLIP-Driven Multi-modal Deep Learning Framework for Temporal Depression Prediction from Social Media Posts
Dr Bharathi M P, Amulya M, Bhumika U
DOI: 10.17148/IJARCCE.2026.15807
Abstract: Depression remains one of the most widespread mental health conditions worldwide, and social media has become an unusually rich window into how people express emotional and behavioral change. Most existing detection systems however still look at a single channel either the wording of a post or a single accompanying image and largely ignore how a personβs posting behavior shifts over time. This paper addresses that gap with a temporal, multi-modal pipeline that reads depression risk not from one post in isolation but from a sequence of a userβs posts, combining what is written, what is pictured, the emotional tone of both, and how these change across time. Images are converted into descriptive captions using BLIP, which are then encoded alongside raw visual features and a set of behavioral and temporal signals processed through a GRU. When these three streams text, image, and temporal-behavioral are fused, the resulting model reaches 71% classification accuracy. The results support the view that behavioral and temporal cues, even when individually weak predictors, meaningfully sharpen a multi-modal systemβs ability to separate depressed from non-depressed posting patterns.
Keywords: Depression Detection; Social Media Analytics; Multi-modal Fusion; Temporal Behaviour Modelling; Image Captioning; Sentiment Analysis; BERT; GRU.
Keywords: Depression Detection; Social Media Analytics; Multi-modal Fusion; Temporal Behaviour Modelling; Image Captioning; Sentiment Analysis; BERT; GRU.
Abstract
Mathematical Modeling of Plasma and Cellular Transport in Pulsatile Arterial Blood Flow
Vishanu Mohan Sharma*, Dr. Dileep Singh
DOI: 10.17148/IJARCCE.2026.15808
Abstract: Blood is a biologically-complex two-phase fluid consisting of suspended cellular (blood) components and plasma. Blood's properties of transport affect the function of the heart while pumping under pulsatile flow conditions. A general and total analytical model for plasma transport and blood-cell transport through a cylindrical artery experiencing pulsatile pressure drops is developed. Equations used for developing the model are the incompressible Navier-Stokes equations with viscosity depending on hematocrit (the volume fraction of red blood cells), a convective- diffusive-migrative equation describing cellular transport, and an advective-diffusive equation describing plasma solute transport. Analytical solutions for the parameters are determined by separating variables and solving them using the axisymmetric fully developed pulsatile flow analytical solution of Womersley. In addition to giving the spatial distributions of velocity, hematocrit, plasma concentration, cellular concentration, wall shear stress, and volumetric flow rate the solutions provide information concerning changes that occur due to varying hematocrit levels. Numerical simulation results indicate that higher hematocrit values result in increased effective viscosity of blood. Increased blood viscosity causes decreased velocities in the direction of flow, decreases the maximum flow rates occurring during the pulse cycle, and increases frictional resistance to flow. Additionally, numerical simulation results demonstrate the existence of a "cell-rich core" along the centerline of vessels and a "plasma-depleted zone" near vessel walls. These zones develop because shear-induced migration of red blood cells towards the centerline. Plasma concentration decreases continuously downstream from the inlet due to a combination of advection, diffusion, and wall transport.
Keywords: Pulsatile blood flow, Plasma transport, Cellular transport, Hematocrit, Mathematical modeling, Non- Newtonian blood
Keywords: Pulsatile blood flow, Plasma transport, Cellular transport, Hematocrit, Mathematical modeling, Non- Newtonian blood
Abstract
AI Fitness and E-Commerce Platform
Anvith K S, Shobha Rani B R
DOI: 10.17148/IJARCCE.2026.15809
Abstract: The rapid growth of digital technology has transformed the way people manage their health and fitness. While many online platforms allow users to purchase fitness supplements, they often fail to provide personalized guidance based on an individual's health condition and fitness goals. This project, AI Fitness and E-Commerce Platform (POP GYM), addresses this gap by combining an online supplement store with an AI powered diet recommendation system.The platform enables users to create an account, browse and purchase fitness supplements, manage their shopping cart, and receive customized diet plans based on personal details such as age, height, weight, gender, and fitness objectives. The diet recommendations are generated using Google's Gemini AI, making them more relevant and personalized for each user. In addition to user features, the system includes an admin dashboard that allows efficient management of products, users, and customer orders.The application is developed using React.js for the frontend, Node.js and Express.js for the backend, and MySQL for database management. This combination ensures a responsive, secure, and scalable web application. By integrating e-commerce functionality with AI-driven health recommendations, the platform offers users a convenient and intelligent solution to support both their fitness shopping and nutritional planning.The project demonstrates how Artificial Intelligence can enhance user experience by providing personalized recommendations while simplifying the management of fitness-related products and services. It also provides a strong foundation for future enhancements such as online payment integration, workout tracking, wearable device connectivity, and mobile application support.
Keywords: Artificial Intelligence, Fitness Management, E-Commerce Platform, Personalized Diet Recommendation, Google Gemini API, React.js, Node.js, Express.js, MySQL, Web Application, Health and Nutrition, Supplement Management, User Authentication, Admin Dashboard, Responsive Web Development.
Keywords: Artificial Intelligence, Fitness Management, E-Commerce Platform, Personalized Diet Recommendation, Google Gemini API, React.js, Node.js, Express.js, MySQL, Web Application, Health and Nutrition, Supplement Management, User Authentication, Admin Dashboard, Responsive Web Development.
Abstract
Energy-Efficient Federated Learning Framework for Personalized Human Activity Recognition Using Smartphone Sensor Data
Prof. Priyanka Mohan, Miss Yashaswini V, Miss Bhavyashree A
DOI: 10.17148/IJARCCE.2026.15810
Abstract: Imagine what your smartphone already knows about you based on how you move. Each step, each climb up the stairs, each slouch into the chair β it is all recorded by your smartphone sensors. To unlock the immense possibilities of using this data for the identification of human actions in real-time, while keeping your privacy intact is an extremely promising endeavour . In this paper, EE-FL-HAR is presented, a system that allows training activity recognition models right inside the smartphone, without transferring any sensor data out of the device itself. Only the model updates are exchanged between devices through the network. The results were tested on a 2,004 sample dataset with 695 features and covering six different daily activities. The Random Forest model achieved 83.29% accuracy with the precision of 0.8348, recall of 0.8329 and F1-score of 0.8326. Federated learning simulation of 50 rounds approaches 93% of the accuracy of a centralized baseline, reducing the total energy consumption by 56%. Six diagnostic figures show what is happening in every step. The conclusion is clear: activity recognition and privacy do not need to be opposing forces anymore.
Keywords: Federated Learning, Human Activity Recognition, Sensor Suite of Smartphone, Energy Efficiency, Personalization, Privacy-preserving Machine Learning, Random Forest, Communication Efficiency.
Keywords: Federated Learning, Human Activity Recognition, Sensor Suite of Smartphone, Energy Efficiency, Personalization, Privacy-preserving Machine Learning, Random Forest, Communication Efficiency.
