VOLUME 14, ISSUE 11, NOVEMBER 2025

Abstract: In recent years, the Metaverse has emerged as a new educational environment that combines virtual and augmented reality, artificial intelligence, and collaborative interfaces, creating dynamic learning spaces that transcend the physical and temporal boundaries of the traditional classroom. In the educational context, the Metaverse is not just a technological innovation, but a pedagogical opportunity for the development of experiential, experiential, and collaborative learning. However, the success of such environments depends on the theoretical foundation of their design and an understanding of how students accept and interact with them. This study explores the relationship between Distributed Cognition theory and the acceptance of Metaverse educational worlds by secondary school students, through the theoretical framework of UTAUT (Unified Theory of Acceptance and Use of Technology). The research question was to create and validate a model that connects the theory of Distributed Knowledge with key factors of UTAUT, in order to investigate the impact of this theory on the design of virtual learning environments in the metaverse. The research provides guidelines for the design of virtual learning environments that combine cognitive artifacts, collaboration interfaces, and metacognitive tools, thus advocating the creation of distributed cognitive learning ecosystems.

Keywords: Metaverse, Distributed Knowledge Theory, UTAUT.

Abstract: Urban traffic congestion severely affects emergency vehicle response times, leading to preventable loss of life and reduced efficiency of public safety systems. Traditional traffic management relies on static signal plans and fixed routing, which cannot adapt to sudden congestion, roadblocks, or peak-hour variations. Recent research on Multi-Agent Systems (MAS), Reinforcement Learning (RL), and vehicle-to-infrastructure (V2I) coordination has shown promising results for dynamic and intelligent emergency mobility. This review analyzes ten influential studies across three domains: emergency vehicle routing algorithms, learning-based traffic signal control, and cooperative multi-agent negotiation frameworks. Prior work such as EMVLight and MARL-based traffic control demonstrates that decentralized agents can learn to reduce delays, but most systems are limited to small grids, isolated intersections, or single-agent routing. Few provide city-scale simulations integrating ambulances, fire trucks, and traffic lights within a shared communication environment. To address these gaps, this paper highlights the need for a unified, scalable simulation combining SUMO traffic modeling with SPADE-based agent communication, enabling adaptive routing, green-wave negotiation, and real-time scenario testing. The review concludes that multi-agent simulation is a practical and scalable approach for optimizing emergency response in future smart cities.

Keywords: Multi-Agent Systems, SUMO, SPADE, Reinforcement Learning, Emergency Vehicle Routing, Traffic Signal Control, Vehicle-to-Infrastructure Communication, Intelligent Transportation Systems, Q-Learning, Smart Cities

Abstract: In moment’s digital period, the verification of documents similar as instruments, identity attestations, and contracts is a major challenge due to phony, manipulation, and lack of translucency. The proposed Blockchain- Grounded Document Verification and Authentication System ensure secure storehouse, inflexible record keeping, and tamper- evidence confirmation of digital documents. The system leverages blockchain technology combined with cryptographic mincing (SHA- 256) and smart contracts to produce a decentralized verification process. Each uploaded document is converted to a unique hash and stored on the blockchain, icing authenticity and traceability. The system is designed for educational institutions, government agencies, and private associations to corroborate instruments and legal documents without interposers.

Keywords: Blockchain, Smart Contract, IPFS, SHA-256, Ethereum, Document Verification, Decentralization.

Abstract: Ransomware has become one of the most destructive and financially damaging cyber threats, capable of encrypting critical data and demanding ransom payments for recovery. Traditional antivirus solutions based on static signatures are often ineffective against newly emerging or polymorphic ransomware variants. To address these limitations, this research presents a Machine Learning (ML)–based Ransomware Detection and Mitigation Framework implemented using Python. The proposed system performs behavioral analysis of file operations and process activities, monitoring parameters such as file-write frequency, entropy levels, extension modifications, CPU utilization, and process lineage. These behavioral features are used to train a Random Forest classifier that distinguishes between normal user operations and ransomware-like activity. The trained model, integrated with Python’s Watchdog library, continuously monitors the file system in real time and automatically quarantines or isolates suspicious files upon detection. Additionally, the framework incorporates backup and recovery mechanisms that periodically create immutable file snapshots, ensuring data integrity and supporting post-attack restoration. The combination of Python’s data processing ecosystem and ML algorithms provides a scalable, adaptive, and proactive defense mechanism against evolving ransomware threats. Experimental evaluations demonstrate that the model effectively detects ransomware-like behavior with high accuracy and minimal false positives. Overall, this work contributes to a practical, intelligent, and automated cybersecurity framework for ransomware prevention, early detection, and mitigation, thereby reducing potential data loss and enhancing system resilience against modern cyberattacks.

Keywords: Ransomware, Machine Learning, Cybersecurity, Python, Random Forest, File Behavior Analysis, Data Protection, Threat Mitigation.

Abstract: In today’s digital era, the widespread use of social media and online platforms has enabled rapid dissemination of information—but also facilitated the spread of misinformation, commonly known as fake news. Such false information can distort public opinion, disrupt political processes, and cause widespread confusion. This research addresses the growing challenge of fake news detection by developing a binary classification system using machine learning techniques. The study compares the performance of two supervised algorithms—Naïve Bayes and Logistic Regression—applied to the Constraint@AAAI 2021 shared task dataset on COVID-19 fake news. The dataset underwent rigorous preprocessing, including text normalization, noise removal, stopword elimination, and TF-IDF feature extraction. Experimental results demonstrate that both models perform effectively in classifying real and fake news, with Naïve Bayes achieving an accuracy of 92.37% and Logistic Regression slightly outperforming it with 93.85%. These findings highlight the potential of lightweight machine learning models for reliable and efficient fake news detection, contributing to the fight against online misinformation and promoting trustworthy digital communication.

Keywords: Fake news, Machine Learning, Naïve bayes, Logistic Regression.

Abstract: Traffic simulation plays a vital role in evaluating and optimizing urban mobility systems. This study presents a comparative analysis of key traffic simulation parameters using the Simulation of Urban Mobility (SUMO) along with its Traffic Control Interface (TraCI) API. By adjusting parameters such as vehicle speed, acceleration, lane-changing behavior, and traffic signal timing, the research examines their impact on traffic performance of vehicles, specifically overall travel time, queue length, and vehicular throughput. The study leverages TraCI to dynamically control and monitor simulation components in real time, enabling a more detailed understanding of how individual parameters influence system behavior. Multiple traffic scenarios are tested using both static and adaptive control strategies to provide a comprehensive comparison. Results were compared to identify which parameters have the most significant effect on traffic flow. Fine-tuning these parameters how them can lead to more efficient and realistic simulations.

Keywords: Traffic Violations, Data Analytics, Over speeding, Signal Jumping, Jaywalking

Abstract: This paper presents an end- to- end intelligent system that integrates generative artificial intelligence (AI) and immersive virtual reality (VR) for automated interior design. The proposed system, called AIVID (AI Virtual Interior developer), combines procedural layout generation, rule- grounded refinement, and real- time immersive evaluation. The system accepts room figure and design constraints as input, automatically generates layout proffers using a tentative variational autoencoder (CVAE) and a graph- grounded layout refinement network, and allows druggies to fantasize, edit, and estimate designs interactively within a VR terrain. stoner feedback attained in real- time is used to acclimatize posterior design proffers. An airman study comparing AIVID with a traditional homemade design workflow demonstrated a 34 reduction in decision time, a 12- point increase in usability (SUS score), and a 0.42- point increase in stoner satisfaction. The study validates the eventuality of AI- driven immersive systems to accelerate and epitomize interior design workflows.

Keywords: Virtual Reality, Generative AI, Interior Design, Procedural Generation, Layout Optimization, Human- Computer Interaction.

Abstract: The Problem is that as data analytics for analysis the data we need to use the various tools like Power BI, Tabulate for Visualization. Also, in excel we analyze the data and visualize also. There is a tool to analysis the data but the checking of fairness in data and bias in dataset this tool is not support. The making AI powered Automated data visualization and fairness Analysis platform. It helps the platform to in HR, Healthcare, Finance. The Platform aim to that we fair all the Employees in company, the loan approval in bank. To check the statistics automatically no need to calculated separately. Keyword: Fairness in ML, AI Visualization Dataset, Automated Analytics, Detect Bias and Fairness

Abstract: Heart disease is one of the leading causes of death worldwide. Early diagnosis and prediction can play a vital role in preventing life-threatening conditions. The traditional methods for predicting heart disease are often manual, time-consuming, and prone to errors. In this research, a machine learning-based model is proposed to predict the likelihood of heart disease based on clinical data such as age, gender, blood pressure, cholesterol, and other medical attributes. Various algorithms like Logistic Regression, Decision Tree, Random Forest, and Support Vector Machine (SVM) were implemented and compared. The dataset used was the UCI Heart Disease Dataset. The results show that Random Forest Classifier achieved the highest accuracy of 88.5%, making it a reliable model for real-world applications. The study aims to assist medical practitioners in making better and faster diagnostic decisions.

Keywords: Heart disease prediction, Data mining, Risk factors, Feature selection, Real-world healthcare data, Neural network, Deep learning

Abstract: Fire detection is a crucial safety application aimed at minimizing the risk of human casualties and property loss. Conventional systems primarily rely on smoke or heat sensors, which often fail to detect fires in open or smoke-free environments. This research proposes an intelligent, Machine Learning-based vision system for real-time fire detection using computer vision techniques. The proposed model leverages a Convolutional Neural Network (CNN) trained on diverse datasets of fire and non-fire images to accurately classify fire instances from live video streams. Implemented in Python using OpenCV for image acquisition and TensorFlow/Keras for deep learning inference, the system triggers an alarm alert when fire is detected. Experimental results demonstrate over 92% detection accuracy, robust performance across varying lighting conditions, and minimal false positives. The system’s low computational cost and high responsiveness make it ideal for integration into smart surveillance, industrial safety, and IoT-based monitoring systems.

Keywords: Fire Detection, Machine Learning, Computer Vision, Convolutional Neural Network (CNN), OpenCV, TensorFlow, Real-Time Monitoring, Safety System.

Abstract: Generative Artificial Intelligence (AI) is revolutionizing education by transforming traditional teaching methods, enhancing student engagement, and enabling personalized learning experiences. This paper explores the applications, challenges, and future prospects of Generative AI in education. It highlights how large language models (LLMs) such as ChatGPT, Copilot, and Gemini are being integrated into academic systems to create adaptive learning environments and intelligent tutoring systems. Furthermore, this paper discusses the ethical implications and the potential of AI to complement human educators rather than replace them.

Keywords: Generative AI, Education Technology, Personalized Learning, ChatGPT, Artificial Intelligence, Adaptive Systems

Abstract: Deepfake images, particularly those generated through face-swapping techniques, have become increasingly realistic and widespread, raising serious concerns about digital trust, personal privacy, and public safety. As these manipulated visuals grow more sophisticated, detecting them reliably has become a pressing challenge. This paper proposes a deep learning-based approach for the automatic detection of face-swapped deepfakes using Convolutional Neural Networks (CNNs). Our method focuses on identifying subtle visual cues and inconsistencies introduced during the face manipulation process—artifacts that are often imperceptible to the human eye. To enhance detection accuracy and robustness, we integrate advanced image preprocessing, feature extraction, and data augmentation techniques. The model is trained and evaluated on widely used benchmark datasets containing a mix of authentic and manipulated images. Experimental results demonstrate high accuracy and generalization capability, reinforcing the practical value of the proposed solution for real-world applications in digital content verification. By automating the detection process, this work contributes meaningfully to the field of media forensics and supports ongoing efforts to preserve the authenticity and integrity of visual media in the age of synthetic content.

Keywords: Deepfake Detection, Face-Swapping, Convolutional Neural Networks (CNN), Image Forensics, Synthetic Media, Digital Content Verification, Image Preprocessing, Media Integrity, AI-generated Images, Feature Extraction.

Abstract: As cyber threats rapidly evolve in complexity, Gen- erative Adversarial Networks (GANs) have emerged as trans- formative tools in network security, serving both as formidable defenses and novel avenues for attack. This survey introduces a comprehensive taxonomy of GAN applications in network secu- rity, classifying contemporary research across critical domains such as adversarial sample generation, intrusion detection, syn- thetic traffic modeling, federated security architectures, IoT and edge protection, and encrypted traffic analysis. By systematically mapping these domains, the paper illustrates how diverse GAN variants are leveraged for simulating threats, resolving class imbalance, and circumventing conventional detection strategies. The taxonomy uncovers key trends in the development and deployment of GAN-driven security models, providing a robust framework for assessing progress and identifying persistent challenges. The review concludes by outlining emerging research directions rooted in the taxonomy, and calls for standardized benchmarks and ethical guidelines to support secure, scalable integration of GANs into modern network defenses. Index Terms: GANs, Network Security, Intrusion detection, Generative AI, Cybersecurity, Adversarial attacks.

Abstract: In this paper, we present an Intelligent Medicine Recommendation System with Salt Composition Analysis (IMRS-SCA), production-ready with FastAPI-based deployment of multi-label classification combined with semantic matching. We create a consolidated dataset of 253,973 Indian pharmaceutical products with compositional metadata along with 14,683 disease-drug associations aggregated from national pharmaceutical databases and clinical prescription records. Our system uses a three-tier matching pipeline extending the raw pharmaceutical attributes comprising salt composition, manufacturer, and disease indications through a normalization framework that incorporates 28+ canonical salt forms, fuzzy string matching with a threshold ≥ 0.85, and synonym-aware semantic encoding. Training was done using a Random Forest multi-output classifier with stratified train-validation-test splits in order to handle class imbalance across more than 100 disease categories. The proposed model gives F1-Score = 0.9108, Precision = 0.9269, Recall = 0.8996, and a mean confidence score = 0.91 for top-ranked recommendations on the reserved test set, outperforming baseline exact-match retrieval (Precision = 0.52). FastAPI-based deployment achieved a mean response latency of 120 ms per query under concurrent load, confirming suitability for real-time clinical decision support. In ablation studies, maximum marginal gain was observed due to the salt normalization layer, which improved alternative medicine discovery for generic substitution scenarios by 34%. The system requires no proprietary medical data, runs on commodity hardware with a <100MB model footprint, and includes comprehensive fallback mechanisms for robustness, providing a reproducible and scalable framework for pharmaceutical informatics and accessibility initiatives. Future directions will include integrating drug-drug interaction prediction, patient-specific contraindication filtering, and reinforcement learning-based dosage optimization.

Keywords: Pharmaceutical Informatics, Salt Composition Analysis, Multi-Label Classification, Medicine Recommendation, Random Forest, and Alternative Drug Discovery.

