Recent Trends in Artificial Intelligence & It’s Applications (e-ISSN: 2583-4819, p-ISSN: 3107-7234)

Correlation-SVM: A Multicollinearity-Aware Feature Selection Framework for SVM-Based Medical Diagnosis

Satish Kumar Kalagotla, Thoudam Basanta, Mutum Bidyarani Devi — Fri, 22 May 2026 00:00:00 +0000

Medical datasets often contain redundant or highly correlated features, leading to multicollinearity that adversely affects Support Vector Machine (SVM) classifiers by causing unstable decision boundaries, inflated coefficient variances, reduced interpretability, and degraded generalization performance, yet traditional feature selection methods inadequately address this issue. This paper proposes Correlation-SVM, a novel multicollinearity-aware feature selection framework that integrates Pearson correlation analysis and Variance Inflation Factor (VIF) computation within a hierarchical elimination process specifically optimized for SVM-based medical diagnosis. The framework operates in four stages: Pearson correlation analysis to identify highly correlated feature pairs; VIF computation to quantify multicollinearity severity; hierarchical feature elimination to iteratively remove redundant features while recomputing VIF after each removal; and SVM training with cross-validation evaluation. Evaluated on four benchmark medical datasets (Wisconsin Breast Cancer, PIMA Indian Diabetes, Hepatitis, and Mammographic Mass) and compared against six state-of-the-art methods (CFS, FCBF, mRMR, SVM-RFE, LASSO, and GA-SVM) using 10-fold cross-validation with five repeats, Correlation-SVM achieved 97.42% accuracy on the Wisconsin dataset with only 5 features (44.4% reduction), outperforming all comparison methods. Multicollinearity was substantially reduced, with maximum VIF decreasing from 8.3 to 2.3 (72.3% reduction) on Wisconsin, from 12.5 to 2.1 (83.2% reduction) on Hepatitis, and from 4.2 to 1.6 (61.9% reduction) on PIMA, achieving VIF values below the acceptable threshold of 2.5. The framework requires only 38.7 seconds of computational time, making it 84% faster than GA-SVM and 79% faster than SVM-RFE, thus achieving wrapper-like performance with filter-like speed. Additionally, the selected feature subsets align with established medical knowledge across all four datasets, enhancing clinical interpretability and trust. Correlation-SVM provides an effective, computationally efficient framework for multicollinearity-aware feature selection in SVM-based medical diagnosis, achieving substantial feature reduction, eliminating multicollinearity, and improving classification accuracy while maintaining interpretability.

Intelligent Real-Time Exam Surveillance and Anomaly Detection Using Deep Learning with Pose-Aware Behavioral Analysis

Anuradha M. Sandi, Amulya Ratna — Tue, 23 Jun 2026 00:00:00 +0000

Maintaining academic integrity in examination halls is a key priority for educational institutions, yet traditional invigilation remains resource-intensive, subject to cognitive fatigue, and prone to oversight. While automated solutions have been proposed, they frequently rely on simple bounding box classifications that fail to capture the subtle, temporal postures associated with cheating. This paper presents an intelligent, real-time exam surveillance framework that integrates object detection, multi-object tracking, and pose-aware behavioral analysis into a unified, edge-deployable pipeline. The proposed system utilizes a fine-tuned YOLO11 Nano network to isolate students and unauthorized objects, a ByteTrack multi-object tracker to maintain persistent student identities across frames, and a secondary YOLO11-pose estimation model to extract seventeen anatomical keypoints in real time. A geometric behavior analysis engine computes head turn deflections, lateral body leaning ratios, and arm reaches based on anatomical joint vectors, passing these inputs into an exponential-decay suspicion scoring engine to filter out innocent, transient movements. Deployed as a lightweight Flask application with a real-time WebSocket dashboard, the system achieves an overall classification accuracy of 96.2 %, with an empirical precision of 89.5 %, representing a significant improvement over baseline YOLOv8 bounding-box models that suffer from excessive false alerts.

AI-Based Automated Defective Exhibit Identification System for Galleries

Pooja Patil, Abhishek Jadhav, Furqan Shaikh, Janhavi Mohite — Wed, 24 Jun 2026 00:00:00 +0000

The preservation of cultural heritage exhibits such as paintings, sculptures, and historical artifacts is a critical task for museums and galleries. Over time, these exhibits may suffer from cracks, discoloration, surface erosion, or structural damage due to environmental exposure and human interaction. Traditional inspection methods rely on manual observation by experts, which is time-consuming, subjective, and often unable to detect early-stage defects. This paper presents an AI-Based Automated Defective Exhibit Identification System for Galleries, developed as an Android application using Java/XML and Firebase Realtime Database. The system allows gallery staff to upload baseline reference images of exhibits and later compare them with newly captured images using AI-based computer vision techniques. The comparison process identifies visual deviations that indicate possible defects and automatically records them with exhibit details, timestamps, and defect type. Detected issues are logged in real time, and notifications are sent to administrators for timely review and restoration tracking. Experimental evaluation shows improved detection accuracy, reduced inspection effort, and faster maintenance response. The proposed system offers a scalable, efficient, and cost-effective solution for intelligent gallery management and digital heritage preservation.