Abstract
A Comprehensive Survey of an Agentic AI Powered Multimodal RAG Learning System
Mrs. Supriya, Likitha R, Madhan N, Raksha D O, Shylashree E
DOI: 10.17148/IJARCCE.2026.15811
Abstract: Modern-day education has changed due to the utilization of artificial intelligence by implementing adaptive learning systems tailored towards individual students. Traditional learning methods generally present the same content to every learner without considering individual learning speed, understanding, or preferences. To address this limitation, this project proposes Data Spark Assistant β An Agentic AI Powered Multimodal RAG Learning System, an intelligent learning platform that delivers personalised educational support using Artificial Intelligence (AI). The system combines Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), Natural Language Processing (NLP), and Agentic AI to analyse learner performance and generate customised study materials. Through intelligent quizzes and assessments, the platform identifies each learner's strengths, weaknesses, and knowledge gaps. Based on this analysis, it recommends personalised lessons, practice questions, and learning resources that help students improve in weaker areas. By integrating adaptive learning techniques with AI-driven content generation, the proposed system creates an interactive and personalised learning environment. This approach enhances learning efficiency, encourages self-paced study, promotes critical thinking, and supports better academic performance.
Keywords: Adaptive Learning, Agentic AI, Retrieval-Augmented Generation (RAG), Large Language Models (LLMs), Natural Language Processing (NLP), Knowledge Gap Analysis, User Profiling, Multimodal Learning.
Keywords: Adaptive Learning, Agentic AI, Retrieval-Augmented Generation (RAG), Large Language Models (LLMs), Natural Language Processing (NLP), Knowledge Gap Analysis, User Profiling, Multimodal Learning.
Abstract
Secure File Sharing System
Archana Bhat, Pooja S
DOI: 10.17148/IJARCCE.2026.15812
Abstract: Nowadays, sharing important files over the internet has become very common, but it also increases the risk of unauthorized access and data theft. Many existing file-sharing systems have security issues such as weak authentication and poor access control. This paper presents a Secure File Sharing System with Multi-Factor Authentication (MFA) to improve the security of file sharing. The system is developed using Next.js, React, Better- SQLite3, Bcrypt, JSON Web Token (JWT), and Time-based One-Time Password (TOTP). It allows users to securely upload, share, and download files after successful authentication. Passwords are securely stored using Bcrypt, while JWT provides secure user sessions. TOTP-based Two-Factor Authentication adds an extra layer of security to prevent unauthorized access. The proposed system is easy to use, provides secure file sharing, and offers better protection against common cyber threats.
Keywords: Secure File Sharing, Multi-Factor Authentication, JWT, Bcrypt, TOTP, Next.js, Better-SQLite3, Web Security
Keywords: Secure File Sharing, Multi-Factor Authentication, JWT, Bcrypt, TOTP, Next.js, Better-SQLite3, Web Security
Abstract
Neuro-EvoSwarm Optimizer (NESO): A Hybrid Deep Learning, Genetic Algorithm and Particle Swarm Optimization Framework for Seasonal Multicropping Strategy Optimization under Drip Irrigation
Mrs.N. Amirtha Gowri M.Sc., M.Phil., (Ph.D), Dr. R. Nandhakumar, MCA., M.Phil., MBA., Ph.D., SET
DOI: 10.17148/IJARCCE.2026.15813
Abstract: Agricultural planning under variable soil and climatic conditions requires simultaneous consideration of productivity, profitability and limited irrigation resources. This paper presents the Neuro-EvoSwarm Optimizer (NESO), a hybrid decision-support framework that integrates Deep Learning (DL), Genetic Algorithm (GA), and Particle Swarm Optimization (PSO) for seasonal multicropping strategy optimization under drip irrigation. The framework uses soil characteristics, climatic variables and crop-management information to predict crop yield and economic profit and then searches for high-quality crop combinations. Data preprocessing includes missing-value handling, outlier detection, categorical encoding, feature scaling and normalization. The Deep Learning component learns nonlinear relationships between agricultural conditions and target outcomes. GA provides broad evolutionary exploration through population initialization, fitness evaluation, selection, crossover, mutation and replacement, while PSO refines the best GA solutions using personal-best and global-best information. A multi-objective fitness formulation is used to maximize yield and profit while minimizing irrigation water use. The source document describes the model architecture, workflow, training configuration and decision-support outputs, but does not provide a complete set of numerical test results; therefore, this condensed paper does not fabricate accuracy or optimization values. The framework provides a scalable basis for precision agriculture and can be extended with real-time IoT observations, weather forecasts, GIS and additional sustainability indicators.
Keywords: seasonal multicropping; drip irrigation; Deep Learning; Genetic Algorithm; Particle Swarm Optimization; NESO; crop yield prediction; agricultural decision support; water-use efficiency
Keywords: seasonal multicropping; drip irrigation; Deep Learning; Genetic Algorithm; Particle Swarm Optimization; NESO; crop yield prediction; agricultural decision support; water-use efficiency
Abstract
Enhancement of Vehicle Number Plate Recognition System
Manpreet Kaur, Jatinder Singh Saini, Sandeep Kaur Dhanda
DOI: 10.17148/IJARCCE.2026.15814
Abstract: Automatic Number Plate Recognition (ANPR) is an important computer vision technology used in Intelligent Transportation Systems (ITS) for traffic monitoring, law enforcement, parking management, toll collection, and vehicle access control. Previous ANPR systems have primarily relied on conventional image processing techniques, including thresholding, edge detection, morphological operations, and rule-based plate localization. Although these methods can perform effectively under controlled conditions, their accuracy is often affected by variations in illumination, vehicle orientation, background complexity, plate size, and image quality. To address these limitations, this proposed research presents an enhanced Automatic Number Plate Recognition system that integrates the YOLOv9(You Look Only Once Version 9) object detection algorithm with image processing and Optical Character Recognition (OCR) techniques. YOLOv9 is used to automatically detect and accurately localize vehicle number plates from input images, eliminating the need for predefined assumptions regarding the position and size of the plate. The detected plate region is then processed using grayscale conversion, noise reduction, histogram equalization, morphological operations, and Sobel edge detection to enhance the image and improve character separation. The segmented alphanumeric characters are subsequently recognized using an OCR module, and the extracted registration number can be matched with a database for vehicle information retrieval. By combining deep learning-based object detection (YOLO) with median filter and OCR, the proposed approach improves the robustness and reliability of the ANPR system under varying image conditions. To test the hypothesis, this research utilized the publicly available zenodo dataset for the vehicle images used to train, validate, and test the Automatic Number Plate Recognition (ANPR) system. It provides the information required for the YOLOv9 model to learn how to detect number plates and for the OCR module to recognize the characters correctly. It is a diverse dataset containing different vehicle types, lighting conditions, viewing angles, and backgrounds helps improve the accuracy and robustness of the proposed system. Experimental results demonstrate that the proposed system achieves 94.1% number plate detection accuracy and 97.2% Character recognition accuracy, indicating significant enhancement over previously developed ANPR techniques and demonstrating the effectiveness of the integrated YOLO- based ANPR framework for practical vehicle identification applications.