Abstract: The fifth-generation (5G) network marks a significant evolution in mobile communication technology, introducing higher data speeds, ultra-low latency, and improved connectivity compared to previous generations. This study presents an analytical examination of 5G network architecture, its core components, and the enabling technologies such as millimeter-wave communication, massive MIMO, and network slicing. The paper also highlights the transformative potential of 5G across multiple sectors, including healthcare, transportation, education, and industrial automation, where it enables real-time communication and supports large-scale Internet of Things (IoT) applications. Globally, 5G technology is driving economic growth, supporting digital transformation, and fostering innovations such as autonomous vehicles and smart city infrastructures. However, the study also analyzes the challenges related to spectrum allocation, cybersecurity threats, and the high cost of infrastructure deployment that limit its rapid adoption in developing regions. By evaluating both the opportunities and obstacles, this paper aims to provide a comprehensive understanding of 5G’s global impact, emphasizing its role in shaping a more connected, intelligent, and sustainable digital future. The introduction of 5G technology represents a revolutionary step in global communication. It is designed to deliver extremely high data rates, low latency, and seamless connectivity among billions of devices. Unlike its predecessors, 5G is not just an upgrade in speed—it is an enabling platform for emerging technologies such as the Internet of Things (IoT), Artificial Intelligence (AI), autonomous vehicles, and smart cities. This paper analyzes the architecture of 5G, its deployment status, and its global impacts on economic, industrial, and social sectors. Using analytical methods and secondary data, the study examines the benefits and challenges of 5G adoption across countries. The findings suggest that 5G has a transformative role in driving global economic growth and digital transformation, but sustainability, energy use, and security concerns must be addressed for long-term success.

Keywords: 5G Network, IoT, URLLC, eMBB, mMTC, Smart Cities, Digital Transformation, Global Economy, Cybersecurity, Industry 4.0.

Abstract: This paper addresses the critical challenge of conducting digital investigations in an era of ubiquitous data and stringent privacy regulations. Open-Source Intelligence (OSINT) provides powerful capabilities for gathering information, with the mobile number serving as a unique digital identifier and a crucial pivot point. However, existing tools are often fragmented, lack integrated ethical guardrails, and risk violating legal frameworks like the GDPR. We propose a novel, integrated software architecture for a mobile number intelligence tool designed on the principle of "Ethical-by-Design." The framework features a modular architecture comprising data collection, processing, AI-augmented analysis using an on-premise Large Language Model (LLM) to ensure operational security, and a visualization dashboard. We present a structured investigative workflow that embeds legal compliance checks, such as documenting the lawful basis for processing, directly into the process. Through conceptual case studies in law enforcement and cybersecurity, we demonstrate the framework's efficacy in generating actionable intelligence while adhering to principles of data minimization and proportionality. This research contributes a blueprint for the next generation of OSINT tools that harmonize advanced investigative capabilities with the fundamental right to privacy.

Keywords: Data privacy, digital forensics, ethical hacking, mobile number intelligence, open-source intelligence (OSINT).

Abstract: The integrity, confidentiality, and availability of information transmitted across interconnected computer systems. As digital transformation expands globally, the growing dependence on cloud computing, Network security represents the collective measures and protocols established to protect mobile connectivity, and the Internet of Things (IoT) has introduced complex cybersecurity challenges. Modern attackers leverage Artificial Intelligence (AI), ransomware-as-a-service, and advanced persistent threats (APTs) to breach traditional defenses. This paper explores the layered architecture of network security, the evolution of encryption techniques, and the development of proactive intrusion detection and prevention systems. Through a detailed discussion of case studies, it also examines the role of emerging technologies such as blockchain and quantum cryptography in redefining future cybersecurity paradigms.

Keywords: Network Security, Cryptography, IDS/IPS, Cybersecurity, Firewalls, AI, Blockchain, Quantum Cryptography

Abstract: In the modern era of big data, organizations require rapid insights from continuously generated data streams. Real-time data analytics has become essential for decision-making in sectors such as finance, healthcare, IoT, and social media. Apache Spark, a powerful open-source distributed data processing framework, provides in-memory computation and supports both batch and stream processing. This paper explores the use of Apache Spark for real-time data analytics, focusing on its architecture, components, and advantages over traditional frameworks like Hadoop MapReduce. Through integration with tools such as Apache Kafka and HDFS, Spark enables scalable, fault-tolerant, and low-latency processing. Experimental analysis shows Spark’s capability to handle large-scale, high-velocity data with minimal delay, offering significant improvements in throughput and processing speed. The results confirm that Apache Spark is a highly efficient and scalable platform for real-time big data analytics.

Keywords: Real-Time Analytics, Apache Spark , Stream Processing, Kafka, Hadoop

Abstract: Efficient traffic management is a critical challenge in modern urban environments due to increasing vehicle density and dynamic traffic patterns. This paper presents a Smart Traffic Control System that integrates pixel-labeling for precise vehicle segmentation with the Simple Online and Realtime Tracking (SORT) algorithm for accurate and efficient vehicle tracking. The proposed framework leverages computer vision techniques to detect, classify, and monitor vehicles in real-time from video feeds, eliminating the need for expensive sensor-based infrastructure. Pixel-labeling enables semantic understanding of the scene by assigning class labels to each pixel, allowing robust differentiation between vehicles, pedestrians, and background elements. The SORT tracker further enhances system performance by maintaining consistent object identities across frames, even under occlusion and varying lighting conditions. Experimental evaluations demonstrate that the system achieves high detection accuracy, reduced processing latency, and improved traffic flow estimation compared to traditional methods. The results suggest that the proposed approach provides a scalable, cost-effective, and adaptive solution for intelligent traffic control in smart city applications.

Keywords: smart traffic control system, SORT Tracker, pixel labeling.

Abstract: Serverless computing has redefined cloud ap- plication deployment by abstracting infrastructure and enabling on-demand, event-driven execution, thereby en- hancing developer agility and scalability. However, main- taining consistent application performance in serverless environments remains a significant challenge. The dynamic and transient nature of serverless functions makes it difficult to distinguish between benign and anomalous behavior, which in turn undermines the effectiveness of traditional anomaly detection methods. These conventional approaches, designed for stateful and long-running ser- vices, struggle in serverless settings where executions are short-lived, functions are isolated, and observability is limited. In this first comprehensive vision paper on anomaly detection for serverless systems, we systematically explore the unique challenges posed by this paradigm, including the absence of persistent state, inconsistent monitoring granularity, and the difficulty of correlating behaviors across distributed functions. We further examine a range of threats that manifest as anomalies, from classical Denial- of-Service (DoS) attacks to serverless-specific threats such as Denial-of-Wallet (DoW) and cold start amplification. Building on these observations, we articulate a research agenda for next-generation detection frameworks that ad- dress the need for context-aware, multi-source data fusion, real-time, lightweight, privacy-preserving, and edge-cloud adaptive capabilities. Through the identification of key research directions and design principles, we aim to lay the foundation for the next generation of anomaly detection in cloud-native, serverless ecosystems.

Keywords: Serverless Computing, Cloud Computing, Edge Computing, Function-as-a-service, Anomaly Detec- tion, DoS, Data Fusion, System Monitoring, Observability.

Abstract: The increasing deployment of IoT devices has heightened the need for effective security mechanisms to identify malicious activities within network traffic. With the rapid growth of IoT devices, safeguarding networks against malicious traffic has become increasingly critical. This study develops an intrusion detection system (IDS) using the UNSW-NB15 dataset, applying both supervised and unsupervised learning techniques. The dataset was preprocessed through feature selection, encoding, and cleaning, followed by exploratory analysis to reveal class imbalance and key traffic characteristics. A Long Short-Term Memory (LSTM) model was trained for binary classification of normal versus attack traffic, while a Bayesian Gaussian Mixture Model (BMM) was applied for anomaly detection using normal data. Evaluation employed accuracy, precision, recall, F1-score, ROC curves, and Youden’s Index for optimal threshold selection. Results show the LSTM delivered strong classification performance, while the BMM provided effective anomaly detection when thresholds were optimized. These findings highlight the potential of combining deep learning and probabilistic models to enhance IDS performance and strengthen network security.

Keywords: Intrusion Detection System (IDS), IoT Security, Network Traffic Analysis, UNSW-NB15 Dataset, Long Short-Term Memory (LSTM), Bayesian Gaussian Mixture Model (BMM),Anomaly Detection

Abstract: Cyber threats and online fraud have become critical challenges in the digital era. Traditional security systems such as firewalls and signature-based methods are insufficient to counter increasingly sophisticated attacks including malware, phishing, ransomware, and fraudulent transactions in online shopping platforms. Artificial Intelligence (AI) and Machine Learning (ML) offer predictive, adaptive, and intelligent solutions capable of detecting cyber threats in real-time. This paper provides a comprehensive review of AI/ML techniques for cyber threat and fraud detection, explores their applications in online shopping platforms, discusses commonly used datasets and evaluation metrics, and highlights emerging trends and future directions for research.

Keywords: Cybersecurity, Fraud Detection, Artificial Intelligence, Machine Learning, Online Shopping, Anomaly Detection, Predictive Security.

Abstract: The Internet of Things (IoT) represents a transformative paradigm in modern computing, enabling seamless connectivity between billions of physical devices through the internet. This research paper provides a comprehensive analysis of IoT concepts, underlying technologies, architectural frameworks, and real-world applications across diverse sectors. The study examines the four-layer IoT architecture comprising the sensing layer, network layer, data processing layer, and application layer, along with key communication protocols such as MQTT and CoAP. Furthermore, this paper explores practical implementations in smart cities, healthcare, agriculture, and manufacturing industries, demonstrating how IoT solutions enhance operational efficiency, sustainability, and quality of life. Security challenges and future trends, including edge computing, 5G integration, and artificial intelligence convergence, are critically analyzed. The findings reveal that IoT technology is rapidly evolving from experimental

Keywords: Internet of Things, IoT Architecture, Smart Cities, Healthcare Monitoring, Precision Agriculture.

Abstract: The MSBTE Guide website serves as a comprehensive educational platform catering to the needs of students pursuing various courses under the Maharashtra State Board of Technical Education (MSBTE). This multifaceted online resource is meticulously designed to provide a wealth of academic support, offering a diverse range of content crucial for exam preparation and coursework enhancement. At the core of the website's offerings are the previous year question papers, a treasure trove of invaluable resources that empower students with insights into the examination patterns, question formats, and key topics. These papers act as an indispensable tool for honing problem-solving skills and fostering a deeper understanding of the subject matter.in addition to the question papers, the platform hosts an extensive collection of comprehensive notes meticulously crafted to complement the curriculum. These notes serve as a concise yet thorough guide, aiding students in grasping intricate concepts, revising crucial topics, and reinforcing their overall understanding of the subject. The website also acts as a repository for syllabus information, ensuring that students have easy access to the prescribed curriculum for their respective courses. This feature is particularly beneficial for individuals looking to streamline their study plans and focus on the specific areas outlined by the MSBTE.

Keywords: Pothole detection, Web application, Urban maintenance, Infrastructure reporting, Node.js, MongoDB, Citizen feedback system

Abstract: Generative Artificial Intelligence (AI) represents one of the most significant breakthroughs in modern computer science. It enables machines to create original content—text, images, audio, video, and even code—that resembles human creativity. Over the past few years, rapid advancements in deep learning architectures have turned generative models into tools of automation, innovation, and exploration. This paper explores how generative AI is revolutionizing industries such as education, healthcare, entertainment, and business analytics. Furthermore, it examines the ethical, social, and technical challenges associated with its implementation. The paper concludes that the long-term sustainability of generative AI depends on transparency, accountability, and the establishment of strict ethical frameworks.

Keywords: Generative AI, Deep Learning, Machine Learning, Automation, Ethics, Innovation, Creativity

Abstract: The rapid growth of cyber threats has left traditional signature-based malware detection methods less effective against new and complex attacks. To address this issue, the AI-Powered Malware Detection System identifies malicious software using machine learning, focusing on behavior instead of static signatures. Developed in Python, the system detects and classifies malware and non-malicious software using algorithms like Random Forest and XG Boost. By training on datasets such as the Microsoft Malware dataset and CIC-MalMem, the model identifies complex patterns in system behavior, including file operations, network activity, and process interactions associated with malware. The features extracted are then processed to create a high-performance model that detects malware with low false positives. This system is also resilient to future variant developments, making it more effective than traditional methods. With applications in cybersecurity defense systems, enterprise IT infrastructure, and cloud security, this paper enhances proactive malware detection and improves system resilience against cyberattacks.

Keywords: Malware, cloud security, cyberattacks, Ai, machine learning, cybersecurity.

Abstract: Recent advancements in computer vision and image processing have significantly transformed healthcare, enhancing diagnostic precision, reducing costs, and improving efficiency. Among medical imaging techniques, Magnetic Resonance Imaging (MRI) stands out for its capability to identify even minute brain abnormalities. This study presents a comparative analysis of two advanced object detection models, YOLOv5 and YOLOv7, for brain tumor detection and classification using MRI scans. The dataset includes three major tumor categories—meningioma, glioma, and pituitary tumors. Preprocessing techniques and mask alignment methods were applied to enhance segmentation accuracy before model training. Experimental evaluation shows YOLOv5 achieved a recall of 0.905 for box detection and 0.906 for mask segmentation, with a precision of 0.94 and 0.936 respectively. At an IoU threshold of 0.5, it attained a mean Average Precision (mAP) of 0.947, while YOLOv7 achieved slightly higher accuracies with 0.936 and 0.935 in detection and segmentation. YOLOv7 also produced better mAP scores across varying IoU ranges. Comparative analysis with traditional models such as RCNN, Faster RCNN, and Mask RCNN further confirms the efficiency and reliability of YOLO-based architectures for accurate brain tumor identification.

Keywords: Brain Tumor, Deep Learning, Image Processing, MRI, YOLO, Object Detection, Segmentation, mAP, RCNN.

Abstract: This paper presents an emotion-aware approach to movie genre classification that leverages the linguistic and affective patterns embedded within dialogues. Unlike conventional genre prediction models that depend on metadata or plots, this study utilizes a fusion of lexical and sentiment-based features to predict movie genres. The system combines TF–IDF representations with emotion cues derived from VADER sentiment analysis, thereby enhancing contextual and affective understanding. Multiple machine learning models, including Naive Bayes, Logistic Regression, Linear Support Vector Machine (SVM), and an Ensemble classifier, were trained and compared. The best-performing model, an Ensemble combining Logistic Regression and SVM, achieved an overall accuracy of 53.23% across ten genres. The findings demonstrate that emotion-informed textual features significantly enhance the accuracy and interpretability of movie genre classification systems.

Keywords: Movie dialogues · Emotion analysis · Genre classification · Sentiment features · Machine learning · Natural Language Processing (NLP)

Abstract: Artificial Intelligence (AI) has rapidly become an integral part of daily life, influencing nearly every sector from communication to healthcare. This paper explores the diverse applications of AI in everyday environments and examines how these technologies improve efficiency, accuracy, and decision-making. The study discusses AI's impact in smart homes, education, finance, healthcare, and transportation, emphasizing real-world use cases like voice assistants, predictive text, and autonomous vehicles. Moreover, it addresses ethical and societal challenges such as data privacy, job automation, and algorithmic bias. The purpose of this paper is to provide a comprehensive understanding of AI’s transformative role in modern society and to analyze its potential future advancements that could shape human lifestyles in the coming decades.