Revolutionizing CGI and VFX with AI - Advancing Neural Rendering, Procedural Animation and Generation in Filmmaking

Abhay Kumar Mourya, Nisha Rathore — Wed, 24 Jun 2026 00:00:00 +0000

This paper examines the far-reaching influence of Artificial Intelligence (AI) on Visual Effects (VFX) and Computer-Generated Imagery (CGI), with particular attention to realism, production efficiency, and the creative latitude afforded to filmmakers. AI-driven approaches — spanning Generative Adversarial Networks (GANs), Neural Radiance Fields (NeRFs), and procedural content generation — make it possible to construct photorealistic digital worlds and hyper-detailed virtual characters while keeping computational expenditure in check. Real-time rendering pipelines, strengthened by ray tracing and AI-powered denoising, allow instant on-set feedback, compressing post-production schedules. Suitless motion capture, physics-based animation, and automated compositing further liberate artists from repetitive technical tasks. By examining these technologies through both a theoretical and case-study lens, this review reveals how AI is repositioning itself not merely as a productivity tool but as a genuine creative collaborator in the filmmaking process. Persisting challenges — deepfake misuse, workforce displacement, and the black-box nature of certain neural models — are discussed alongside a forward-looking perspective on responsible AI integration.

Enhanced Plant Disease Classification Using Deep Learning: An Attention-Guided Transfer Learning Approach

Rajan Kumar, Ritu Kadiyan — Sat, 18 Jul 2026 00:00:00 +0000

Timely and accurate identification of crop diseases from leaf images remains one of the more stubborn problems in precision agriculture, largely because the visual differences between many diseases are subtle and the imaging conditions in the field are anything but controlled. This paper presents an enhanced deep learning framework for plant leaf disease classification that couples a lightweight EfficientNet-B0 backbone with a Convolutional Block Attention Module (CBAM) and a carefully tuned training pipeline. The attention module lets the network concentrate on lesion-bearing regions of the leaf rather than spreading its capacity across the whole image, while the training pipeline — combining transfer learning, an aggressive-but-realistic augmentation strategy, label smoothing, and mixup regularisation — pushes generalisation without inflating the parameter budget. Evaluation of the model is performed on the PlantVillage dataset (54,305 images across 38 disease-and-crop categories spanning 14 species), with benchmarking against six widely used convolutional architectures: VGG16, InceptionV3, ResNet50, MobileNetV2, DenseNet121, and a plain EfficientNet-B0. The proposed model reaches a test accuracy of 99.56% with a macro-averaged F1-score of 99.54%, outperforming every baseline while keeping the parameter count at roughly 5.4 million — an order of magnitude smaller than VGG16. An ablation study isolates the contribution of each component, and Grad-CAM visualisations confirm that the attention-augmented network attends to the diseased tissue rather than the background. A deliberately unflattering cross-dataset evaluation on field-condition images is also presented, where accuracy decreases to 71.8%, highlighting the challenges and implications for real-world deployment. The results suggest that modest architectural additions, applied thoughtfully, can deliver a favourable accuracy-to-cost trade-off suitable for mobile and edge diagnosis.

MODEX: A Web-Based Resume Generation System Powered by Transformer Models and Adaptive Template Rendering

Tue, 04 Aug 2026 00:00:00 +0000

Creating an effective résumé is more difficult than it seems because it involves several important aspects: the proper writing style; proper formatting; and adding all the correct sections based on the job you are applying to. Most job seekers do not have the skill set necessary to create an effective résumé, causing many job seekers to have disorganized and ineffective résumés. To solve this problem, they have developed a web-based application known as MODEX which is built on a Node.js and Express.js backend with a React.js frontend and helps to make and improve résumés. The two main functions of MODEX are to generate a résumé from scratch via an online form and to improve existing résumés through the use of an intelligent parser that reformats an existing résumé. MODEX uses Large Language Model (LLM)-based techniques via API integration to perform resume parsing, named entity extraction, and content summarization to accomplish this task. Additionally, a React.js-based component-driven templating system allows rendering the created résumé into multiple customizable HTML layouts. All user-generated data will be collected and stored in a database for future retrieval. This paper will present the overall design of the system, the design and implementation of the RESTful service, and results from a user study with 35 participants that show how users were able to create their résumés more easily and efficiently with MODEX than they were before MODEX.

A Review of the Application of Artificial Intelligence Across Engineering Disciplines

Ukoima Kelvin Nkalo — Tue, 04 Aug 2026 00:00:00 +0000

This review examines the application of Artificial Intelligence (AI) across major engineering disciplines, including mechanical, civil, electrical, industrial, computer, energy, healthcare, agricultural, environmental, chemical, aerospace, and biomedical engineering. Using a structured narrative review, the study synthesizes findings from research published between 2020 and 2025. The review shows that AI, particularly machine learning, deep learning, reinforcement learning, and computer vision, has improved design optimization, predictive maintenance, fault detection, and system automation across these fields. However, the extent of adoption differs because of variations in data availability, risk tolerance, validation requirements, and regulatory constraints. The review also discusses key challenges, including data quality, model interpretability, integration with legacy systems, and workforce readiness. Emerging developments such as explainable AI, quantum-enhanced simulation, and human-AI collaboration are identified as promising directions for future research and practice. By providing a comparative overview of current progress, limitations, and opportunities, this paper offers a practical framework for the responsible integration of AI into engineering practice.