Keywords: Image Processing, Text Extraction, Character recognition, YOLOv9, ANPR, Number Plate Detection, MATLAB
Keywords: Image Processing, Text Extraction, Character recognition, YOLOv9, ANPR, Number Plate Detection, MATLAB
Abstract
Age-Aware Multimodal Web-Based Framework for Parkinson's Disease Detection Using Handwriting and Voice Analysis
Korupala Pavani, Dr G Suvarna Kumar
DOI: 10.17148/IJARCCE.2026.15815
Abstract: Parkinson's Disease (PD) is a progressive neurodegenerative disorder affecting millions worldwide, characterized by motor impairments including tremors, rigidity, and postural instability. Early and accurate diagnosis is critical for improving patient outcomes, yet traditional clinical diagnosis often lacks objective, quantitative measures. This paper proposes an Age-Aware Multimodal Web-Based Framework for automated Parkinson's Disease detection by integrating three complementary modalities: voice acoustic features, handwriting drawing analysis (spiral and wave patterns), and clinical biomarkers (Jitter, Shimmer, HNR). Three age groups β Gen Z (18β26 years), Adults (28β59 years), and Seniors (60+ years) β are incorporated to capture demographic diversity. Feature extraction employs Mel- Frequency Cepstral Coefficients (MFCC) for voice signals and Histogram of Oriented Gradients (HOG) for handwriting images. Classification models including Support Vector Machine (SVM) and Random Forest (RF) are trained per modality and fused via a weighted ensemble (Voice: 40%, Spiral: 30%, Wave: 30%). The clinical SVM achieves the highest accuracy of 86.36%, spiral RF achieves 85.71%, and wave RF achieves 76.19%. The complete system is deployed as a Flask-based web application enabling real-time multimodal patient assessment. Experiments over a total of 392 samples (107 clinical, 81 voice, 102 spiral images, 102 wave images) demonstrate the effectiveness of the proposed age- aware multimodal approach for non-invasive PD screening.
Keywords: Parkinson's Disease Detection, Multimodal Classification, MFCC, HOG, SVM, Random Forest, Ensemble Learning, Flask Web Application, Age-Aware Framework.
Keywords: Parkinson's Disease Detection, Multimodal Classification, MFCC, HOG, SVM, Random Forest, Ensemble Learning, Flask Web Application, Age-Aware Framework.
Abstract
An Intelligent Digital Twin Framework with Biomarker-Aware XGBoost for Ovarian Cancer Prediction
Bhavana B R*, Sindhu A G and Akash Gowda S G
DOI: 10.17148/IJARCCE.2026.15816
Abstract: Ovarian cancer is a major killer in women worldwide, and the rate of cancer deaths is especially low because of the lack of good tools to detect it in the early stages and the late diagnosis. Currently, the most used screening tests are CA-125 and transvaginal ultrasound which are not sufficiently precise to make an early diagnosis, so more intelligent prediction methods are needed. In this study, we have proposed a framework for the Digital Twin and XGBoost machine learning to make the personalized prediction of ovarian cancer risk and generation of biomarkers. The clinical data set consisted of 349 patients with 51 biomarkers; 10 were identified as key predictors by Recursive Feature Elimination, including CA125, HE4, ALB and CEA. The accuracy of the XGBoost model is 91.43%. Clinicians can use the Digital Twin component to estimate a patientβs current cancer risk and simulate the evolution of cancer risk over time, with the understanding of changes in biomarkers, to obtain interpretable clinical insights. The proposed framework performs well in terms of prediction and is a dynamic and personalized clinical decision support tool for ovarian cancer. .
Keywords: Ovarian cancer, XGBoost, Bio markers, Digital Twin.
Keywords: Ovarian cancer, XGBoost, Bio markers, Digital Twin.
Abstract
AI-Powered Resume Analyzer with Intelligent Job Recommendation and Skill Gap Analysis: A Survey of NLP and Machine Learning Approaches
M B Chethana, P Harshitha, R Manvitha, Mrs. Sougandika Narayan
DOI: 10.17148/IJARCCE.2026.15817
Abstract: The contemporary employment ecosystem presents a paradox wherein a large proportion of competent candidates fail to advance past automated screening stages, not due to skill deficiency but due to inadequate resume formulation. Applicant Tracking Systems (ATS) employed by recruiters systematically eliminate resumes that do not conform to keyword density thresholds and structural conventions, creating an opaque barrier for job seekers lacking professional guidance. This paper surveys six representative studies spanning 2021β2026 on NLP-based resume analysis, machine learning-driven job role prediction, semantic candidate-job matching, and automated skill gap identification. Each study is evaluated across methodology, algorithmic approach, dataset characteristics, and core limitations. Based on this survey, we propose a unified AI-Powered Resume Analyzer that integrates a multi-stage NLP extraction pipeline, a Scikit-learn classification model for job role prediction, a structured skill-gap benchmarking module, and a Flask-based web interface. The proposed system is designed to provide candidates with quantitative ATS compatibility scores, predicted suitable job roles, identified competency gaps, and prioritized improvement recommendations in a single candidate-facing platformβaddressing critical shortcomings in existing recruiter-centric approaches.
Keywords: Natural Language Processing, Resume Analysis, Applicant Tracking System, Skill Gap Analysis, Job Role Prediction, Machine Learning, SpaCy, NLTK, Flask, TF-IDF, BERT, Scikit-learn, Career Guidance, Named Entity Recognition
Keywords: Natural Language Processing, Resume Analysis, Applicant Tracking System, Skill Gap Analysis, Job Role Prediction, Machine Learning, SpaCy, NLTK, Flask, TF-IDF, BERT, Scikit-learn, Career Guidance, Named Entity Recognition
Abstract
A Comparative Study of Classical and Deep Learning Approaches for Non-Uniform Motion Blur Restoration
Sarthak Agarwal, Ms.Charulatha RT
DOI: 10.17148/IJARCCE.2026.15818
Abstract: Motion blur caused by fast or irregular relative motion between a camera and its subject remains a challenging image restoration problem, particularly when the blur is spatially non-uniform and follows a curved rather than linear trajectory β conditions under which standard deblurring assumptions break down. This work presents and rigorously compares two independent restoration pipelines: a classical approach combining per-tile blind kernel estimation (cepstrum-based and gradient-prior methods) with Richardson-Lucy deconvolution, and a lightweight deep learning model (NAFNet-lite, ~0.44M parameters) trained on a custom-generated synthetic dataset simulating non-uniform, curved motion blur via randomized Bezier-path kernels. Both pipelines are evaluated on a shared held-out benchmark using PSNR, SSIM, and an auxiliary readability-based metric. Results show the deep learning model consistently outperforms the classical baseline on pixel-fidelity metrics (SSIM 0.955 vs. 0.705), while the classical pipeline achieves comparable or better rates of individual-sample improvement despite requiring no training. An ablation study reveals that total-variation regularization, commonly used to suppress deconvolution ringing, measurably degrades fine-detail recovery in this setting. Further, an attempt to improve real-world generalization by sequentially fine-tuning the deep model on a real-world photographic blur/sharp dataset resulted in significant catastrophic forgetting of performance on the synthetic benchmark, highlighting a practical limitation of naive fine-tuning and motivating joint or replay-based training strategies as future work. A complete, interactive web application was developed to demonstrate both restoration pipelines side by side.
Keywords: Motion Deblurring, Blind Deconvolution, Non-Uniform Blur, Richardson-Lucy Deconvolution, Deep Learning, NAFNet, Synthetic Data Generation, PSNR, SSIM, Ablation Study, Catastrophic Forgetting, Transfer Learning, Web-Based Deployment.
Keywords: Motion Deblurring, Blind Deconvolution, Non-Uniform Blur, Richardson-Lucy Deconvolution, Deep Learning, NAFNet, Synthetic Data Generation, PSNR, SSIM, Ablation Study, Catastrophic Forgetting, Transfer Learning, Web-Based Deployment.