Keywords: Artificial Intelligence, Machine Learning, Deep Learning, Automation, Smart Systems, Everyday Applications

Abstract: Learning Management System (LMS) has been a widely-used learning media so a study is required to know the trend of its development. The present study aimed to analyze the types of documents, languages, con-tributing countries, top affiliates, sponsorship funding, top productive authors, research citations, subject areas, top source titles, trend mapping visualization, and top-cited 100 publications, and review some publications on LMS research during 1991–2021 using bibliometric analysis. The metadata were obtained by Scopus database and analyzed by VOS Viewer within 2.689 documents. The bib-biometric analysis results showed that LMS research had conference papers as the most widely published document type and English was the most commonly used language. The country with the most publications was the United States of America. National Natural Science Foundation of China became the top funding sponsor. The top affiliate was Bina Nusantara University. The most productive authors were Sabine Graf. Top cited author achieved by Fred D. Davis, and the top subject areas were Computer Science. Then, Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics became the title of the top source. Trends of LMS research in 1991–2021 were: 1) related to E-learning; 2) implementation of learning active-ties and student-teacher cases; 3) technology integration in learning; 4) distance learning; 5) technology education; 6) online learning environment; and 7) inter-active learning environment.

Keywords: Learning Management Systems, E-learning, Online Education, Educational Technology, Personalized Learning, Student Engagement, LMS Development, Artificial Intelligence, Mobile Learning, Intelligent Teaching System (ITS), Gamification.

Abstract: In the modern era, healthcare has become one of the most critical sectors demanding innovation and technological intervention. Artificial Intelligence (AI) and the Internet of Things (IoT) have emerged as transformative technologies that can help provide proactive healthcare solutions. The Smart Health Monitoring System using AI aims to continuously monitor vital health parameters of individuals such as body temperature, heart rate, oxygen level (SpO₂), and blood pressure. These parameters are collected through IoT-based wearable sensors and analyzed using AI algorithms to detect early signs of abnormalities or diseases. The system automatically generates alerts to healthcare professionals or family members when abnormal readings are detected. This AI-enabled solution helps in early diagnosis, preventive healthcare, and reduction in hospital readmissions. Keyword: Smart Health Monitoring, Artificial Intelligence (AI), Internet of Things (IoT), Machine Learning, Real-time Health Tracking, Remote Patient Monitoring, Predictive Healthcare, Biomedical Sensors, Data Analytics, Disease Prediction, Telemedicine, Deep Learning, Wearable Devices, Healthcare Automation, Health Data Security.

Abstract: This document gives formatting instructions for authors preparing papers for publication in the Proceedings of an International Journal. The authors must follow the instructions given in the document for the papers to be published. You can use this document as both an instruction set and as a template into which you can type your own text.

Keywords: Generative Artificial Intelligence (AI) by IoT.

Abstract: Diabetes is a diseases that affect the body’s ability to produce or use insulin, a hormone that regulates blood sugar or glucose levels .Diabetic Retinopathy (DR) is an eye disease in humans with diabetes which may harm the retina of the eye and may cause total visual impairment. Therefore it is critical to detect diabetic retinopathy in the early phase to avoid blindness in humans. Our aim is to detect the presence of diabetic retinopathy by applying Machine learning algorithms. Hence we try and summarize the various models and techniques used along with methodologies used by them and analyze the accuracies and results. It will give us exactness of which algorithm will be appropriate and more accurate for prediction. Machine learning consists of a number of stages to detect retinopathy in the images that includes converting image to suitable input format, various preprocessing techniques. It also includes training a model with a training set and validating with a different testing set. Method proposed in this project is Resnet 152.Berfore applying alorithum retinal images must be Preprocessing, and Feature Extraction. First, the images are preprocessed. They are converted. Proper resizing of image is also done. As the images are heterogeneous they compressed into a suitable size and format. Data set used for this project is taken from Kaggle. The main objective of this work is to build a stable and noise compatible system for detection of diabetic retinopathy.

Keywords: Machine learning, Diabetic Retinopathy, Resnet-152

Abstract: Accurate estimation of energy expenditure and calories burned during exercise is essential for fitness tracking and health monitoring. Reliable calorie estimation enables professionals to design personalized fitness plans and helps individuals optimize their workouts. This study proposes a machine learning approach to predict calories burned based on physiological and exercise-related features such as gender, age, height, weight, exercise duration, heart rate, and body temperature. Several ensemble regression models are employed, including Gradient Boosting Decision Trees Regression (GBDTR), Extreme Gradient Boosting Regression (XGBOOSTR), Stacking Regression (STACKINGR), Random Forest Regression (RFR), Bagging Regression (BAGGINGR), and Voting Regression (VOTINGR). Among these models, XGBOOSTR demonstrates the highest performance with a Mean Squared Error (MSE) of 14.224, Mean Absolute Error (MAE) of 2.022, R-squared (R²) value of 0.9964, Peak Signal to Noise Ratio (PSNR) of 37.41, and Signal to Noise Ratio (SNR) of 29.29. Explainable Artificial Intelligence (XAI) techniques, including Local Interpretable Model Agnostic Explanations (LIME) and Shapley Additive Explanations (SHAP), are applied to interpret model predictions and identify the most influential features, such as exercise duration, heart rate, and body temperature. The findings of this research provide valuable insights for developing wearable health applications, enhancing personalized fitness tracking, and assisting medical professionals in promoting healthier lifestyles.

Keywords: Machine Learning, XAI, Regression, Prediction.

Abstract: The ServiceHub Application: Comprehensive Solutions for Every Home Need Using Android is designed to make everyday life easier by connecting people with reliable service providers for their household needs. In today’s fast-paced world, finding trustworthy professionals for basic home services like plumbing, electrical work, cleaning, and maintenance can often be frustrating and time-consuming. Service Hub aims to solve this problem by offering a simple, accessible, and efficient mobile platform where users can search for nearby service providers, compare their ratings, and book appointments instantly through their smartphones.

Abstract: The fashion industry is rapidly evolving with the integration of artificial intelligence (AI) technologies. The AI Fashion Hub project aims to provide users with a smart platform that recommends fashion outfits based on personal preferences, current trends, and visual inputs. By applying machine learning and image recognition techniques, the system analyzes clothing patterns, colors, and styles to deliver personalized outfit suggestions. The project demonstrates how AI can revolutionize fashion retail by enhancing user experience, promoting digital styling, and improving online shopping decisions.

Keywords: Artificial Intelligence, Fashion Recommendation, Machine Learning, Deep Learning, Image Recognition, Personalization.

Abstract: Agriculture is a vital sector that directly influ- ences a country’s economy and food security. Crop pre- diction plays an important role in helping farmers make informed decisions about sowing, irrigation, and harvest- ing. Traditional prediction methods are often inaccurate due to changing weather, soil variations, and pest attacks. Artificial Intelligence (AI) techniques such as Machine Learning (ML) and Deep Learning (DL) provide new ways to analyze complex agricultural data for better yield prediction. This review paper discusses recent research developments in AI-based crop prediction systems, com- monly used algorithms, data sources, methodologies, and challenges. It also highlights the emerging trends such as IoT-based smart farming, real-time data processing, and the integration of AI with remote sensing technologies for more accurate and scalable prediction models

Keywords: Artificial Intelligence (AI), Machine Learning, Deep Learning, Crop Prediction, Precision Agriculture, Re- mote Sensing, IoT Sensors, Weather Data Analysis, Yield Forecasting, Smart Farming.

Abstract: The automotive market’s increasing digitalization necessitates accurate, data-driven vehicle valuation models to en- hance market transparency and support strategic decision-making. This paper presents a machine learning framework designed to predict car prices based on a comprehensive set of technical and physical specifications. The core of this research is a comparative analysis of a baseline Linear Regression model against a more sophisticated Random Forest Regressor to evaluate their predictive efficacy. Using a structured dataset of over 200 vehicle records, our methodology incorporates a robust preprocessing pipeline, including one-hot encoding for categorical features and standard scaling for numerical attributes. The empirical results demonstrate the superior performance of the Random Forest model, which achieved a coefficient of determination (R2) of 0.96, alongside a Root Mean Squared Error (RMSE) of 1791.80 and a Mean Absolute Error (MAE) of 1251.66. Feature importance analysis reveals that engine size and horsepower are the most significant determinants of vehicle price. This framework serves as a foundational tool for broader business ap- plications, including the subsequent forecasting of sales volumes and revenue, thereby offering a scalable solution for stakeholders across the automotive industry.a

Keywords: Car Price Prediction, Machine Learning, Random Forest, Regression Analysis, Feature Importance, Automotive Analytics, Revenue Forecasting.

Abstract: Cricket, being a data-intensive sport, offers a substantial opportunity for the application of machine learning in predictive analytics. This study employs data-driven machine learning methodologies to predict match outcomes in the Indian Premier League (IPL). To create the predictive models, data from the IPL from 2008 to 2024 was collected and prepared. This data included information about team stats, player performance, venue details, and toss results. We used and tested several algorithms, such as Logistic Regression, Random Forest, and XGBoost, to see how well they could predict the chances of a team winning. The XGBoost model did the best, with an accuracy rate of about 78%. This was better than traditional models, mostly because it was better at dealing with the non-linear relationships between match features. The system does more than just predict who will win a match; it also gives clear information about what factors have the biggest impact on how well a team plays. This study shows how machine learning could help analysts, coaches, and fans make strategic decisions, play fantasy sports, and comment on games.

Keywords: Machine Learning, IPL Prediction, Sports Analytics, XGBoost, Cricket Match Outcome, Data-Driven Decision Making.

Abstract: Automated testing methods have evolved from expensive, proprietary "bed of nails" and "flying probe" testers to sophisticated Automated Optical Inspection (AOI) systems and software-based solutions. The rise of open-source hardware platforms like the NodeMCU (based on ESP8266/ESP32 microcontrollers) and powerful programming languages like Python has made it possible to create highly customized, cost-effective, and flexible automation systems. This project leverages these open-source tools to build an accessible and efficient automated bare board testing solution.

Keywords: PCB Testing, Automated Optical Inspection, PCB testing using AI, Hybrid test method, In circuit testing;

Abstract: This paper presents the design and development of Autonomous Drone based Delivery System to replace the traditional delivery system with Unmanned Aerial Vehicle (UAV). With rapidly growing technology automation have become the new normal and everyone is using it in their day to day work. The demand for faster delivery system which will be efficient, fast, conventional as well as environmentally sustainable have increased rapidly. Hence Autonomous Drone Delivery System serves as an alternative to all these things and provides a solution which is efficient and environmentally sustainable as well as will have minimal human interference in it. The core objective of this project is to implement a delivery system which is reliable, efficient and have a delivery system integrated claw mechanism that can carry lightweight package to a specific destination autonomously.

Keywords: Autonomous Drone Delivery System, environmentally sustainable, claw mechanism, UAV

Abstract: Credit scoring plays a crucial role in the financial sector, helping institutions assess both the repayment ability and risk profile of borrowers. Over the years, machine learning has brought major improvements to this process. However, many existing models still face challenges related to data quality, interpretability, and genuine predictive stability. This study proposes a practical machine learning framework that combines automated data correction, guided by expert domain knowledge, with powerful ensemble-based learning techniques. The system achieves complete classification accuracy on a real-world credit dataset, marking a significant step forward in data-driven lending analysis. By integrating traditional banking logic with modern supervised algorithms, the framework ensures highly accurate predictions, clear interpretability, and robust financial outcomes. Experimental analysis confirms that proper data refinement and consensus modeling can effectively distinguish between reliable and risky borrowers. The proposed approach can serve as a foundation for future AI-driven credit scoring systems that meet both operational and regulatory expectations.

Keywords: Credit Scoring, Machine Learning, Ensemble Learning, Domain Expertise, Data Correction, Feature Engineering, Explainable AI, Uncertainty Quantification, FinTech, Predictive Modeling, Supervised Learning, XGBoost, LightGBM, Random Forest, Gradient Boosting, Financial Inclusion, Risk Assessment, Credit Risk Modeling, Model Interpretability.

Abstract: Add-on courses in higher education can bridge skill gaps and enhance student employability, yet students often struggle to identify offerings aligned with their backgrounds, interests, and career goals. In higher education, add-on courses play an important role in improving students’ skills and employability. However, students often face difficulty in selecting suitable courses from a wide range of options. The research paper highlights the importance of add-on courses in higher education for improving student skills and employability. The study proposes a recommendation system to guide students in selecting appropriate courses. The system’s design is based on the analysis of student preferences for delivery modes, course formats and career aspirations which helps to provide insights for developing flexible, relevant, and student-centric programs. Overall, the research suggests that academic institutions should design flexible and industry-aligned add-on programs to meet student expectations.

Keywords: Add-on courses, Higher Education, Recommendation System, K-means Clustering, Collaborative Filtering

Abstract: Managing queues and handling guests efficiently have become essential for restaurants aiming to deliver smooth and satisfying dining experiences. Traditional walk-in and manual reservation systems often lead to long waiting times, confusion during peak hours, and an overall decline in service quality. This paper presents a smart queue and guest management system designed to simplify restaurant operations through the integration of IoT sensors, cloud computing, and artificial intelligence. The system enables guests to join a virtual queue using a mobile application or self-service kiosk, view real-time waiting updates, and receive digital notifications when their table is ready. At the same time, predictive algorithms help restaurant managers allocate seating and staff resources more effectively. Experimental testing in a simulated environment showed a significant reduction in customer waiting times and improved table-turnover efficiency. The proposed approach demonstrates how digital queue systems can modernize restaurant service, reduce operational stress, and enhance overall customer satisfaction.

Keywords: Augmented Reality, Artificial Intelligence, Virtual Try-On, Fashion Technology, 3D Body Modeling, Clothing Simulation

Abstract: Cardiac arrhythmia is a common cardiovascular disorder that results from abnormalities in the electrical conduction system of the heart, leading to irregular heartbeat patterns. Accurate and timely detection of arrhythmia is crucial for effective diagnosis and treatment, yet manual interpretation of electrocardiogram (ECG) signals remains a challenging and time-consuming process due to the complex, dynamic, and non-stationary nature of ECG data. This study proposes a robust automated deep learning framework for the classification of ECG signals into three clinically significant categories: cardiac arrhythmia (ARR), congestive heart failure (CHF), and normal sinus rhythm (NSR). ECG data were obtained from the publicly available MIT-BIH and BIDMC databases on PhysioNet and underwent a comprehensive preprocessing pipeline that included noise removal, normalization, and segmentation to ensure data quality and consistency. Two pretrained convolutional neural network architectures, ResNet-50 and AlexNet, were fine-tuned using transfer learning techniques to leverage their deep hierarchical feature extraction capabilities for ECG classification. The models were trained and validated using a stratified dataset, and their performance was assessed through a multi-class confusion matrix employing evaluation metrics such as accuracy, precision, recall, sensitivity, specificity, and F-measure. Experimental results demonstrated that the proposed deep learning model achieved outstanding performance with an overall classification accuracy of 99.2%, average sensitivity of 99.2%, specificity of 99.6%, and precision, recall, and F-measure all at 99.2%. These results indicate that the model can effectively differentiate between normal and pathological cardiac conditions with high reliability. In conclusion, the proposed system offers a powerful and efficient tool for automated arrhythmia detection, significantly reducing diagnostic time and minimizing errors associated with manual ECG interpretation, thereby supporting clinicians in the rapid and accurate diagnosis of cardiac disorders.