Artificial Intelligence Techniques for Adaptive Personalized Learning Systems

Tue, 04 Aug 2026 00:00:00 +0000

Artificial Intelligence (AI) is one of the solutions that has undergone a paradigm shift in the current education systems, providing smart solutions that enhance learning and academic efficiency. The conventional methods of study planning are based on fixed schedules, which are not responsive to personal learning capabilities, performance, and time constraints. Due to this, several students have difficulty managing their time, prioritizing their subjects poorly and using poor learning strategies. The study suggests the creation and deployment of a personalized learning planner, an AI-based study planner capable of dynamically creating adaptive study plans by analyzing student learning behavior and student performance analytics. The suggested system will combine machine learning, learning analytics, and recommendation algorithms to process student educational data and generate student-specific study plans. The system allows a dynamic and responsive learning process by constantly tracking the progress of students and reshaping their schedules. The proposed system has the following architecture: data collection, user profiling, machine learning analysis, and intelligent schedule generation modules. Empirical research indicates that Learning systems based on AI planning of studies can be used to enhance student productivity, efficiency in time management and academic achievements with significant effects. The study indicates that artificial intelligence could help revolutionize the educational planning system and facilitate personal learning processes. The further evolution of this field can incorporate new powerful deep learning solutions and real-time analytics to expand the possibilities of personalization.

AI-Driven Adaptive Tuning Systems for Real-Time Musical Instrument Optimization

Rittwik Mahmud, Ahamad Shariful Alam — Tue, 04 Aug 2026 00:00:00 +0000

This work investigates the impact of Artificial Intelligence (AI) on audio engineering and musical performance through the development of an adaptive tuning system for real-time pitch optimization. Unlike conventional tuning methods that depend on fixed references, the proposed system continuously adjusts instrument pitch in response to environmental variations, performer interaction, and tonal context. This dynamic capability enables more precise and responsive tuning during both live performances and studio recordings. The system employs a hybrid approach that combines machine learning models with advanced signal processing techniques. It analyzes incoming audio signals to detect pitch deviations and applies immediate corrective adjustments, ensuring consistent tonal accuracy. By learning from patterns in performance and environmental conditions, the system improves its responsiveness and maintains stability across varying scenarios. Experimental results demonstrate that the adaptive tuning system significantly enhances tuning accuracy while maintaining low latency, making it suitable for real-time applications. Furthermore, the system improves overall auditory quality by minimizing tuning inconsistencies without compromising musical expressiveness. The findings highlight the limitations of static tuning approaches and emphasize the advantages of intelligent, context-aware systems. By enabling continuous and automated pitch correction, the proposed method offers greater flexibility and reliability for musicians and audio engineers. This research underscores the transformative potential of AI-driven adaptive tuning technologies, providing an innovative solution that aligns with the evolving demands of modern music production and performance environments.

Algorithmic Bias in Artificial Intelligence: Reinforcing Gender Stereotypes in Digital Systems

Shilpi Saxena, Tejpal Sharma, Abhay Yadav, Aniket Sharma — Tue, 04 Aug 2026 00:00:00 +0000

Artificial Intelligence (AI) systems have transitioned from specialized computational tools to foundational socio-technical infrastructures that increasingly mediate access to employment, financial capital, public information, and digital representation. Although these automated systems are frequently framed by developers and commercial entities as objective, value-free, and neutral arbiters of human activity, a growing body of rigorous empirical evidence demonstrates that AI systems systematically reproduce, reify, and amplify existing gender inequalities. These disparities are deeply embedded within historical training datasets, algorithmic architectures, and institutional power structures. This study investigates the specific operational pathways through which contemporary digital systems reinforce regressive gender stereotypes under the guise of mathematical optimization. Utilizing a structured, interdisciplinary review of peer-reviewed research published between 2015 and 2025, combined with a cross-domain comparative analysis, this paper identifies and conceptualizes three primary mechanisms of algorithmic bias production: data-driven encoding, structural inheritance, and design homogeneity. The findings indicate that contemporary AI systems do not merely act as passive mirrors of societal prejudice; rather, they actively intensify inequalities through the compounding effects of algorithmic scale, institutional authority, and recursive data feedback loops. When biased automated outputs are deployed globally, they alter the digital ecosystem, generating new, corrupted data that feed back into future model iterations, thereby creating a self-perpetuating cycle of digital marginalization. To address these vulnerabilities, this paper outlines a comprehensive, multi-layered mitigation framework that rejects simple technical fixes in favor of a holistic approach. This framework integrates advanced technical auditing tools, deep institutional development reforms, and gender-responsive global governance protocols. Ultimately, this study demonstrates that mitigating gender bias within artificial intelligence demands a systemic transformation in how technology is conceived, built, and regulated, moving far beyond superficial dataset sanitization toward true computational justice.