Abstract
AI-Based Disaster Relief Demand Prediction System
M. Sri Chaitanya Varma, Pyla Jyothi, N.J.V.D. Jagadeesh, N. Enosh Raju, K. Jagadeesh, G. Omkar Venkat
DOI: 10.17148/IJARCCE.2026.15819
Abstract: Natural disasters like floods, tsunamis, cyclones, earthquakes, droughts, and wildfires have a drastic impact on a country like India, where the occurrence of disasters is frequent. We have made progress to some extent in predicting disasters in advance and planning mitigation strategies by notifying authorities and the public to be prepared. However, during and after a disaster, people may lose their lives not because there is no help available, but because the help does not reach the right place at the right time. Relief in terms of food, water, rescue personnel, medical teams, and shelters is meant to reach the affected people. This research proposes a system to predict accurate relief demand using past disaster records, live disaster data, and geographical information. Machine learning models are used, where a Random Forest classifier is applied to classify the type of disaster from live news data, followed by five XGBoost predictors for different types of relief resources, and K-Nearest Neighbors for geographical similarity matching to predict relief for newly impacted areas. A web-based dashboard with analytics and map visualization is developed to support decision-making and provide better insights for efficient resource allocation. The XGBoost regression models for relief prediction achieved high predictive performance, with an overall RΒ² score of approximately 0.93 across all resource categories.
Keywords: Disaster Management, Disaster Response System, Machine Learning, Resource Demand Prediction, XGBoost, Random Forest, K-Nearest Neighbors, Geographical Analysis.
Keywords: Disaster Management, Disaster Response System, Machine Learning, Resource Demand Prediction, XGBoost, Random Forest, K-Nearest Neighbors, Geographical Analysis.
Abstract
Explainable AI-Based Customer Churn Prediction Using Machine Learning and SHAP Analysis
Dr. Bharathi M P, Narendra C, Abhishekaraddi Maraddi
DOI: 10.17148/IJARCCE.2026.15820
Abstract: Over Customer churn prediction plays a critical role in withstanding the competition in the contemporary business world. This paper presents an Explainable Artificial Intelligence (XAI) framework for predicting customer churn using advanced machine learning algorithms. The framework employed includes XGBoost, LightGBM, Random Forest, and Ensemble Learning to develop an accurate predictive model. The study used customer demographics and behavior data to train the algorithm. Additionally, the study utilized SHAP (SHapley Additive ExPlanations) to interpret the results by illustrating the key factors associated with customer churn. The results indicated that all the predictive models had good Accuracy, F1-score, and ROC-AUC performance metrics. However, the Ensemble Learning model had the highest ROC-AUC of 93.8%. Thus, the study concludes that the proposed solution is an effective approach to predicting customer churn and aids businesses in making the right decisions based on the results.
Keywords: Explainable AI, SHAP, XGBoost, LightGBM, Random Forest, Ensemble Learning, Machine Learning, Data Analytics.
Keywords: Explainable AI, SHAP, XGBoost, LightGBM, Random Forest, Ensemble Learning, Machine Learning, Data Analytics.
Abstract
A Comparative Analysis of Starling Swarm Algorithm for Optimisation in Dynamic Environment
Karatu Musa Tanimu, Wei Pang, and George M. Coghill
DOI: 10.17148/IJARCCE.2026.15821
Abstract.
Many real-world phenomena
can be modelled as dynamic optimisation problems, where the environment of a
problem is dynamic and therefore, conventional methods are not capable of
dealing with such situations. In this paper, a novel starlings swarm algorithm
is proposed for dealing with such dynamic environments. The proposed starlings
algorithm consists of a swarm, which has two distinct behaviours to change in
their environment that is mutually exclusive, namely: collective response in
terms of communication within the seven nearest neighbours and murmuration in
terms of patrolling the search space. This way a diverse population is created
while the seven nearest neighbours progressively track the dynamic changes. We
established that in problems with a moderate shift severity the best variants
are those with topology similar to starlings and results confirmed a
significant improvement in those cases that used starlings swarm algorithm,
which eliminates the use of independent multi-swarm in the search space.
Keywords: Swarm Intelligence, starlings, Optimisation algorithm, bio-inspired, Environment dynamic.
Abstract
Automated Business Card Digitization and Smart Contact Follow Up Using AI
Basavaraj R Palled, Shobha Rani B R
DOI: 10.17148/IJARCCE.2026.15822
Abstract: Automated Business Card Digitization and Smart Contact Follow Up Using AI is a cloud based, multi tenant system built to automate the handling of physical business cards. It digitizes card images through browser side Optical Character Recognition using Tesseract.js, structures the extracted text into verified JSON contact records using Google Gemini 2.5 Flash with an enforced JSON response schema, and stores records in a PostgreSQL database where Row Level Security policies provide complete tenant isolation. A large language model driven follow up composer generates personalized outreach messages, which users preview and dispatch via WhatsApp click to chat links, the Resend transactional email API, or LinkedIn profile deep links. Testing across a representative sample of business card images showed that the Gemini augmented extraction pipeline achieved 96.4% accuracy and 95.6% F1 score, improving significantly over standalone Tesseract.js OCR at 67.9%. The end to end pipeline, from client side image compression through extraction and database persistence, completed in an average of 1.22 seconds. Cross tenant isolation tests returned zero data leaks across all evaluated scenarios.
Keywords: business card digitization, OCR, large language model, JSON schema extraction, multi tenant SaaS, Row Level Security, CRM automation, serverless edge functions, Gemini 2.5 Flash, Supabase, multi channel follow up
Keywords: business card digitization, OCR, large language model, JSON schema extraction, multi tenant SaaS, Row Level Security, CRM automation, serverless edge functions, Gemini 2.5 Flash, Supabase, multi channel follow up
Abstract
SECURE API COMMUNICATION IN CLOUD APPLICATIONS
Mr. Naveen J, Manjunatha H, Sojo T Johnson
DOI: 10.17148/IJARCCE.2026.15823
Abstract: The rapid expansion of cloud-native applications has made API security a critical discipline in modern software engineering. APIs serve as the communication backbone of microservices, connecting mobile clients, third-party platforms, and enterprise back-ends across distributed cloud environments. This research investigates secure API communication mechanisms for cloud applications, comparing REST, GraphQL, and gRPC paradigms through a security lens. The OWASP API Security Top 10 is analysed with targeted mitigations, and four authentication mechanisms β OAuth 2.0 with PKCE, JWT, Mutual TLS, and Zero Trust β are evaluated both theoretically and experimentally. An empirical testbed built on Kubernetes, Istio, and Kong measured latency, throughput, and security scores across five protocol configurations. Results show that gRPC with mTLS achieves the best security-performance balance (9.2/10 security, 28 ms latency, 3,500 req/s).
The Unified Secure API Framework (USAF) β A Practical Reference Architecture for Real-World Cloud Implementations
Keywords: API Security, Cloud Computing, OAuth 2.0, JSON Web Token (JWT), Mutual TLS (mTLS), Zero Trust Architecture, OWASP API Security, REST, GraphQL, gRPC, Microservices, API Gateway, Secrets Management, PKCE, Rate Limiting, CloudNative Security
The Unified Secure API Framework (USAF) β A Practical Reference Architecture for Real-World Cloud Implementations
Keywords: API Security, Cloud Computing, OAuth 2.0, JSON Web Token (JWT), Mutual TLS (mTLS), Zero Trust Architecture, OWASP API Security, REST, GraphQL, gRPC, Microservices, API Gateway, Secrets Management, PKCE, Rate Limiting, CloudNative Security
Abstract
Reducing Round-Trip Time in Cloud Environments Using Software-Defined Networking with Intelligent Prefetching and Caching: A Review
Chandu H M, Nagendra Swamy M S, Chandan Hegde
DOI: 10.17148/IJARCCE.2026.15824
Abstract: Round-Trip Time (RTT) is considered the main metric for measuring the performance of applications for end users in today's cloud environments. In order to improve the user experience, RTT has to be reduced by future network architectures. Current architectures are mainly designed to be reactive, meaning that they counteract congestion, user mobility and changes in service demand and resource requirements only after the quality of service that is being provided to users has already been affected. In this paper, we review works that have contributed in this regard and also present a completely integrated proactive network architecture design based on Software-Defined Networking (SDN), machine-learning-based prefetching and an edge cache manager. The architecture is especially suitable for scenarios with high user mobility as well as with high variability of access traffic. In our work, we first present notable technology progress supporting reduced round-trip times followed by a detailed description of the novel architecture. The work also describes a simulation-based approach for evaluating the performance of the architecture. In the end we outline the expected results as well as the challenges that have to be faced during the implementation of the architecture and how these challenges affect future cloud as well as edge networks.