Keywords: Electrocardiogram (ECG), Deep learning (or deep neural network), Convolutional Neural Network (CNN) model, ARRHYTHMIAS, accuracy, Time Frequency Representations, ResNet50, AlexNet and Morse Wavelet.

Abstract: This paper presents a Convolutional neural network-based automated waste segregation system. For efficient waste management. Food waste, metal, plastic, and paper are the four categories into which the model divides waste images. It uses a deep learning technique to classify images. Training and testing are conducted using the Waste Segregation Large Dataset from Kaggle, which includes more than 56,000 labeled images. To efficiently extract and classify features, the CNN architecture includes multiple convolutional, pooling, and dense layers. The suggested system's consistent accuracy demonstrates CNNs' high level of precision in waste segregation automation. Additionally, a variety of data improvement techniques are applied to lessen overfitting and boost the model's generalization ability. The model's strength is ensured by assessing its performance using metrics such as accuracy, precision, recall, and F1-score. The system can integrate into smart waste management setups, where cameras and sensors automatically capture and classify waste in real time. This automated process reduces manual work and human error while optimizing recycling tasks by ensuring precise sorting. Overall, the proposed approach helps promote sustainable recycling, conserve resources, and support cleaner urban areas.

Keywords: Waste Segregation · Convolutional Neural Network · Image Classification · Deep Learning · Smart Waste Management

Abstract: In domestic applications, induction motor is very important. We are going to use very efficiency while change in speed of induction motor with respect to change in the firing angle of the TRIAC. We are using the ESP8266 chip which will be mounted on PCB board and manually on the board we are going to control speed as well as display characteristics. Also by using Blynk IOT we are going to control the speed.

Keywords: ESP8266, AC Motor, Speed Control, Blynk Application, ThingSpeak, TRIAC, Optocoupler, Automation, Smart System, Remote Monitoring

Abstract: This paper is about Sahayak, an AI teaching assistant to give teachers a leg up in India's under-funded, multi-grade classrooms. The system works with Google AI (Gemini, Vertex AI, and Firebase) to help with making content, dealing with different learning levels, heavy workloads, and getting access to info. Sahayak is meant to be a go-to, cloud-based helper that makes lesson planning easier, creates local content in native languages, and makes different worksheets for various grade levels. The method is iterative and modular, focusing on user design, being able to handle growth, working with low bandwidth, and sticking to ethical AI, including the Digital Personal Data Protection (DPDP) Act 2023. By cutting down on busywork and giving fast, learning stuff, Sahayak wants to make teaching better and help students do well.

Keywords: Artificial Intelligence (AI), Multi-Grade Teaching, EdTech, Gemini, Vertex AI, Content Differentiation, Localization, Teacher Empowerment, Firebase.

Abstract: This paper presents a data-driven framework for intelligent real estate price prediction and investment recommendation using machine learning techniques. Unlike conventional valuation methods that rely on manual appraisals and subjective judgments, the proposed system leverages structured datasets containing attributes such as property size, location, area, and the number of rooms. The model integrates Linear Regression and Random Forest algorithms to enhance prediction accuracy and provide reliable valuation insights. Additionally, the system offers investment recommendations based on predictive analysis, thereby assisting buyers and investors in informed decision-making. Comparative evaluation of the models demonstrates that the Random Forest approach outperforms Linear Regression in terms of accuracy and stability. The results indicate that the proposed Smart Estate system can significantly improve transparency, minimize pricing bias, and modernize real estate transactions in the Indian market.

Keywords: Real Estate · Machine Learning · Price Prediction · Investment Recommendation · Regression Models

Abstract: This paper presents a comprehensive research paper on gesture-guided unmanned aerial vehicle ( UAV ) control systems utilizing the ESP32 microcontroller and MPU6050 inertial measurement unit (IMU). The integration of wearable gesture recognition with wireless drone control offers intuitive, hands-free operation for aerial vehicles. Through detailed analysis of hardware architecture, sensor fusion algorithms, machine learning approaches, and real-time signal processing, this work demonstrates a cost-effective alternative to traditional remote control interfaces. Key findings show that gesture recognition systems combining accelerometer and gyroscope data achieve recognition accuracies exceeding 98 percentage with processing latencies under 50 milliseconds. The hybrid approach utilizing 1D convolutional neural networks and LSTM architectures enables both static and dynamic gesture classification suitable for real-time drone navigation. This research establishes foundational principles for accessible, intuitive aerial vehicle control systems with applications extending to entertainment, surveillance, search-and-rescue, and emergency response operations.

Keywords: Gesture Recognition, Smart Glove, ESP32, MPU6050, UAV Control, Human-Machine Interaction, Wearable Sensors

Abstract: Computer vision and pose estimation are essential in analyzing exercise form quality. Tools like MediaPipe can enhance these analyses by providing real-time feedback on body posture. Machine learning techniques, such as random forest algorithms, can be employed to evaluate exercise performance. Cosine similarity can help in comparing different exercise postures to determine alignment and efficiency. Diabetes mellitus is a chronic metabolic disorder where insulin production or effectiveness is compromised. This leads to elevated blood glucose levels, which can result in various health complications. Understanding the role of insulin is crucial for managing diabetes and maintaining overall health. Regular exercise and proper form can significantly contribute to better metabolic control. .

Keywords: Computer Vision · Machine Learning · Predictive Analysis · Diabetes Mellitus · Random Forest

Abstract: Artificial Intelligence (AI) has revolutionized the way humans interact with machines by creating systems that can understand and respond to natural language. This paper presents the design and implementation of a simplified, cost-effective conversational AI prototype named “MindMesh.” The objective of this research is to develop a local, real-time AI assistant that utilizes natural language processing (NLP), speech recognition, and pseudo-holographic projection to enable a screenless conversational experience. Unlike advanced systems requiring high-end computation or expensive sensors, MindMesh demonstrates that intelligent interaction can be achieved using affordable tools like Python libraries, Raspberry Pi, and mini projectors. The model is designed primarily for educational demonstration, research, and practical exposure to AI concepts within limited resources. The paper also explores AI integration techniques, display mechanisms, and implementation challenges encountered during the prototype’s development.

Keywords: ai Engineering, Touchable Holograms, Screenless Display, Volumetric Display, Holographic Interface, Mid-Air Interaction, Natural Language Processing (NLP), Human-Computer Interaction (HCI), Laser Projection Technology.

Abstract: In this research, a novel system that uses machine learning-recognized hand gestures to ease file transfer and computer control between two nearby PCs is displayed. By the use of gestures like "catch" and "throw," users may transfer files between computers and handle fundamental PC functions like media control, file browsing, and cursor movement. Bluetooth PAN (Personal Area Network) is used to establish communication between devices, allowing easy data sharing without the need for cable connections or difficult file browsing. The system uses a trained machine learning model for classification, OpenCV for image processing, and a standard webcam for gesture detection. Within an 8–10 m range, experimental results showcase reliable file transfer capability and accurate gesture detection. The goal of this initiative is to enhance computer-human interaction.

Keywords: Machine Learning, Gesture Recognition, File Transfer, Bluetooth PAN, Human-Computer Interaction, Computer Vision.

Abstract: This research, entitled "ENHANCING GRADUATE SCHOOL THESIS ACCESSIBILITY THROUGH DIGITALIZATION: A CASE STUDY OF WUP LIBRARY SYSTEMS," aimed to develop a system that could help graduate school students to access the graduate school theses easily and in a user-friendly manner. The System Development Life Cycle (SDLC) Model (Royce, 1970) was employed in this study. The System Development Life Cycle (SDLC) is an organized method for designing, developing, testing, and implementing information systems. The researchers adopt the following five stages of the SDLC model: First, planning; second, analysis; third, design; fourth, development; and fifth, testing. The study found that beginning with a clear objective to enhance academic research access, the planning and analysis phases ensured that the system would meet user needs while integrating seamlessly with existing platforms. The design phase established a structured framework for content organization, secure access, and ongoing maintenance. Utilizing a no-code approach with Google Sites and Google Drive, the development phase produced a fully functional and easily manageable digital repository. Finally, testing results confirmed the system's stability, efficiency, and usability, affirming its effectiveness in supporting academic research and digital access for graduate students.

Keywords: Case Study, Digitalization, Graduate School Theses, SDLC Model, WUP Library System.

Abstract: This research paper examines the features and user experience of RentLONA, an online platform that connects tenants and landlords. We analyze the benefits and challenges of using RentLONA and evaluate its potential for growth in the rental market. The study aims to provide insights into the platform's impact on the rental industry and identify areas for improvement. RentLONA's user-friendly interface and secure payment processing make it a popular choice among tenants and landlords. However, the platform's limitations and challenges need to be addressed to ensure its continued success. By exploring the features and user experience of RentLONA, this study provides valuable insights for stakeholders in the rental industry. The findings of this study can inform the development of online rental platforms and improve the rental experience for users. RentLONA's success can be attributed to its ability to simplify the rental process and provide a wide range of options for users.

Keywords: Survey Analysis, Real-Time Notifications, MongoDB, Node.js, React.js, Express.js.

Abstract: Flying drones in tight spaces is tricky because they need to avoid obstacles, but the usual tech for this is often too bulky, pricey, or power-hungry for small drones. Enter the SPARC module—a smart, efficient solution designed for these smaller drones, like those using a Raspberry Pi. SPARC uses four tiny LiDAR sensors to create a 360-degree view of the surroundings. The magic happens on a custom 4-layer PCB with a powerful STM32H7 microcontroller. This chip runs a fast algorithm to clean up sensor data and quickly identify the nearest obstacle, all in under 50 milliseconds. This means the main computer on the drone doesn’t have to work as hard. Important safety info is then sent to the drone’s flight controller using a reliable communication protocol. The whole module is just 50mm by 50mm, making it a great fit for small drones by keeping size, weight, and power use low. This makes advanced obstacle avoidance affordable and reliable for the small drone market.

Keywords: Obstacle Perception, LiDAR Sensors, Microcontroller, Collision Avoidance, Drone CAN Protocol.

Abstract: The advancement of mobile technology has opened new opportunities for addressing social challenges. One such challenge is the unequal distribution of essential resources such as food, clothes, and books. The proposed system, Helping Hands, is a cross-platform Android application that facilitates the donation and distribution of these resources to the underprivileged. This digital platform connects donors, NGOs, and volunteers in a transparent and organized manner. Developed using the Flutter SDK and integrated with Firebase services, the system offers real-time updates, secure authentication, and efficient data management. The application simplifies the donation process, minimizes waste, and promotes social responsibility through technology. Testing results confirm the platform's reliability, scalability, and usability, making it an effective solution for sustainable community support.

Keywords: Donation Management, Flutter, Firebase, Mobile Application, NGO Collaboration, Sustainable Development.

Abstract: The stock market is a complex, dynamic system influenced by economic indicators, political factors, and investor sentiment. Predicting its behavior has been a long-standing challenge due to its nonlinear and volatile nature. Recent advancements in Machine Learning (ML) and Deep Learning (DL) have enabled researchers to develop more robust and adaptive models for stock market prediction. This paper provides a comprehensive review of various ML and DL techniques, their comparative performance, challenges, and future directions in financial forecasting.

Keywords: Stock Market Prediction, Machine Learning, Deep Learning, LSTM, ARIMA, Sentiment Analysis

Abstract: Social media has become an essential part of modern communication, allowing people to connect, share, and express themselves. However, the growing presence of fake accounts on platforms like Instagram has become a serious issue, leading to the spread of misinformation, scams, and privacy risks. This project aims to detect fake Instagram profiles using machine learning techniques. Three algorithms—Support Vector Machine (SVM), Random Forest, and Decision Tree—are used to classify accounts as real or fake. The dataset includes user activity details, engagement patterns, and content-based features. The models are trained and compared based on their accuracy and efficiency. The results show that machine learning methods can effectively identify fake profiles and improve safety and trust on social media. Future work can involve integrating deep learning models and extending the system to other social platforms.

Keywords: Fake Accounts, Machine Learning, Instagram, SVM, Random Forest, Decision Tree, Social Media Security, User Behavior.

Abstract: Cyberattacks are among the most significant problems of the modern digital age, posing a threat to people, organizations, governments, and critical infrastructure. From test viruses during the 1970s, cyberattacks have evolved into global operations with crime syndicates and nation-states. Today, advanced campaigns employ artificial intelligence, exploit supply chains, and strike at systemic vulnerabilities, blurring the line between war and crime. This essay charts the past of cyberattacks, discusses common types of attacks, scans the emerging threat of ransomware, analyzes defense systems, and probes new threats facilitated by artificial intelligence, quantum computing, and international interconnections.

Keywords: Cybersecurity, Cyberattacks, Ransomware, Artificial Intelligence, Cyber Defense, Quantum Computing, Cyber Warfare, Network Security

Abstract: Excessive moisture, stemming from heavy rainfall or unintentional overwatering, is a primary and often overlooked contributor to significant plant distress, leading to common ailments such as root rot, fungal infections, and mildew. Conventional methods for managing this problem rely on manual intervention, such as moving pots or applying tarps, which are labor-intensive, unreliable, and impractical for users who are away. This project introduces the “Smart Plant Cover System,” an autonomous, intelligent, and low-cost mechatronic solution designed to proactively protect plants from excess water. The system is architecturally designed around five core conceptual units: an Environmental Data Acquisition Unit (Capacitive Soil Moisture Sensor) to gather real-time data, a Decision Intelligence Unit (Microcontroller) to process this data, a Control Execution Unit (Servo Motor) to provide physical action, a Visualization & Feedback Unit (LCD Display) for status monitoring and calibration, and an Adaptive Response Unit (Hysteresis Logic) to ensure system stability. This report details the systematic process of the system's operation. The Decision Intelligence Unit continuously compares real-time moisture data against a user-defined, pre-calibrated "wet threshold." If the soil moisture level exceeds this critical point, the Control Execution Unit is automatically triggered, deploying a physical cover to shield the plant from further water saturation. Conversely, when the soil moisture level returns to an acceptable, pre-defined "dry threshold," the cover is automatically retracted, allowing the plant to resume normal respiration and photosynthesis. This paper presents the complete design, the critical methodology for sensor calibration, the modular architecture, and the successful implementation and testing of this responsive, standalone solution for intelligent plant protection.

Keywords: Plant cover, excessive moisture, soil moisture sensor, Microcontroller, Hysteresis Logic.

Abstract: The project “Crime Suspection” is designed to enhance crime detection by analyzing human behavior using advanced technologies such as computer vision and machine learning. It aims to identify suspicious activities in public places through continuous video surveillance. The system automatically detects unusual behavior patterns and notifies authorities in real-time. This intelligent approach to surveillance can significantly reduce manual monitoring, improve reaction time, and help in preventing crimes before they occur.