Keywords: Software-Defined Networking (SDN); Predictive Prefetching; Edge-Centric Caching; Round-Trip Time (RTT); Cloud Computing; Quality of Service (QoS); Machine Learning; Network Mobility.
Keywords: Software-Defined Networking (SDN); Predictive Prefetching; Edge-Centric Caching; Round-Trip Time (RTT); Cloud Computing; Quality of Service (QoS); Machine Learning; Network Mobility.
Abstract
Smart Underground Cable Fault Detector
Mrs. Mamatha N, B Kiran Kumar, Charan K V, Bharath Gowda, Jnanesh R
DOI: 10.17148/IJARCCE.2026.15825
Abstract: the Smart Underground Cable Fault Detector is an innovative solution designed for the rapid and precise identification of issues in buried power lines. While subterranean cables offer superior safety and aesthetic advantages, they are vulnerable to faults like short circuits or insulation failure caused by aging and moisture. Traditional troubleshooting is often laborintensive and expensive.
This system leverages an Arduino Uno and the voltage drop principle to pinpoint fault locations. By using an onboard ADC to process signals and an ESP8266 for wireless connectivity, the device provides real-time updates and distance calculations via an LCD. This automation significantly reduces the need for manual inspection and disruptive excavation, making it ideal for smart city infrastructure and industrial power grids."
This system leverages an Arduino Uno and the voltage drop principle to pinpoint fault locations. By using an onboard ADC to process signals and an ESP8266 for wireless connectivity, the device provides real-time updates and distance calculations via an LCD. This automation significantly reduces the need for manual inspection and disruptive excavation, making it ideal for smart city infrastructure and industrial power grids."
Abstract
Cloud Resource Allocation and Scheduling: A Comprehensive Study
Srinivasa G., KareemPasha, Chandan Hegde
DOI: 10.17148/IJARCCE.2026.15826
Abstract: Cloud computing has now become a basic element of IT services delivery, enabling the provision of on demand, scalable computing resources over the Internet, in an efficient manner. The challenge for the efficient management of the resources in cloud computing environments, which are scalable and dynamic by nature, arises from their increasing scale. Traditional methods fail to address the issue of the efficient scheduling of tasks for the load balancing in such environments. In order to overcome this challenge, numerous scheduling methods that are based on metaheuristics have been proposed and effectively applied to various types of cloud computing environments. The aim of this paper is to present a comprehensive survey and comparison of five well known, efficient scheduling methods, which are based on metaheuristics. In the paper, five typical methods, namely, Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Ant Colony Optimization (ACO), Simulated Annealing (SA), and Artificial Bee Colony (ABC), are studied in detail. The principles, advantages, and disadvantages of the five methods are discussed in the paper, and a comparison of the methods is also presented, based on a number of important factors, such as the quality of the solutions found, the execution time, the resource utilization, the scalability of the methods, and the corresponding computational costs.
Keywords: Cloud Computing, Resource Allocation, Task Scheduling, Metaheuristic Algorithms, Optimization.
Keywords: Cloud Computing, Resource Allocation, Task Scheduling, Metaheuristic Algorithms, Optimization.
Abstract
Short-Term Price Movement Prediction Using Order Book Heatmap Vision
R.T.N Sura Reddy, Sharavana Ragav S, Ms. Charulatha R .T
DOI: 10.17148/IJARCCE.2026.15827
Abstract: Financial markets generate high-frequency data at a scale that makes manual feature engineering increasingly inadequate. This paper presents a computer vision-based framework for short-term price movement prediction by transforming Limit Order Book (LOB) snapshots into grayscale heatmap images. Unlike conventional approaches that depend on handcrafted technical indicators, the proposed method learns directly from raw market microstructure patterns using deep learning. Three architectures are investigated progressively: a lightweight baseline convolutional neural network (CNN), an SE-ResNet-lite model, and a Dual-Axis Fusion Network (DAFNet) tailored to the distinct semantics of price and time axes in LOB heatmaps. Experimental results indicate that the proposed pipeline improves predictive performance from a near-random 50-60% baseline to more than 70% test accuracy. These findings are consistent with recent literature showing that order book representation and model design substantially influence short-horizon forecasting quality.
Keywords: Limit Order Book, Computer Vision, Deep Learning, High-Frequency Trading, Convolutional Neural Networks, DAFNet, Price Prediction.
Keywords: Limit Order Book, Computer Vision, Deep Learning, High-Frequency Trading, Convolutional Neural Networks, DAFNet, Price Prediction.
Abstract
FrameDeblur: NAFNet-Based Selective Video Frame Deblurring and Restoration Framework
S.Roshan Pranao, Harihara Balan S, Yelavarthi Sai Dheeraj, Ms. Charulatha R T
DOI: 10.17148/IJARCCE.2026.15828
Abstract: This paper presents FrameDeblur, a video-based image restoration framework designed to identify and deblur individual blurred frames from a video sequence. Unlike conventional image deblurring approaches that directly operate on a single image or process an entire video sequence, the proposed system accepts a video as input and provides an interactive mechanism for manually selecting the frame requiring restoration. The selected frame is extracted and subjected to preprocessing before being passed through a pretrained NAFNet (Nonlinear Activation Free Network) for deep image restoration. The restored output is subsequently refined using classical image restoration and enhancement operations to improve sharpness, structural details, and overall visual quality. The system integrates video frame extraction, manual frame navigation, deep learning-based restoration, and final image generation within a unified interactive workflow. By concentrating computational processing on the user-selected frame, the framework avoids unnecessary restoration of the complete video sequence and provides a practical approach for selective frame-level deblurring. The system is implemented with an interactive Gradio-based interface, enabling users to upload a video, navigate through its frames, select a blurred frame, and obtain the corresponding deblurred image as output. The proposed framework provides a flexible foundation for future extensions involving automatic blur detection, sequential video restoration, temporal consistency, and real-time deblurring.
Keywords: Video Deblurring, Frame Selection, Image Restoration, NAFNet, Deep Learning, Motion Blur, Video Processing, Frame Enhancement, Selective Restoration, Gradio.
Keywords: Video Deblurring, Frame Selection, Image Restoration, NAFNet, Deep Learning, Motion Blur, Video Processing, Frame Enhancement, Selective Restoration, Gradio.
Abstract
ARTIFICIAL INTELLIGENCE IN TRANSFORMING THE GIG ECONOMY
Chandrakala G S
DOI: 10.17148/IJARCCE.2026.15829
Abstract: Digital platforms have changed the way short-term work is organised, and artificial intelligence has become an important mechanism behind this transformation. This paper studies how AI is used in India's gig economy for assigning work, adjusting prices, evaluating performance and identifying fraud. The discussion uses Uber, Zomato and Upwork as representative platform examples and draws on secondary information from the sources listed in the study. The analysis shows that AI helps platforms coordinate large numbers of workers and customers, reduce operational delays, personalise opportunities, and strengthen platform security. At the same time, automated decisions can create concerns relating to fairness, explainability, employment stability and privacy. The paper therefore argues that technological efficiency alone is not sufficient for sustainable platform growth. Transparent decision systems, regular algorithmic checks, human review of adverse decisions and responsible handling of worker data are needed to build greater trust. The study concludes that AI should be developed as a worker-supporting technology rather than only as an instrument for platform control.
Keywords: Artificial Intelligence, Gig Economy, Platform Work, Algorithmic Management, Job Matching, Dynamic Pricing, Fraud Detection, Uber, Zomato, Upwork, India
Keywords: Artificial Intelligence, Gig Economy, Platform Work, Algorithmic Management, Job Matching, Dynamic Pricing, Fraud Detection, Uber, Zomato, Upwork, India
Abstract
PaperPilot AI: An AI-Powered Multimodal Research Paper Mentor for Intelligent Academic Content Understanding, Mathematical Equation Decoding, and Comparative Literature Analysis
George, Dr. Beenarani Manoj
DOI: 10.17148/IJARCCE.2026.15830
Abstract: PaperPilot AI is an end-to-end multimodal research paper mentorship and intelligent retrieval platform designed to accelerate scientific paper comprehension, mathematical formula deconstruction, and comparative literature analysis. Modern academic literature relies heavily on complex LaTeX equations, multi-axis figures, empirical benchmark tables, and specialized domain terminology that present severe barriers to students and researchers. PaperPilot AI addresses these challenges by combining PyMuPDF document extraction, regex-driven mathematical formula detection, a Retrieval-Augmented Generation (RAG) Q&A engine with page-level citations, and high-performance LLMs (Groq Llama 3.3 70B and heuristic professor fallbacks). The system features multi-tiered explanation modules ('Like I'm 15', 'College Student', 'Simple English', and 'Real-World Analogies'), line-by-line LaTeX variable symbol breakdowns with worked numerical examples, visual plot axis decoders, experimental table metric evaluators, automated research methodology flowcharts, interactive Anki flashcards, multi-type quiz generators (MCQ, True/False, Coding, Interview), and automated citation generators (BibTeX, APA, IEEE, MLA). Furthermore, the platform integrates a comparative literature engine that benchmarks paper contributions against baseline models. The system is delivered via a FastAPI backend and a modern React Vite frontend, offering a responsive, professor-grade interactive research companion.
Keywords: Research Paper Mentor, Retrieval-Augmented Generation (RAG), Multimodal LLM, LaTeX Equation Decoding, Figure & Table Intelligence, Comparative Literature Analysis, Automated Quiz & Flashcard Generation.
Keywords: Research Paper Mentor, Retrieval-Augmented Generation (RAG), Multimodal LLM, LaTeX Equation Decoding, Figure & Table Intelligence, Comparative Literature Analysis, Automated Quiz & Flashcard Generation.
Abstract
The influence of blockchain technology and high- involvement human resource practices on supply chain resilience and organizational performance
Mohammad Ali Yousef Yamin, Abd Arahman Hussain Al Amri, Safar Said Alamri
DOI: 10.17148/IJARCCE.2026.15831
Abstract: The rising global competition and turbulent business environment have enabled organizations to develop resilient and crisis induced supply chain strategies. Therefore, current study strives to understand how blockchain technology, artificial intelligence and high involvement HR practices impact supply chain resilience and organizational performance. Moreover, moderating effect of environmental dynamism is examined between the relationship of supply chain resilience and organizational performance. The research model is empirically tested with 293 responses collected from managers working in manufacturing firms. Research framework is developed following positivist research paradigm. Data are computed through structural equation modeling approach. Results of the empirical analysis have revealed that blockchain novelty, blockchain efficiency, artificial intelligence, employee skills, employee incentives and employee participation explained π
2 51.1% variance in supply chain resilience. Therefore, environmental dynamism and resilience have explained π
2 48.8% variance in organizational performance. The findings of this research have suggested that policy makers could enhance supply chain resilience through blockchain novelty, blockchain efficiency, artificial intelligence, employee skills and employee incentives. Moreover, this study has suggested that if logistics firms comprise characteristics of environmental dynamism they would have better ability to work in uncertain environment. This research is pioneering as it has examined the impact of blockchain novelty and blockchain efficiency towards supply chain resilience. Similarly, developing an integrative resilient logistic research model with blockchain technology, artificial intelligence, high involvement HR practices and environmental dynamism makes this research more unique and valuable.
Keywords: Blockchain novelty; Blockchain efficiency; Artificial intelligence; High involvement HR practices; Supply chain resilience; Environmental dynamism.
Keywords: Blockchain novelty; Blockchain efficiency; Artificial intelligence; High involvement HR practices; Supply chain resilience; Environmental dynamism.
Abstract
A COMPREHENSIVE SURVEY OF AN AQUA CYCLE INTELLIGENCE: AI BASED SYSTEM FOR RO REJECT WATER OPTIMIZATION AND SUSTAINABILITY
Ankit, Ayush Gupta, Darshan K S, Dharmaraj, Mrs.Supriya
DOI: 10.17148/IJARCCE.2026.15832
Abstract: Reverse Osmosis (RO) water purifiers are widely used in households, industries, educational institutions, and commercial buildings for obtaining clean drinking water. However, one of the major disadvantages of RO purification systems is the excessive generation of reject water. Approximately 50β75% of the input water is discharged as reject water during the purification process, leading to significant water wastage and environmental concerns. In most cases, reject water is directly discarded without analyzing its quality or exploring its potential for secondary usage. Existing systems mainly focus on water purification and fail to provide intelligent monitoring, optimization, and sustainability- driven reuse mechanisms. This paper proposes AQUA CYCLE INTELLIGENCE, an Artificial Intelligence-based system designed to optimize the utilization of RO reject water through smart monitoring, analysis, and recommendation techniques. The proposed system integrates IoT sensors, Artificial Intelligence algorithms, cloud computing, and real- time dashboards to monitor parameters such as Total Dissolved Solids (TDS), pH level, temperature, turbidity, and water flow rate. Using AI and Machine Learning techniques, the system predicts the quality and usability of reject water and recommends suitable secondary applications such as floor cleaning, gardening, flushing, cooling systems, and washing purposes. The proposed framework also enables real-time monitoring through web and mobile dashboards, tracks water savings and environmental impact, and generates analytical reports for users and authorities. By intelligently managing RO reject water, the system promotes water conservation, reduces unnecessary wastage, supports sustainable development goals, and encourages efficient resource utilization. The integration of AI, IoT, and cloud technologies transforms RO reject water from a discarded byproduct into a valuable reusable resource.
Abstract
Atlas AI: Watching the planet evolve through AI-Powered Satellite Intelligence
Deekshith M, Dhruv Kaushik, Likhith Gowda A K, Varun M, Asha K H
DOI: 10.17148/IJARCCE.2026.15833
Abstract: Environmental changes such as deforestation, urban expansion, and reduction in water bodies have increased in recent years. Monitoring these changes using traditional field surgeys is often time-consuming and difficult for large areas. Because of this, satellite remote sensing combined with artificial intelligence has become an effective approach for environmental monitoring. This paper reviews recent studies that use satellite pfatforms such as Sentinel-1, Sentinef-2, and Landsat along with models including CAT, U-Net, Vision Transformers (ViT), and LSTM for change detectton The reviewed studies suggest that combining multispectral and SAR data improves monitoring accuracy in different land- cover conditions. it is also obsewed that transformer and hybrid deed learning models oflen Perform better than traditional approaches. Oreraff, the combination of AI and satellite imagery provides a useful way to monitor environmental changes and support better planning and consewation decisions.