Keywords: Crime analysis, Predictive policing, Fraud detection, Cybersecurity.

Abstract: Over the past decade, Ghana has undertaken substantial investments in digitalization as a strategic initiative aimed at enhancing public financial management, reducing payroll leakages, and increasing fiscal accountability. Initiatives such as Ghana.gov, Smart Workplace, and the Human Resource Management Information System (HRMIS) have been implemented to automate payments, minimize cash handling, and link payroll data with verified personnel records. Utilizing publicly accessible fiscal data from 2019 to 2025, alongside government records and secondary analyses conducted by the World Bank and IMF, this study investigates whether these e-government reforms have effectively reinforced payroll discipline and expanded fiscal space. The findings indicate that although digital transactions via Ghana.gov increased from GH¢ 1.5 billion in 2019 to over GH¢ 126.5 billion in 2025, and the number of participating institutions expanded from fewer than 100 to over 1,750, the wage bill still rose from GH¢ 52 billion in 2019 to GH¢ 89.6 billion in 2025. This suggests that while digitalization has enhanced fiscal transparency, it has not necessarily curtailed fiscal expenditures. Although digital tools have improved revenue monitoring and limited opportunities for leakage, their influence on total payroll costs remains constrained by structural and political challenges. Evidence from Rwanda, Kenya, and Mauritius indicates that digitalization alone is insufficient; sustainable fiscal benefits require institutional integration, performance-based remuneration systems, and robust enforcement mechanisms. The paper concludes with policy recommendations for Ghana’s subsequent phase of digital budget management.

Keywords: Ghana, digitalization, payroll reform, fiscal accountability, Ghana.gov, Smart Workplace, e-governance, wage bill management, public financial management.

Abstract: The easy access to generative adversarial networks (GANs) has resulted in the creation of very realistic deepfake videos. This poses a serious threat to the accuracy of information and public trust. Detecting these altered videos is crucial because traditional methods cannot identify the subtle changes made by deep learning models. This work presents a new hybrid model for detecting deepfake videos. Our approach employs a ResNext convolutional neural network (CNN) to extract important spatial features from individual video frames, particularly focusing on small mismatched areas on faces. These features are then analyzed by a Long Short-Term Memory (LSTM) recurrent neural network (RNN) to track how these features change over time and identify issues between frames that are common in GAN-generated fakes. The model is trained and tested on a large dataset of real and fake videos. We demonstrate how effective our spatiotemporal analysis is, and we also introduce a web-based platform for practical use. Future work will include adding audio and visual analysis to check all types of media.

Keywords: Deepfake, Generative Adversarial Networks, Video Forensics, Convolutional Neural Networks, Long Short-Term Memory, Spatiotemporal Analysis.

Abstract: The practice of hiding communication by enclosing data in other data is known as steganography. There are many different carrier file kinds available, but due to their popularity on the Internet, digital photographs are the most popular. From ancient times to the present, the protection of secret information has always been a major concern. The basic goal of steganography is to hide the existence of the message so that an attacker cannot detect it. To incorporate hidden information, any type of cover item, such as text, image, or video, can be used. In this paper, a brief overview of steganography which is one of the main branches of information hiding is explained and covers its primary forms, categorization, and uses.

Abstract: Due to the COVID-19 outbreak, greater and more dependable tools were required to predict the severity of the disease and help specialists in their decision-making. Conventional approaches might be ineffective and time-consuming in the event of handling large and complex patient data. This paper presents a computer-aided diagnostic model, which is based on machine learning and the application of deep learning to predict the correct outcome. The model uses UNET to process medical images to segment them, and CNN/ResNet50 to classify the chest X-ray or CT scan. In order to enhance accuracy, a hybrid approach is formed by using optimized features with a Random Forest classifier. The system is programmed and developed to have a simple Graphical User Interface (GUI) that allows an individual to upload medical images and get an automated prediction with visualized output. Accuracy, precision, recall, and F1-score are used to compare performance, and the findings indicate that the proposed model is better than the current methods that can be used to offer an effective and convenient means of detecting and predicting the severity of COVID-19 at an early stage.

Keywords: COVID-19 Prediction; Machine learning; Deep learning; UNET; CNN; ResNet50; Random Forest; Image Segmentation; Medical Diagnosis; Computer-Aided Detection.

Abstract: Healthcare is rapidly moving toward patient-centred care, and technology has been essential in developing the quality of personalised healthcare. This is a report about an artefact, which is a healthcare chatbot that utilises state-of-the-art natural language processing technology. Chatbot was developed with an aim of enhancing patient participation, provisioning relevant messages in time, as well building linkage between doctors and patients. It leverages highly advanced NLP systems that can understand the sentiments of conversations and respond back as a real human would, making communication easy and relevant within its context. It can be integrated in existing healthcare platforms and hence made accessible by patients for queries such as health information to specific medical inquiries. Through artificial intelligence or machine learning algorithms, it is possible for the chatbot to keep improving on its performance adjusting according to the ever-changing language and user’s preferences. The healthcare chatbot's key features include symptom analysis, prescription reminders, and health-related FAQs. By adhering to healthcare standards and using encryption mechanisms for sensitive information, the system assures data security and privacy compliance.

Keywords: Natural Language Processing, Artificial Intelligence, Chatbot, Multilingual Functionalities.

Abstract: Digital transformation has profoundly reshaped the fashion industry by blending computational methods with traditional design practice. In particular, Artificial Intelligence (AI) and related technologies (computer vision, deep learning, AR/VR, 3D modeling and generative methods) have enabled virtual try-on systems that promise improvements in personalization, reduced returns, and more sustainable production workflows. This review synthesizes existing literature on virtual try-on systems and their enabling technologies, presents a systematic methodology used to collect and analyze literature, extracts six core themes emerging from research (Design, Consumers, Body, Virtual, Printing, Supply), discusses the primary findings and technological trends, identifies technical and ethical challenges, and outlines key directions for future research. The paper keeps the focus on core developments and critical evaluations drawn from peer-reviewed studies, positioning virtual try-on as a central applied research area at the intersection of AI, graphics, and fashion retail.

Keywords: Artificial Intelligence, Virtual Try-On, Digital Fashion, Augmented Reality, Machine Learning, Fashion Retail, 3D Modeling, Sustainability.

Abstract: Remote Patient Monitoring (RPM) is an emerging healthcare approach that enables continuous tracking of patient health data outside hospital environments. The integration of Augmented Reality (AR) and Virtual Reality (VR) enhances this system by providing immersive visualization, interactive therapy, and real-time doctor–patient communication. This paper presents the design, methodology, and simulation of an AR/VR-based RPM system that uses wearable sensors to collect physiological data, visualize them through AR overlays, and offer virtual rehabilitation environments. The proposed approach demonstrates the potential of immersive technologies in improving accessibility, patient engagement, and quality of remote healthcare.

Keywords: Remote Patient Monitoring, Augmented Reality, Virtual Reality, IoT, Digital Healthcare, Rehabilitation.

Abstract: In the era of rapidly expanding student populations and increasing academic work- loads, traditional methods of evaluating handwritten answer sheets have become inef- ficient, inconsistent, and resource-intensive. AutoGrad addresses these challenges by leveraging cutting-edge Generative AI to automate the assessment process with high ac- curacy and scalability. The system integrates the Gemini model for Optical Character Recognition (OCR), effectively digitizing diverse handwriting styles, and the LLaMA-7B language model for semantic answer evaluation.AutoGrad introduces a novel Mixture-of-Experts (MoE) architecture to significantly reduce character recognition errors and uses adaptive thresholding to fine-tune evaluation rigor based on question types. The solution further integrates a rule-based and AI-driven hybrid evaluation engine, ensuring both factual correctness and semantic coherence in student answers. With a Flask-based user interface, vector similarity matching, and a real-time feedback generation system, AutoGrad offers an end-to-end, scalable solution for academic institutions.Empirical results from real-world deployments show a reduction in grading time and a correlation with faculty evaluations. AutoGrad not only automates evaluation but en- hances it—providing detailed feedback, promoting personalized learning, and supporting academic integrity at scale.

Keywords: Automated Grading, Generative AI(or LLM), Optical Character Recognition(OCR), Semantic Evaluation,Scalable Assessment, Mixture-of-Experts(MoE), Distributed Processing

Abstract: The integration of Artificial Intelligence (AI) and Augmented Reality (AR) in higher education represents a groundbreaking innovation in modern learning environments. This project focuses on designing and developing an intelligent and immersive learning platform that leverages these technologies to enhance student engagement, understanding, and retention. AR technology allows students to visualize abstract concepts in an interactive 3D environment, bridging the gap between theoretical knowledge and real-world application. Meanwhile, AI enables personalization by analyzing learner data to recommend suitable content, monitor progress, and provide adaptive feedback. Together, these technologies make education more accessible, interactive, and effective. The proposed system incorporates AI-driven analytics, AR visualization modules, and a web-based dashboard for instructors to manage content and track student performance. This combination not only enriches the learning experience but also promotes active learning and critical thinking. The outcome of this project is a scalable and intelligent educational platform that empowers institutions to modernize teaching methods and redefine higher education in the era of digital transformation.

Abstract: Deepfake leverage machine learning algorithms, particularly Generative Adversarial Networks (GANs), to create highly realistic images, videos, or audio recordings of individuals. By learning from vast datasets, these models can generate media that mimics real-life behavior, expressions, and voices, making them difficult to identify as fake. The primary objective of this project is to develop a Deepfake Detection System that can effectively identify and classify manipulated media using advanced deep learning and computer vision techniques. This system aims to address the security, ethical, and social implications posed by synthetic media by providing a reliable tool for the detection and analysis of deepfakes in both static images and video sequences.

Keywords: Deepfake, Fake Image Detection, Fake Video Detection, Image Dataset, Video Dataset.

Abstract: This paper presents a real-time framework for American Sign Language (ASL) recognition and translation that leverages a Convolutional Neural Network (CNN)-based deep learning approach to achieve robust, accurate, and efficient gesture interpretation. The proposed system utilizes a vision-driven interface to capture hand gestures via webcam, applying image preprocessing and CNN-based feature extraction to recognize static and dynamic ASL signs. The model is trained and validated on a custom dataset comprising ASL alphabets, numerals, and commonly used words, employing data augmentation and cross-validation to enhance resilience against variations in lighting, background, and signer morphology. Recognition results are mapped to corresponding textual output, with optional speech synthesis for improved accessibility. Experimental evaluations demonstrate average recognition accuracies exceeding 90% under real-world conditions, outperforming traditional methods in both speed and reliability. This practical framework bridges communication gaps for the deaf and hard-of-hearing community, providing an accessible solution for human-computer interaction and assistive technology.

Keywords: American Sign Language, Human-Computer Interface, Convolutional Neural Network

Abstract: Industrial machines experience gradual degradation due to continuous operation, mechanical wear, and varying load conditions. Unplanned breakdowns result in production losses, increased maintenance costs, and reduced equipment lifespan. To address these challenges, this work presents an IoT-based predictive maintenance system that continuously monitors machine health parameters and performs real-time analysis using the ThingSpeak cloud platform. An ESP32 microcontroller is integrated with sensors such as an ADXL345 three-axis accelerometer for vibration measurement, a temperature sensor, and a current sensor to capture critical machine health indicators. The acquired data is transmitted to ThingSpeak through Wi-Fi, where MATLAB Analytics is used to extract features such as vibration RMS, spectral peak frequencies, temperature trends, and load variations. These features are further analyzed to detect anomalies, estimate machine degradation, and predict possible failure conditions. Threshold-based logic and machine learning algorithms are implemented on the cloud to classify machine states into healthy, warning, and fault categories. The system also triggers alerts using ThingHTTP and webhooks, enabling immediate maintenance actions. Experimental results show that the proposed solution provides accurate early-warning detection, reduces downtime, and offers a scalable, low-cost architecture suitable for industrial automation environments. The research demonstrates that IoT-enabled predictive maintenance significantly improves reliability, enhances operational efficiency, and supports data-driven decision-making in industrial machine monitoring.

Keywords: Predictive Maintenance, Industrial Machines, IoT-Based Monitoring, ThingSpeak Cloud, ESP32 Microcontroller, ADXL345 Accelerometer, Vibration Analysis, MATLAB Analytics;

Abstract: Sign language is a vital medium of communication for individuals with hearing and speech impairments, but the lack of knowledge among non-signers creates barriers. This project proposes a real-time sign language recognition system that can detect alphabets (A-Z) and numerics (0-9) from webcam video. The system combines modern deep learning techniques with web technologies to provide an accurate, fast, and user-friendly solution. The frontend uses WebRTC to capture video streams directly in a browser, making the system platform-independent and usable with any standard laptop or external camera. The backend uses FastAPI with WebSockets to enable real-time communication between the browser and the deep learning model, ensuring low-latency predictions. The recognition model integrates EfficientNet (transfer learning) for feature extraction, combined with a CNN+RNN to capture spatial and temporal patterns. An attention mechanism enhances performance by focusing on the most informative frames, while a GRU classifier predicts the final alphabet or number with high accuracy. Training and validation are carried out using benchmark datasets along with self-collected samples to ensure adaptability in real-world conditions. The system prototype displays recognized signs as text beneath the video feed, with emphasis on accuracy, robustness, and real-time performance for applications in education, healthcare, and accessibility services.

Keywords: Sign Language Recognition, Real-Time Gesture Recognition, EfficientNet, CNN-GRU Hybrid Model, Attention Mechanism, Spatial-Temporal Feature Extraction, WebRTC, Low-Latency Inference

Abstract: Correct identification of spam messages is a critical component of modern digital communication. With the rapid growth of mobile devices and instant messaging platforms, users are increasingly exposed to unsolicited messages that often contain phishing links, fraudulent schemes, or promotional content. These spam messages not only pose security threats but also result in time loss, decreased productivity, and potential financial damage. Therefore, developing automated systems capable of accurately detecting and filtering spam messages is essential for safeguarding users and maintaining the integrity of communication networks.

Abstract: Mental health problems are climbing fast, and for a lot of people, real therapy just isn’t an option. Money, a lack of therapists, and the weight of stigma keep too many folks from getting help. Lately, though, AI has started to change the equation.New conversational agents can bring support to more people, right where they are. In this study, we built and tested an AI mental health chatbot that blends natural language processing with cognitive behavioral therapy ideas. It recognizes emotion, offers structured support, and keeps privacy and ethics front and center. Early results show that these chatbots make it easier for people to get help, but there’s still work to do—especially around empathy, personalization, and making sure the advice is truly reliable.