Keywords: Environmental Monitoring, Remote Sensing, Satellite Imagery, Artificial Intelligence, Deep Learning, Deforestation Detection, Urban Expansion, Water Body Shrinkage, Sentinel-1, Sentinel-2, Landsat, Convolutional Neural Network (CNN), U-Net 3D, Vision Transformer (ViT), Long Short-Term Memory (LSTM), Transformer-Based Models, Change Detection, Sustainable Development, Urban Planning, Land Use Land Cover (LULC).
Keywords: Environmental Monitoring, Remote Sensing, Satellite Imagery, Artificial Intelligence, Deep Learning, Deforestation Detection, Urban Expansion, Water Body Shrinkage, Sentinel-1, Sentinel-2, Landsat, Convolutional Neural Network (CNN), U-Net 3D, Vision Transformer (ViT), Long Short-Term Memory (LSTM), Transformer-Based Models, Change Detection, Sustainable Development, Urban Planning, Land Use Land Cover (LULC).
Abstract
Design and FPGA Implementation of a Low- Power Dynamic Clock-Gated Data Processing Unit
Macherla Sunitha, Dr B Ramprasad
DOI: 10.17148/IJARCCE.2026.15834
Abstract: Power consumption has become a critical design constraint in modern digital systems, particularly in battery-operated embedded devices, Internet of Things (IoT) applications, and FPGA-based computing platforms. Clock networks contribute a significant portion of the total dynamic power consumption due to continuous switching activity, even when functional modules remain idle. This paper presents the design and FPGA implementation of a Low-Power Dynamic Clock-Gated Data Processing Unit (DPU) using Verilog HDL. The proposed architecture incorporates a dynamic clock-gating mechanism that selectively disables clock signals to inactive processing blocks, thereby reducing unnecessary switching activity and minimizing power dissipation. The Data Processing Unit integrates arithmetic, logical, and data manipulation functions controlled through a centralized control unit and intelligent clock-gating circuitry. The design is modeled at the Register Transfer Level (RTL) and synthesized for FPGA implementation. Functional verification is performed using comprehensive simulation scenarios to validate the correctness of all processing operations and clock-gating behavior. Experimental results indicate that the proposed architecture achieves approximately 35β45% reduction in dynamic power consumption compared to conventional non- clock-gated designs while maintaining equivalent computational performance. Furthermore, the design operates at a maximum clock frequency of 180 MHz with less than 60 FPGA logic elements utilized. The combination of low power consumption, reduced switching activity, and efficient hardware utilization demonstrates the effectiveness of the proposed dynamic clock-gated DPU for energy-efficient FPGA-based digital systems and embedded applications.
Keywords: Dynamic Clock Gating, Data Processing Unit, FPGA Implementation, Low-Power Design, Verilog HDL, Energy-Efficient Computing.
Keywords: Dynamic Clock Gating, Data Processing Unit, FPGA Implementation, Low-Power Design, Verilog HDL, Energy-Efficient Computing.
Abstract
DEEPFAKE IMAGE DETECTION USING DEEP LEARNING
Pavaman D.K, Yashodara.R, Prakash Naik, Puneeth.A, Divyashree.K, Ullas Reddy .G
DOI: 10.17148/IJARCCE.2026.15835
Abstract: The rapid growth of artificial intelligence has made it possible to create images and videos that look increasingly realistic. One of the most noticeable examples of this technology is deepfake media, where a person's face or identity can be digitally changed using artificial intelligence. Although deepfake technology has useful applications in areas such as entertainment, education and digital content creation, it can also be misused to spread false information, impersonate people and manipulate digital evidence Because modern deepfakes can look very convincing, identifying them simply by looking at an image is becoming more difficult. This has led researchers to explore automated detection methods based on machine learning and deep learning. In particular, convolutional neural networks have been widely used because they can learn visual patterns that may not be obvious to human observers.
This literature review examines existing research on deepfake generation and detection, commonly used datasets, deep learning approaches and the challenges faced by current detection systems. It also discusses EfficientNet as a lightweight deep learning architecture and its suitability for developing a practical deepfake image detection system such as DeepGuard.
This literature review examines existing research on deepfake generation and detection, commonly used datasets, deep learning approaches and the challenges faced by current detection systems. It also discusses EfficientNet as a lightweight deep learning architecture and its suitability for developing a practical deepfake image detection system such as DeepGuard.
Abstract
A Comparative Study of Lightweight Machine Learning Models for Detecting Suspicious Network Traffic
Dadavali S. P.
DOI: 10.17148/IJARCCE.2026.15836
Abstract: The increasing use of computer networks and Internet-based services has increased the risk of suspicious and potentially malicious network activities. Traditional monitoring approaches may require continuous manual analysis and may become difficult to apply to large volumes of traffic. This paper presents a comparative evaluation of five lightweight machine learning models for binary classification of network traffic as Normal or Attack. The evaluated models are Logistic Regression, Decision Tree, K-Nearest Neighbors (KNN), Random Forest, and Support Vector Machine (SVM). Experiments are conducted using the UNSW-NB15 dataset. A reproducible stratified sample of 20,000 training records and 10,000 testing records is used to keep the experiment computationally manageable. The models are evaluated using accuracy, precision, recall, F1-score, and confusion matrices. In the experiment, Random Forest achieves the highest accuracy (86.30%) and F1-score (88.81%), while SVM obtains the highest recall (99.47%). The findings show that conventional machine learning models can provide useful suspicious-traffic classification without requiring complex deep-learning architectures. Random Forest provides the best overall balance among the evaluated metrics under the experimental conditions of this study.
Keywords: Machine learning, network traffic, suspicious traffic detection, network security, classification, UNSW- NB15, intrusion detection.
Keywords: Machine learning, network traffic, suspicious traffic detection, network security, classification, UNSW- NB15, intrusion detection.
Abstract
Deep Learning-Based Image Classification and Real-Time Object Detection Using CNN and YOLOv8
Madan S, Prem Singh M, T Chandraiah, Poornima B H, Basavanna M
DOI: 10.17148/IJARCCE.2026.15837
Abstract: Image classification is one of the fundamental areas of application of Machine Learning and Deep Learning in the field of computer vision. The exponential rise in the number of digital images produced by smartphones, cameras, satellites, and medical devices has made manual classification tedious, time-consuming, and error-prone, while conventional approaches depend on manually engineered features and give inaccurate results for complex images. This work presents a web-based Image Classification and Object Detection System that classifies Natural, Medical, and Satellite images using a trained Convolutional Neural Network and performs real-time object detection using YOLOv8. The system was trained on a multi-domain dataset of 12,783 RGB images spanning 14 classes across three groups, with all images resized to 224 Γ 224 pixels for the M standard input. It is deployed as a Flask web application in which users register, log in, upload an image, and receive the predicted category together with a confidence score, while every prediction is stored in an SQLite database and made available through a prediction history, categories, performance, and profile dashboard. Experimental results show confidence levels between 85.1% and 100% across the tested classes, confirming that the system performs accurate and efficient image classification with a user-friendly interface and no requirement for manual feature extraction.