Keywords: Conversational AI, Mental Health Support, NLP Chatbot, Artificial Intelligence, Natural Language Processing, Social and Behavioral Sciences, Real-time Emotion Detection

Abstract: This project presents an intelligent assistive system for stroke patients using a smart glove integrated with three flex sensors, heart rate and temperature sensors, a camera, and a Raspberry Pi 4B. The flex sensors are attached to three fingers to detect bending motions, representing binary combinations to trigger predefined commands. These commands, alongside vital signs such as temperature and heart rate, are displayed on an LCD screen and transmitted to an IoT platform (ThingSpeak) for remote monitoring. A camera module captures live video of the patient, which is streamed in real-time. Additionally, a MATLAB based GUI application is developed to display all sensor data and commands on a computer, providing real time monitoring and support. This system offers an efficient, low-cost solution to enhance communication and health tracking for stroke patients.

Keywords: Smart Glove, Stroke Rehabilitation, IoT Healthcare Monitoring, Real time Patient Assistance Security.

Abstract: The global adoption of electric vehicles (EVs) is steadily rising because they offer an environmentally sustainable alternative with minimal carbon emissions. However, challenges related to energy storage and efficient charging remain major obstacles to wider EV adoption. This paper provides solutions to charging systems with hybrid sources, plug-in hybrid electric vehicles (PHEVs), and all-electric vehicles (EVs). This project explores how Artificial Intelligence (Al) and the Internet of Things (IoT) can enhance EV performance monitoring and support autonomous vehicle operation through sensor integration. A self-charging system can be implemented and the exchange of information between the vehicle and its surroundings. Artificial Intelligence (AI) refers to the human mind that can perform tasks and decision-making like human intelligence through different logic and programs. Al technologies play a crucial role in advancing electric vehicles toward full automation. In the future, Al and loT enabled autonomous vehicles could help minimize charging delays, improve parking management, and support the development of smart city infrastructure.

Keywords: Electric vehicles (EVs), Internet of things (IoT), Artificial Intelligence (AI), Plug-in Hybrid Electric Vehicles (PHEVs), Wireless Charging, Battery Monitoring System.

Abstract: This work presents a low-cost wearable sensor system for real-time kinematic analysis of cricket bowling. The system integrates MPU6050 IMU sensors placed on the wrist, elbow, and spine, along with a MAX30100 heart-rate sensor and DHT11 temperature–humidity sensor. Data is captured through an ESP8266 NodeMCU and transmitted to a cloud platform for processing. Machine learning algorithms classify bowling phases and detect technique deviations, while a synchronized 30-FPS camera provides visual verification. A web dashboard displays real-time biomechanics, physiological status, and automated performance reports. The system provides an affordable alternative to high-end motion-capture solutions, supporting injury prevention and performance improvement.

Keywords: Data Visualization, Kinematic Analysis, Machine Learning, Wearable Inertial Sensors, Cloud Storage.

Abstract: Agriculture is a vital sector that supports the livelihood of millions of people, particularly in developing countries like India, where farming remains a primary source of income. However, the sector faces persistent challenges such as improper crop selection, soil nutrient imbalances, climate variability, water scarcity, and lack of scientific decision-making. Traditional farming practices rely heavily on farmers’ intuition, experience, or generalized recommendations, which often result in poor crop yield, excessive fertilizer usage, and increased vulnerability to environmental fluctuations. To address these issues and support precision agriculture, this research proposes a machine-learning-based Crop Recommendation System that utilizes soil nutrient values—Nitrogen (N), Phosphorus (P), and Potassium (K)—along with environmental factors such as temperature, humidity, pH level, and rainfall to recommend the most suitable crop for cultivation in a given region. The proposed system uses supervised machine learning algorithms, including Random Forest, Decision Tree, Naive Bayes, and Support Vector Machine (SVM), to analyze large agricultural datasets and learn complex patterns between soil–climate features and crop suitability. The dataset undergoes extensive preprocessing, which includes handling missing values, normalizing numeric attributes, removing noise and outliers, and encoding categorical labels. Feature engineering further enhances the prediction quality by identifying the most influential variables such as the N:P:K ratio, soil fertility index, and climate–soil interactions. These features help the model better distinguish crop requirements under varying environmental conditions. During the experimental phase, each algorithm was trained and evaluated using standard performance metrics, including accuracy, precision, recall, and F1-score. Results show that the Random Forest classifier outperformed other models, achieving an accuracy of 97%, largely due to its ensemble nature, robustness, and ability to handle high-dimensional data. The findings highlight that machine learning can significantly improve agricultural decision-making by offering farmers scientific guidance tailored to their land conditions. This system has the potential to enhance crop productivity, minimize risks associated with crop failure, optimize fertilizer usage, and promote long-term soil health. This research demonstrates that integrating machine learning into agriculture provides a practical and scalable solution to modern farming challenges. Future advancements may include IoT-enabled soil sensors, satellite-based remote sensing, real-time data analysis, and deep learning models for yield prediction and dynamic crop recommendation. Overall, the study emphasizes the transformative potential of machine learning in supporting sustainable and smart agriculture. This research aims to develop and evaluate a machine-learning-based crop recommendation system using key soil and climatic features. By comparing multiple ML algorithms and identifying the most accurate model, the study contributes to the growing field of smart agriculture and demonstrates how technology can transform traditional farming practices.

Abstract: Effective and smart way to automate the operation of the parking system that allocates an effective parking space using internet of effects technology. The IoT provides a wireless access to the system and the individuals can keep a track of the vacuity of the parking area. With increase in the population of the vehicles in metropolitan metropolises, road traffic is the major problem that's being faced. The end of this paper is to resolve this issue. The individual generally wastes his time and sweats in hunt of the vacuity of the free space in a specified parking area. The information about the empty places is displayed to individual. Therefore, the waiting time for the individual in hunt of parking space is minimised. To automate the operation of the parking system using IoT technology, the system can be equipped with detectors that can descry the presence of a vehicle in a parking spot. These detectors can be placed in each parking space and can communicate wirelessly with the system’s central server. The data from these detectors can be used to produce a real- time chart of the parking area, showing the vacuity of parking spots to the individuals.

Keywords: Internet of Things (IoT), Smart Parking System, Arduino UNO, IR Sensor, Servo Motor, LCD Display, I2C Module, Embedded C, Real-Time, Traffic Congestion

Abstract: Liver diseases constitute a major global health challenge, responsible for millions of deaths each year and placing a substantial burden on healthcare systems. The liver is a vital organ responsible for metabolic regulation, detoxification, and biochemical synthesis. Any disruption in its functioning can lead to severe disorders such as Hepatitis, Cirrhosis, Liver Cancer, Non-Alcoholic Fatty Liver Disease (NAFLD), and Alcoholic Liver Disease. Early detection of these conditions is crucial because most liver disorders progress silently, showing minimal or non-specific symptoms during their initial stages. Traditional diagnostic methods, including blood tests, imaging scans, and biopsies, are often invasive, costly, time-consuming, and may not always provide clear or timely results. These limitations highlight the need for accurate, efficient, and automated tools that can support clinical decision-making.

Abstract: Agriculture plays a vital role in sustaining the global economy, yet farmers face challenges such as labour shortages, high costs, and inefficient resource utilization. To address these issues, this project presents the design and development of a Smart Multipurpose Agriculture Robot capable of performing multiple farming operations autonomously or semi-autonomously. The proposed system integrates IoT (Internet of Things), embedded systems, and sensor technologies to carry out tasks such as seed sowing, pesticide spraying, soil moisture monitoring, and weed detection efficiently.

Keywords: Arduino Uno, IoT (Internet of Things).

Abstract: The Eco-Bin is a semi-automated smart waste management prototype designed to demonstrate efficient segregation of dry, wet, and e-waste using sensors, motorized mechanisms, and programmed decision-making. Instead of performing real composting, the system simulates the decomposition process to help users understand modern waste-handling methods. The model promotes hygienic disposal, reduces manual sorting, and supports environmental awareness by showcasing how automation can improve sustainability. Eco-Bin serves as an educational tool, offering a simple, cost-effective approach to smart city–oriented waste management.

Keywords: Smart bin, waste segregation, automation, IoT-based prototype, decomposition simulation, sustainable waste management.

Abstract: Driver drowsiness is a major cause of road accidents, resulting in severe injuries and fatalities. This project proposes a Smart Driver Drowsiness Detection System using Raspberry Pi and advanced Object Detection (ADAS & CAOA) algorithms to monitor driver alertness in real time. The system employs Eye Aspect Ratio (EAR) and Mouth Aspect Ratio (MAR) to analyse eye closure, yawning, and head movements through continuous camera monitoring. When fatigue is detected, it triggers alerts, activates hazard lights, or sends SOS messages. Using Virtual Network Computing (VNC) for remote monitoring, the system enhances road safety with an intelligent, scalable, and cost-effective solution.

Keywords: Eye Aspect Ratio (EAR), Virtual Network Computing (VNC), Mouth Aspect Ratio (MAR), Object Detection (ADAS & CAOA).

Abstract: This paper presents an experimental study focused on optimizing Portable Document Format (PDF) file size reduction through the sequential application of multiple online compression tools to achieve user-specified file size targets. Existing online compressors employ distinct algorithms for image re-encoding, font subsetting, and metadata removal, often yielding varied results in terms of compression efficiency and quality preservation, but not to the user-specified size. The proposed approach investigates the cumulative effect of applying multiple compression tools in a defined sequence to achieve enhanced file size reduction without significant quality degradation. A heterogeneous dataset of PDF files, commonly required for uploading as Proof of Address (POA) and Proof of Identity (POI) on Indian government websites, such as Aadhaar Card, PAN Card, and property documents, comprising text-heavy and image-rich content, was compressed using popular online tools, including SmallPDF, iLovePDF, PDFCompressor, and Adobe Acrobat. The performance of each tool and its sequential combinations was evaluated using metrics such as compression ratio, percentage reduction, and visual fidelity indices. Experimental results demonstrate that optimized multi-tool sequences achieved up to 27% greater reduction in file size compared to single-tool compression, while maintaining acceptable readability and structural integrity. The findings highlight the potential of hybrid compression pipelines for efficient ways to compress the files according to the needed file size to upload.

Keywords: PDF compression, multi-tool optimization, hybrid compression pipeline, online compression tools, document size reduction, PSNR, SSIM, digital archiving.

Abstract: The implementation and design of a hybrid detection framework driven by AI that can analyse objects and faces in real time using a single web-based system is introduced in this paper. The proposed architecture uses a dual-model pipeline that combines the speedy object detection capabilities of You Only Look Once, version 7 with the facial recognition accuracy of DeepFace. A full-stack Flask application serves as the foundation for secure user interaction, camera management, and live data visualization. The framework facilitates multi-camera connectivity, dynamic object selection, and instant user enrollment through interactive face capture. Effective object monitoring and human identification in a range of environmental conditions are ensured by the detection and recognition modules operating simultaneously on live video streams. All user credentials and logs are securely maintained using encrypted authentication techniques and Structured Query Language Lite. Additionally, by allowing real-time updates of facial datasets without server outages, the system improves scalability and flexibility. The experimental evaluation demonstrates that the hybrid model consistently provides high recognition accuracy while maintaining low processing delays, making it suitable for real-time applications such as automated attendance, intelligent monitoring, and security-driven surveillance. Its combined structure allows the system to handle both object detection and identity recognition within a single workflow, avoiding the limitations of using separate models. By merging these capabilities, the framework delivers a balanced solution that improves reliability, strengthens security, and ensures smooth real-time operation. Overall, the developed system creates an integrated platform that aligns efficiency, adaptability, and practical usability for diverse AI-based environments.

Keywords: Hybrid Detection Framework, YOLOv7, DeepFace, Real-Time Recognition, AI-Driven Face Analysis

Abstract: Exam malpractice significantly undermines the reliability of digital and remote examination systems. Traditional manual invigilation lacks scalability and accuracy, leading to inconsistencies in supervision. This project proposes an AI-driven solution integrating Real-time Monitoring, Object Detection, and Human Monitoring for automated malpractice detection. The system utilizes computer vision and deep learning algorithms to analyze live or recorded video streams, identifying anomalies such as multiple human presences, unauthorized devices, and irregular motion patterns. Through continuous behavioural tracking and object classification, the framework ensures high- precision detection of suspicious activities, thereby enhancing academic integrity, examination security, and operational efficiency in online and offline assessment environments.

Keywords: Real-time Monitoring, Object Detection, Human Monitoring, Academic integrity.

Abstract: The rise of obesity and metabolic disorders highlights the limitations of generalized dietary advice, which often fails to meet individual health needs. At the same time, health tracking applications and self-reported data provide valuable insights into individuals’ heart rate, physical activity, sleep patterns, and dietary intake. This research presents a machine learning–based personalized nutrition recommendation system that integrates biometric, lifestyle, and dietary data to provide customized diet plans for individuals. By analyzing patterns in user behavior and correlating them with health risk factors, the system predicts potential nutritional deficiencies or risks and generates actionable recommendations tailored to each user. The approach leverages data-driven modeling to bridge the gap between raw health data and effective, personalized dietary guidance. Experimental evaluation on synthetic and public datasets demonstrates that the system can accurately identify individual health risks and suggest targeted nutritional adjustments, promoting preventive healthcare and overall well-being.

Keywords: Personalized Nutrition, Machine Learning in Healthcare, Dietary Recommendation Systems, Preventive Healthcare, Biometric Data Analytics, Health Risk Prediction, Individualized Diet Planning.

Abstract: Identification of people in any organization or colleges for the purpose of attendance marking is one such a software of face recognition. The use of Attendance Management System is to performs the regular activities of attendance marking and analysis with reduced human intervention. In this method the camera is settled and it will capture the image, the faces are recognized along with the database and eternally the attendance is marked. This system is depended on face detection and recognition concept, that detects the employees or student using webcam when they arrive in the office or class room and marks the attendance by recognition.

Keywords: Face Recognition, Attendance Management System, Haar Cascade.

Abstract: Phishing has become one of the most pervasive and damaging forms of cybercrime, targeting unsuspecting internet users through fraudulent websites that mimic legitimate ones. These malicious platforms deceive users into revealing sensitive credentials such as passwords, financial information, and personal identification details. The continuous evolution of phishing tactics—such as sophisticated URL obfuscation, dynamic content manipulation, and social engineering—renders traditional detection mechanisms increasingly ineffective. Conventional defense strategies, including blacklists, heuristic filters, and rule-based approaches, fail to detect newly emerging or “zero-day” phishing websites that are not yet cataloged in known databases. Hence, there is an urgent need for an adaptive, intelligent, and automated solution that can accurately detect phishing websites in real time without dependence on third-party services. This study focuses on developing a machine learning-based phishing website detection system that leverages URL-based, domain-based, and HTML content-based features to distinguish between legitimate and phishing websites. The core idea is to train classification models capable of learning behavioral patterns and structural differences inherent in phishing websites. The dataset used for experimentation consists of over 60,000 URLs, equally divided into phishing and legitimate samples, collected from verified and publicly available repositories. Feature extraction plays a pivotal role in this system. Important features include URL length, use of special symbols, number of subdomains, domain registration age, SSL certificate presence, hyperlink patterns, and term frequency–inverse document frequency (TF-IDF) vectors derived from the website’s HTML content.

Abstract: Automatic Question Generation (AQG) systems aim to convert textual material into meaningful assessment questions, offering a valuable tool for both educators and learners. Unlike conventional text summarization, AQG involves identifying essential information, generating accurate answers, and creating plausible distractors to form high-quality multiple-choice questions. This paper presents an NLP-based AQG framework that processes instructional content through a sequence of linguistic operations, including text pre-processing, syntactic and semantic analysis, and probabilistic language modeling. The proposed system utilizes established NLP libraries such as SpaCy and NLTK to detect key concepts and automatically construct factual questions related to entities, events, and contextual details. By automating question creation, the approach reduces the manual workload of educators and provides learners with an efficient tool for self-assessment. The study also highlights architectural considerations, discusses implementation challenges, and suggests future improvements to enhance the scalability and accuracy of AQG systems.

Keywords: Automatic Question Generation (AQG), Natural Language Processing (NLP), Educational Technology, Semantic Analysis, Question Formulation and Generation

Abstract: The proposed research presents the design and implementation of an intelligent fuzzy system that integrates disease prediction with personalized drug dosage control. The framework utilizes patient-specific data, including clinical, demographic, and physiological parameters, to predict disease probabilities and recommend safe drug doses through fuzzy inference mechanisms. By addressing uncertainties and vagueness in medical data, the system improves diagnostic reliability and ensures therapeutic accuracy. Comparative results show that the fuzzy-based approach achieves superior performance over conventional machine learning models such as ANN and SVM, obtaining 92% accuracy with a significantly lower RMSE of 0.19. The proposed system demonstrates strong potential for clinical decision support by enhancing interpretability, reliability, and precision in diagnosis and dosage recommendation.

Keywords: Fuzzy logic, disease prediction, drug dosage control, medical decision support, fuzzy inference system, machine learning, intelligent healthcare, uncertainty modeling.

Abstract: The proposed research presents the design and implementation of an intelligent fuzzy system that integrates disease prediction with personalized drug dosage control. The framework utilizes patient-specific data, including clinical, demographic, and physiological parameters, to predict disease probabilities and recommend safe drug doses through fuzzy inference mechanisms. By addressing uncertainties and vagueness in medical data, the system improves diagnostic reliability and ensures therapeutic accuracy. Comparative results show that the fuzzy-based approach achieves superior performance over conventional machine learning models such as ANN and SVM, obtaining 92% accuracy with a significantly lower RMSE of 0.19. The proposed system demonstrates strong potential for clinical decision support by enhancing interpretability, reliability, and precision in diagnosis and dosage recommendation.

Keywords: Fuzzy logic, disease prediction, drug dosage control, medical decision support, fuzzy inference system, machine learning, intelligent healthcare, uncertainty modeling.

Abstract: Android is the world’s most widely used mobile operating system, powering billions of smartphones, tablets, smart TVs, and IoT devices. Its open-source nature supports innovation but also increases exposure to many security threats. Because Android applications handle highly sensitive data such as banking information, identity details, authentication tokens, and personal records, a single vulnerability can lead to privacy leaks, financial loss, unauthorized access, or malware attacks. Recent studies (2020–2025) highlight recurring issues such as insecure data storage, weak cryptographic implementation, misuse of runtime permissions, unsafe Inter-Component Communication (ICC), insecure network communication, and application tampering or repackaging. The fast rise of Android malware—often using code obfuscation, dynamic payloads, and repackaging—adds further complexity to application security. To address these issues, researchers have proposed modern mitigation techniques, including encrypted storage, certificate pinning, component protection, secure coding practices, and automated testing tools like MobSF, QARK, and Drozer. Industry standards such as OWASP MASVS and MASTG provide structured guidelines for secure development. Recent work also shows that AI and ML models (SVM, LSTM, CNN) achieve high accuracy in detecting malware. Overall, the literature concludes that most Android vulnerabilities result from improper implementation rather than platform limitations, stressing the need for a security-first development approach

Keywords: Android Security, Malware Analysis, ICC, Mobile Application Vulnerabilities, Cryptography, Secure Development.

Abstract: This project focuses on creating a wireless communication system for natural disaster alerts using IoT and GSM technology. The system integrates sensors that monitor ground vibration, water levels, rainfall, and temperature. Data from these sensors is processed by an Arduino controller and transmitted through an Android application for instant alerts. It ensures timely warnings to people in remote areas, enhancing safety during disasters. Powered by solar energy, it functions even during power failures. The system improves communication reliability, enabling quick emergency responses and reducing damage caused by floods, landslides, or earthquakes in disaster-prone regions.

Keywords: GSM technology, Arduino, solar energy, disaster-prone regions, landslides.

Abstract: This project aims to develop an integrated intelligent safety system for railways that can detect fire hazards and prevent train collisions using modern technologies like LabVIEW and IoT. The system uses temperature and smoke sensors connected to LabVIEW for real-time fire detection and alert generation. Simultaneously, an IoT-based collision avoidance module employs ultrasonic and GPS sensors to detect nearby trains or obstacles and transmit warnings to a central control unit. By combining both systems, the project enhances railway safety, automation, and accident prevention, reducing human error and ensuring-quicker-responses-to emergencies.

Keywords: Collision avoidance, Internet of things (IOT), Automation, Real-Time Monitoring, Temperature Sensor, GPS Module, Ultrasonic Sensor, Smoke Sensor, Alert System, Smart railway, Accident Prevention, LABVIEW

Abstract: Skin diseases are among the most common medical conditions worldwide, affecting millions of people each year. Accurate and timely diagnosis is critical, yet traditional diagnostic methods depend heavily on dermatologists’ expertise and manual examination, which can lead to human error and delayed treatment. This paper presents an automated skin disease detection system using Convolutional Neural Networks (CNNs), a deep learning technique capable of learning complex visual features from medical images. The proposed model classifies skin lesions into different disease categories, such as melanoma, eczema, and psoriasis, using publicly available datasets like HAM10000. The CNN model is trained and validated on dermoscopic images, achieving high accuracy in disease identification.

Keywords: Skin Disease Detection, Deep Learning, Convolutional Neural Network, Image Classification, Medical Diagnosis.

Abstract: Ransomware has evolved from simple file-encryption malware into a sophisticated criminal enterprise, increasingly intertwined with Bitcoin and other cryptocurrencies. While early attacks focused on encrypting data and demanding ransom, modern ransomware often combines extortion with direct cryptocurrency theft, exploiting vulnerabilities in wallets, exchanges, or decentralized finance (DeFi) protocols [1], [2]. Bitcoin’s pseudonymity, global reach, and liquidity make it both a preferred ransom payment medium and a direct target for attackers, who use complex laundering techniques such as mixers, cross-chain swaps, and dark-net marketplaces to obscure funds [3], [4]. Despite improvements in blockchain forensics and law enforcement interventions, attackers continuously adapt, blending ransomware and crypto-heist strategies to maximize profit while complicating attribution [5]. This study surveys the evolution of ransomware, examines the convergence with Bitcoin-based theft, and highlights detection, prevention, and forensic strategies that integrate endpoint monitoring, blockchain intelligence, and cross-jurisdictional coordination to disrupt these hybrid attacks effectively.

Keywords: Ransomware; Bitcoin; Cryptocurrency Heists; Blockchain Forensics; Ransomware-as-a-Service (RaaS); Cybercrime Economy; Money Laundering; DeFi Exploits; Cybersecurity Defense

Abstract: The Metaverse, a dynamic amalgamation of virtual reality, augmented reality, blockchain, and artificial intelligence, represents a paradigm shift in the digital landscape. This abstract encapsulates an exploration into the diverse dimensions of the Metaverse, beginning with its inception and tracing its evolution through history. Delving into the realms of technological convergence, societal impact, and economic dynamism, we navigate the Metaverse's influence on culture, ethics, and user experiences. The narrative encompasses the challenges and considerations inherent in this digital frontier, from legal and ethical frameworks to ensuring accessibility and inclusivity. The Metaverse emerges not merely as a technological spectacle but as a transformative force shaping how we connect, create, and envision the digital future. This abstract offers a comprehensive overview, capturing the essence of the Metaverse's evolution, impact, and the multifaceted considerations that accompany this groundbreaking digital era

Keywords: Metaverse, Virtual Reality, Augmented Reality, Artificial Intelligence.

Abstract: Stock price prediction is one of the most challenging tasks in financial analysis due to the market's highly volatile and nonlinear nature. In this study, we propose a machine learning-based approach for predicting stock prices using historical data. The model utilizes regression-based and deep learning algorithms to capture temporal patterns and market trends. Specifically, we employ Linear Regression, Random Forest, and a Long Short-Term Memory (LSTM) network to model the time-series behavior of stock prices. The LSTM model is particularly suited for this task as its architecture allows it to effectively learn and remember long-term dependencies inherent in sequential financial data. The project aims to assist investors and analysts in making informed decisions by forecasting future stock prices with reasonable accuracy. The performance of all models is rigorously evaluated using standard metrics, including Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and R²-score, on a held-out test set. The results demonstrate that the Long Short-Term Memory (LSTM) network significantly outperforms traditional regression models in capturing sequential dependencies in stock market data, achieving the highest R² score of 0.93 and the lowest RMSE of 0.018. This robust performance underscores the suitability of deep learning for complex financial forecasting.

Keywords: Stock prediction, Sequential Financial Data, Temporal patterns, Forecasting, Deep learning, LSTM.

Abstract: This study applies data analytics and machine learning techniques to analyze coffee shop sales and operational efficiency in India. The research focuses on identifying patterns in customer behavior, sales trends, and cost optimization using data collected from point-of-sale (POS) systems, inventory records, and IoT-based sensors. A predictive model is developed to forecast daily sales and recommend inventory levels based on factors such as time, weather, and customer footfall. Data preprocessing, feature extraction, and regression-based algorithms are used to evaluate relationships between sales, pricing, and operational factors. The study demonstrates how data-driven insights can improve decision-making, reduce wastage, and enhance profitability for coffee shops. The results highlight the potential of integrating computer science tools—such as machine learning, data visualization, and IoT monitoring—into the coffee retail industry for smarter management and sustainable growth.

Keywords: coffee shop, data analytics, machine learning, IoT, sales forecasting, operational efficiency

Abstract: In modern manufacturing environments, ensuring consistent product quality is critical for maintaining competitiveness and customer satisfaction. This study proposes an online machine learning-based approach for real-time industrial product quality analysis. Unlike traditional offline models that require retraining on static datasets, the online learning paradigm enables the system to continuously update itself with incoming data, adapting to process variations and new patterns without significant downtime. The proposed system integrates streaming data from production lines—such as sensor readings, visual inspections, and process parameters—to predict and detect quality anomalies early in the production cycle. Experimental results on industrial datasets demonstrate that the online learning models, including algorithms like Online Gradient Descent and Adaptive Random Forest, achieve high accuracy and robustness while reducing latency in decision-making. This approach enhances operational efficiency, minimizes defective output, and supports predictive maintenance strategies in Industry 4.0 settings.With the growing adoption of Industry 4.0 technologies, real-time quality monitoring has become essential in modern manufacturing systems. This study presents an online machine learning- based approach for industrial product quality analysis, aimed at improving defect detection and process optimization. Unlike traditional batch learning models, online machine learning algorithms continuously update their parameters with streaming data, allowing for adaptive learning in dynamic production environments. The system leverages real-time data from sensors and inspection tools to identify deviations in product quality as they occur. Algorithms such as Online Gradient Descent and Adaptive Random Forest were evaluated for their performance in handling non-stationary data streams. Experimental results show that the proposed method provides high accuracy, low latency, and efficient resource usage, making it suitable for deployment in smart manufacturing systems. This research highlights the potential of online learning to enhance product reliability, reduce waste, and support intelligent decision- making in industrial processes.

Keywords: Online Machine Learning Industrial Quality Control, Real-time Monitoring, Smart Manufacturing, Adaptive Algorithms, Product Defect Detection, Industry 4.0, Data Streams, Predictive Maintenance, Machine Learning in Manufacturing.

Abstract: After 2021, over 90 million passenger automobiles were produced, marking a significant increase in auto production. This growth has led to a flourishing used car market, which has become a highly lucrative sector. One of the most critical and fascinating areas of research within this market is automobile price prediction. Accurate price prediction models can greatly benefit buyers, sellers, and businesses in the used car industry. This paper presents a detailed comparative analysis of two supervised machine learning models: K-Nearest Neighbour and Support Vector Machine regression techniques, to predict used car prices. We utilized a comprehensive dataset of used cars sourced from the Kaggle website for training and testing our models. The K Nearest Neighbour algorithm is known for its simplicity and effectiveness in regression tasks. On the other hand, the Support Vector Machine regression technique uses a different approach, finding the optimal hyperplane that best fits the data. Both methods have their strengths and weaknesses, which we explored in this study. Our results indicated that both KNN and SVM models performed well in predicting used car prices, but with slight variations in accuracy. Consequently, the suggested models fit as the optimum models and have an accuracy of about 83 percent for KNN and 80 percent for SVM. The results indicate that the KNN model slightly outperforms the SVM model in predicting used car prices

Keywords: K Nearest Neighbour, Machine Learning, Prediction, Support Vector Machine, Used Cars Accuracy.

Abstract: AI-Based Financial Law Analyser and Collaborator, an intelligent digital platform designed for ease in understanding and management of financial legal matters, uses advanced techniques in Artificial Intelligence like NLP, OCR, and Legal-NER. The system thus processes complex legal documents, identifies relevant financial laws, and provides a crisp summary, including drafts for potential defences. It enhances accessibility by having multilingual support. Moreover, it includes the lawyer discovery module that links users with qualified legal professionals based on the nature of the case, area of expertise, and applicable jurisdiction. Built using Python, Flask, LangChain, and React, the platform ensures both processing efficiency and a seamless, interactive user experience. Automating document interpretation, performing legal research, and lawyer matchmaking not only greatly cut costs but also democratize access to justice. This is a flagship project on how Artificial Intelligence can change financial legal assistance, making legal insight more understandable, actionable, and affordable for all.

Keywords: Legal tech, AI, NLP, legal informatics, semantic search, chatbots Introduction

Abstract: Heart disease remains one of the leading causes of global mortality, often due to late diagnosis and the absence of early risk assessment. Traditional diagnostic methods require clinical visits, medical equipment, and expert interpretation, which may not always be accessible. This paper presents a Machine Learning (ML)–based heart disease prediction system designed to evaluate an individual’s likelihood of developing heart disease using key medical parameters. The model uses attributes such as age, cholesterol, blood pressure, fasting blood sugar, maximum heart rate, chest pain type, and other clinical indicators to generate accurate predictions. Several ML algorithms—Logistic Regression, Random Forest, KNN, and Support Vector Machine—were trained and evaluated, with Random Forest achieving the highest accuracy. A web-based interface built using Streamlit allows users to enter their health metrics and receive prediction results instantly, along with personalized health recommendations. This system is scalable, user-friendly, and promotes early prediction, ultimately supporting preventive healthcare.

Keywords: Heart Disease Prediction, Machine Learning, Random Forest, Health Monitoring System, Medical Diagnosis, Streamlit.