Keywords: Image Classification, Deep Learning, Convolutional Neural Network, YOLOv8, Object Detection, Transfer Learning.
Keywords: Image Classification, Deep Learning, Convolutional Neural Network, YOLOv8, Object Detection, Transfer Learning.
Abstract
A Robust Multi-Stage Framework for Noise-Resilient Bone Tumor Detection, Segmentation, and Classification
V.Dineshkumar, Dr. N. Kamaraj
DOI: 10.17148/IJARCCE.2026.15838
Abstract: Accurate identification and segmentation of bone tumors from medical images are essential for early diagnosis, treatment planning, and prognosis estimation. However, automated analysis remains difficult because bone lesions show considerable variation in shape, size, texture, anatomical location, and boundary definition, while medical images are frequently affected by noise and low contrast. This research work presents a three-phase framework for bone tumor analysis. In Phase 1, image normalization, noise reduction, contrast enhancement, adaptive superpixels, and stochastic clustering based on Gaussian Mixture Models and Expectation-Maximization are used for noise-resilient candidate- region segmentation. In Phase 2, a Fast Mask R-CNN architecture with a VGG-19 backbone is employed for simultaneous tumor localization, segmentation, and benign/malignant classification. In Phase 3, Snake Swarm Optimization (SSO) is integrated to refine segmentation boundaries and optimize key model parameters. This results show progressive improvement across phases: the Phase 1 model achieved 96.3% accuracy, 90.5% sensitivity, 97.89% Dice coefficient, and 96.86% Jaccard index; the Phase 2 model achieved 96.31% accuracy, 94.25% precision, 97.82% recall, and 95.99% F1-score; and the optimized Phase 3 model achieved 97.8% accuracy, 97.2% precision, 97.9% recall, and 97.5% F1- score. Collectively, the findings suggest that combining adaptive segmentation, instance segmentation, and swarm-based optimization can improve robustness and clinical usefulness in computer-aided bone tumor diagnosis.
Keywords: bone tumor; medical image segmentation; Fast Mask R-CNN; adaptive superpixels; Gaussian mixture model; expectation-maximization; Snake Swarm Optimization; computer-aided diagnosis.
Keywords: bone tumor; medical image segmentation; Fast Mask R-CNN; adaptive superpixels; Gaussian mixture model; expectation-maximization; Snake Swarm Optimization; computer-aided diagnosis.
Abstract
EXTENSIBLE STORAGE MANAGEMENT AND DATA AVAILABILITY TECHNIQUES
Anozie, E. L., Okoronkwo, M. C., Emesiani, J. U., Eze, R. C.
DOI: 10.17148/IJARCCE.2026.15839
Abstract: Storage management is a critical component of modern information technology, encompassing a wide range of techniques and strategies to efficiently store, organize and retrieve data. With the high growth of digital information in recent years, effective storage management has become paramount for businesses and individuals. Effective storage management improves data availability by ensuring the data is properly structured, stored, and accessible, even during periods of high load or during system maintenance. Extensible storage management and data availability techniques encompass a broad range of strategies and technologies designed to manage data storage systems dynamically and ensure data is available whenever needed. In this work, we looked at the techniques, current trends and applications of storage management, benefits and challenges of storage management. We also highlighted the components and challenges of data availability, best practices and tools to ensure data availability and the impact of data availability on business operations.
Keywords: Storage Management, Data Availability, Database Management System
Keywords: Storage Management, Data Availability, Database Management System
Abstract
Reskilling the Banking Workforce in the Age of Artificial Intelligence: Emerging HRM Priorities and Future Skill Requirements in the Indian BFSI Sector
Dr. Avadhesh Vyas, Mr. Virendra Choudhary
DOI: 10.17148/IJARCCE.2026.15840
Abstract: Artificial Intelligence (AI) is transforming the banking, financial services and insurance (BFSI) industry, affecting job profiles, work procedures and client engagement models. AI-driven automation delivers efficiency and innovation benefits, but it also generates new skill requirements and intensifies the demand for continuous workforce upskilling. Reskilling thus became a key Human Resource Management (HRM) response to AI-driven transformation. This study explores the changing skill requirements of employes in the Indian BFSI sector and the new HRM priorities related to AI adoption. The study adopts a secondary-data-based conceptual review design and synthesizes evidence
including reports from the World Economic Forum, the Reserve Bank of India, and the State Bank of India. The review identifies five broad skill categories that are becoming increasingly important in AI-enabled financial organizations: AI and online literacy, data and analytical skills, cybersecurity and risk-governance capabilities, human-centric skills, and adaptability and continuous learning. Reskilling must integrate technical training with domain knowledge, ethical judgment, critical thinking, and human-oversight capability, the paper argues, and proposes a six-stage HRM reskilling framework β skill-gap identification, role-based mapping, targeted reskilling, internal mobility, continuous capability assessment, and a responsible AI culture β supported by a βReskillβRedeployβRetainβ strategic model. The results suggest a change from training-oriented HRM to strategic, skill-based, and adaptive talent management in Indian BFSI organizations.
Keywords: Artificial Intelligence; Reskilling; Human Resource Management; Banking Sector; BFSI; Future Skills; Digital Transformation; Employee Development; India
including reports from the World Economic Forum, the Reserve Bank of India, and the State Bank of India. The review identifies five broad skill categories that are becoming increasingly important in AI-enabled financial organizations: AI and online literacy, data and analytical skills, cybersecurity and risk-governance capabilities, human-centric skills, and adaptability and continuous learning. Reskilling must integrate technical training with domain knowledge, ethical judgment, critical thinking, and human-oversight capability, the paper argues, and proposes a six-stage HRM reskilling framework β skill-gap identification, role-based mapping, targeted reskilling, internal mobility, continuous capability assessment, and a responsible AI culture β supported by a βReskillβRedeployβRetainβ strategic model. The results suggest a change from training-oriented HRM to strategic, skill-based, and adaptive talent management in Indian BFSI organizations.
Keywords: Artificial Intelligence; Reskilling; Human Resource Management; Banking Sector; BFSI; Future Skills; Digital Transformation; Employee Development; India
Abstract
Historical Handwritten Document Age Identification using DCT-based Frequency Analysis
Dr. Pushpalata Gonasagi
DOI: 10.17148/IJARCCE.2026.15841
Abstract: Historical handwritten documents hold invaluable information about the development of language, literature, governance, education and cultural heritage. The automatic age estimation of such documents has become a major research question in digital humanities, archival science and forensic document analysis. In this paper, we propose a frequency-domain approach for age estimation of historical handwritten documents based on the Discrete Cosine Transform (DCT). The proposed framework starts with the preprocessing to minimize the scanning noise and normalize the document images. Each document image is divided into non-overlapping blocks and the 2D DCT is applied to derive frequency coefficients which represent the structural features of the handwriting and paper texture. To create discriminative feature vectors, we employ the statistical descriptors of certain DCT coefficients. SVM classifiers are used to categorize these features. Using the publicly accessible Medieval Paleographic Scale (MPS) dataset, which comprises handwritten manuscripts from several historical eras, we assess the suggested method. The findings of the experiment demonstrate that frequency features based on DCT may reliably classify ancient manuscripts and capture their age-related characteristics.
Keywords: Historical document, DCT, MPS, Document age estimation, SVM.
Keywords: Historical document, DCT, MPS, Document age estimation, SVM.