Abstract: Bitcoin is known for its high volatility and speculative trading behavior. Predicting Bitcoin prices is valuable for investors, traders, and financial analysts. The study uses historical price data, technical indicators, and/or sentiment analysis. Machine learning and statistical models like ARIMA, Linear Regression, and LSTM are applied. Deep learning models, especially LSTM, show better accuracy in capturing time-series patterns

Keywords: Prediction accuracy, Time series analysis, Historical data, Model training and testing.

Abstract: The proposed project, Smart Parenting: IoT Solutions for Infant Safety, aims to ensure the continuous protection, comfort, and well-being of infants through an intelligent monitoring system built around the Raspberry Pi controller. The setup combines a range of sensors and control units that work together to observe the baby’s environment and physical condition in real time. When the system detects irregular behavior such as unusual body posture or distress it automatically sends alerts to parents via a notification platform. In addition, it can manage environmental factors such as temperature and lighting, and an automated mosquito-net mechanism helps shield the baby from insects. Overall, the system provides a reliable, data-driven approach to infant care and comfort.

Keywords: Raspberry Pi, Internet of Things, Infant Safety, Intelligent Monitoring, Automation.

Abstract: This paper presents a broad overview of inventory management within the pharmaceutical industry. Effective inventory management helps organizations determine what products to order, when to order them, and in what quantities. It involves tracking stock from procurement to final sale, monitoring demand patterns, and adjusting accordingly to ensure adequate availability. While stockouts typically lead to lost sales and reduced profits for most businesses, in the pharmaceutical sector they can result in prolonged illness, fatalities, or even malpractice lawsuits—situations hospital administrators strive to avoid. Therefore, precise oversight is essential to manage stock levels, monitor expiration dates, and maintain proper storage conditions. Tools such as barcode scanning at receipt, lot tracking for monitoring product movement within the facility, and accurate sales recording make this possible. Regardless of where a transaction takes place, updates to physical inventory must be reflected across the entire system. Despite numerous theoretical inventory models proposed over the years, many have not been implemented by pharmaceutical organizations due to limited real-world testing. This creates a need to examine the practical relevance and impact of Inventory Theory by validating a simple inventory model in a high-demand hospital setting in India. This paper aims to evaluate the theory, explore its global applications in healthcare institutions, and highlight its advantages.

Keywords: Inventory Control, Pharmaceutical Sector, Supply Chain Optimization, Product Oversight, Capacity Planning, Mixed-Integer Linear Programming Model, RFID Technology, Optimization Techniques, Resource Utilization Efficiency, Strategic Planning Framework

Abstract: Multiple researchers have developed fuzzy logic-based active and reactive controllers in multi-agent systems to help agents act intelligently. Some researchers have been adding the PageRank Algorithm into the fuzzy logic controller in robot soccer during the evolutionary process. It is known that the PageRank Algorithm is focused on the selection of a website to search by means of hyperlinks; popular websites are commonly selected by more users because they usually contain good content. Taking the PageRank Algorithm idea, we will develop a new agent selection method in multi-agent systems. Multi-agent system definition MAS can be defined as a society of agents that interact with one another according to certain rules in order to accomplish some sets of goals autonomously. In a multi-agent system, four types of agent interactions can be established: Interaction, Coordination, Collaboration, and Coalition. These types of interactions are explained in the work of Wooldridge (2002). According to Omicini et al., the cooperation and coordination capabilities inside the multi-agent systems have been improving through the use of new technologies, like the Java language. For instance, the Java Agent Development framework can be used to develop a multi-agent system from Java beans. Key Words: Fuzzy logic, Page Rank, Multi-Agent Systems, JADE, Agent Communication, Agent Selection

Abstract: This project presents the design and development of a Job Portal Application that serves as a digital platform connecting job seekers and employers. The main objective of the system is to simplify the recruitment process by providing an efficient and user-friendly interface for both parties. Job seekers can register, create profiles, upload resumes, and apply for suitable jobs, while employers can post vacancies, search for candidates, and manage applications. The application is developed using modern technologies such as React Native for the front end and Firebase/MySQL for backend data management. Additional features like real-time notifications, job recommendations, and secure authentication enhance usability and system reliability. This project demonstrates how technology can be used to automate and streamline the hiring process, reducing manual work and improving communication between job seekers and employers. Overall, the Job Portal App contributes to a faster, more organized, and transparent employment process.

Keywords: Java Base, Android Operating System, XML Server.

Abstract: This study investigates the application of machine learning algorithms to predict cancer risk levels based on a dataset of various risk factors. Four classification models, namely Support Vector Machines (SVM), Logistic Regression, Random Forest, and XGBoost, were trained and evaluated on a dataset containing patient information and associated risk factors. The data was preprocessed to handle categorical features and scale numerical features before splitting into training and testing sets. The models were trained on the training data and their performance was assessed using accuracy on the test set. Logistic Regression achieved the highest accuracy of 0.9000, followed by SVM (0.8800), XGBoost (0.8775), and Random Forest (0.8575). The results demonstrate the potential of machine learning models, particularly Logistic Regression, in predicting cancer risk levels based on the provided factors. This can aid in identifying individuals at higher risk and potentially facilitate early intervention strategies.

Keywords: Random Forest, XGBoost, Healthcare Analytics, Risk Factors, Data Science.

Abstract: Keeping road in some good condition poses one of the most difficult and costly tasks, particularly with traditional methods that depend on manual surveys. These processes are time consuming, prone to error, and lead to late discovery of certain cracks and potholes that can drive up repair costs and threaten safety. To address this issue, our product is a mettalic roads damage automated system with machine learning. The system takes video pictures of roads, analyzes the data to find flaws, and categorizes issues according to how serious they are. It was created using MATLAB's machine learning capabilities and offers local governments immediate insights via an intuitive interface. By employing a prioritization framework based on the age of the roads, it guarantees that aging infrastructure is addressed quickly, improving resource distribution and repair timelines. This method seeks to reduce expenses, increase accuracy, and expedite road assessments. In addition to ensuring road safety and comfort while driving, this project supports sustainability in the management of municipal infrastructures by creating smarter, safer, and better-maintained roads.

Keywords: Road Damage Detection; Metallic road analysis; machine lesrning; image processing; road safety; aging infrastructure.

Abstract: With the rapid advancement of Artificial Intelligence, deepfake audio generation has become increasingly realistic and difficult to identify. These synthetic voices can be misused for fraud, impersonation, political manipulation, and privacy violations. Traditional audio verification systems based on manual inspection or basic acoustic features are not sufficient to detect these sophisticated manipulations. This research introduces a Machine Learning-based Audio Deepfake Detection System that analyzes speech signals to distinguish between real and synthetic audio. The proposed model uses a CNN + LSTM hybrid architecture, trained on Mel-spectrogram representations of audio clips. The system achieves high accuracy, effectively detecting voice cloning across different speakers and environments. Developed in Python using Librosa, TensorFlow/Keras, and Sklearn, the system processes uploaded audio files and provides a prediction label (“Real Audio” or “Fake Audio”). Experimental results show strong performance, minimal false detections, and suitability for security, forensics, and media authentication tasks.

Keywords: Deepfake Audio, Voice Cloning, CNN-LSTM, Machine Learning, Speech Analysis, Fake Audio Detection, Mel-Spectrogram.

Abstract: The rapid expansion of digital ecosystems has increased the need for intelligent, unified platforms that streamline career development, recruitment processes, and professional networking. Career Connect Hub is a comprehensive MERN-based web application designed to bridge the gap between students, job seekers, mentors, and recruiters through an integrated and data-driven ecosystem. The platform provides secure user authentication, personalized dashboards, AI-enhanced resume analysis, job-role matching, and a centralized training and placement management system. It also enables senior–junior mentorship, professional networking, real-time job postings, application tracking, and an AI-powered help desk for career-related queries. Machine learning modules, deployed as microservices, support resume scoring, fake job detection, and personalized career recommendations. By combining modern web technologies with intelligent decision-support systems, Career Connect Hub offers a scalable and usercentric solution that enhances employability, improves recruitment efficiency, and supports data-driven career planning.

Keywords: Career Development Platform; MERN Stack; AI Resume Analyzer; Job Recommendation System; Fake Job Detection; Professional Networking; Placement Management; Microservices; Machine Learning; StudentRecruiter Interaction.

Abstract: Accurate prediction of residential electricity demand is essential for energy conservation, cost optimization, and effective grid planning. Smart homes generate large volumes of fine-grained consumption data, making them suitable candidates for advanced predictive modeling . This study proposes a hybrid forecasting framework that integrates Long Short-Term Memory (LSTM) networks with Extreme Gradient Boosting (XG Boost). LSTM captures temporal dependencies in consumption sequences, while XG Boost model nonlinear relationships in engineered features. The ensemble produces stable and adaptive predictions suitable for dynamic household environments. A web-based interface supports data upload, real-time forecasting, visualization, and cost estimation. Experimental results demonstrate that the hybrid model consistently outperforms standalone approaches in RMSE, MAE, and MAPE. The system provides interpretable predictions using feature-attribution techniques, enabling users to understand consumption drivers. This research contributes a practical and extensible solution for smart home energy management.

Keywords: Smart Home Energy Forecasting, LSTM, XG Boost, Hybrid Ensemble Model, Deep Learning, Gradient Boosting, Smart Grid Optimization, Demand Response, Feature Engineering.

Abstract: Soil fertility plays a crucial role in agricultural productivity, yet farmers often struggle to identify nutrient deficiencies and select suitable fertilizers. Traditional methods are time-consuming, costly, and lack personalized recommendations. To address this, we propose Soil IQ, an Explainable AI (XAI)–based system that predicts soil nutrient levels (N, P, K, pH, organic carbon) and recommends optimal fertilizers with transparent model explanations. The system uses machine learning algorithms such as Random Forest and Decision Trees, combined with SHAP-based explainability, to generate interpretable recommendations. Experimental results demonstrate high accuracy in nutrient prediction and improved decision-making for fertilizer selection. Soil IQ empowers farmers with data-driven insights, enhances crop productivity, and promotes sustainable fertilizer usage.

Keywords: Soil Analysis, Fertilizer Recommendation, Explainable AI, Machine Learning, Agriculture, SHAP.

Abstract: Accurately estimating student academic performance is central to educational planning and early intervention strategies. Performance outcomes are influenced by multiple academic, behavioral, and lifestyle factors, making predictive modeling an important tool for educators and institutions. This study proposes a machine learning framework integrated with explainable artificial intelligence (XAI) to predict students' performance scores using features such as study duration, previous academic results, extracurricular participation, sleep patterns, and practice of sample question papers. The dataset was preprocessed through imputation of missing values and exploratory visualization techniques, including histograms, kernel density plots and correlation heatmaps, to assess distributions and identify anomalies. A diverse set of regression models, including Linear Regression, Bayesian Ridge, Ridge, Lasso, Elastic Net, Decision Tree Regressor, Random Forest, XGBoost, and LightGBM, was evaluated using MSE, MAE, RMSE, R², PSNR, and SNR. Linear Regression emerged as the best-performing method, achieving an MSE of 4.06, MAE of 1.60, RMSE of 2.01, R² of 0.98, and the highest PSNR and SNR values. To improve interpretability, SHAP and LIME techniques were applied to identify both global and local feature influences. The findings demonstrate that interpretable models supported by XAI can provide accurate predictions while enhancing transparency, thereby offering meaningful insights for educational research and policy formulation.

Keywords: Machine Learning, XAI, SHAP, LIME.

Abstract: The early detection of brain tumors plays a vital role in improving patient survival rates and treatment planning. This project presents a deep learning-based system for Brain Tumor Detection and Classification using Convolutional Neural Networks (CNN) and Vision Transformers (ViT). The system analyzes MRI images to identify and classify tumors automatically. The CNN model effectively extracts local spatial features, while the ViT captures global contextual information, resulting in improved classification accuracy. The proposed approach was trained and tested on MRI datasets, achieving high accuracy and reliability. A Gradio web interface was also developed to provide an interactive platform for real-time image upload and tumor prediction. The experimental results demonstrate that the ViT model outperforms CNN in accuracy and robustness, confirming the potential of transformer-based architectures in medical image diagnosis. This project contributes to the development of an efficient, accurate, and user-friendly system for assisting radiologists in brain tumor detection.

Keywords: Brain Tumor Detection, Convolutional Neural Networks (CNN), Vision Transformer (ViT), Medical Image Analysis, Deep Learning, MRI Classification, Hybrid Architecture, Feature Extraction, Computer-Aided Diagnosis, Transfer Learning, Tumor Classification, Medical Imaging, Neural Networks, Artificial Intelligence in Healthcare, Diagnostic Support System

Abstract: Worldwide, road traffic crashes continue to claim a large number of lives and cause massive economic damage. A major factor that determines survival chances is how quickly emergency services are informed after an incident. Over the past few decades, accident detection techniques have evolved from purely manual reporting to advanced automated solutions that rely on IoT devices, various sensors, computer vision, and artificial intelligence. This survey reviews the historical development, current approaches, and future trends in accident detection and notification systems, including human-dependent methods, sensor-based systems, smartphone applications, Intelligent Transportation Systems (ITS), and AI-enhanced frameworks. A comparative study reveals that modern solutions significantly outperform traditional ones by reducing response time and increasing detection accuracy. The paper concludes with remaining challenges and promising research directions for next-generation systems.

Keywords: Accelerometer, Detection, Emergency Response, IoT, Raspberry Pi, Sensors, Smart Mobility

Abstract: Our current supply chains struggle with hidden inefficiencies and vulnerabilities due to limited transparency. This opacity fuels the rampant issue of counterfeiting, often undetectable by sight. Existing methods like RFID tags and AI offer some solutions, but they're hampered by limitations like replicability and high computational demands. This project proposes a novel approach: harnessing the power of blockchain technology. By creating a decentralized, tamper-proof ledger that tracks every step of a product's journey, this system guarantees transparency and traceability throughout the supply chain. Blockchain based system, makes everything decentralized that may be accessed by several parties at the same time. One of its main advantages is that the recorded data is difficult to change without the consent of all parties concerned which makes the data extremely secure and protect from all vulnerabilities. This paper presents system designed using blockchain technology for detection of counterfeit products.

Keywords: “Blockchain”, “Counterfeit, RFID”, “Artificial Intelligence”, “QR code”.

Abstract: Diabetic Retinopathy (DR) is a progressive ocular disorder resulting from prolonged diabetes, in which damage to the retinal blood vessels can eventually cause permanent vision impairment. Early diagnosis is crucial, yet access to routine screening remains inadequate, especially in rural and resource-constrained communities. RetinoAI is an automated deep-learning framework designed to detect DR and classify its severity using retinal fundus photographs. The system uses Convolutional Neural Networks (CNNs) enhanced through transfer learning, accompanied by an optimized preprocessing pipeline incorporating image resizing, normalization, and augmentation to improve visual quality and model robustness. The DR classification module identifies disease grades ranging from No DR to Proliferative DR and generates corresponding confidence levels. To improve interpretability, RetinoAI employs Grad- CAM-based visual explanations that highlight important retinal regions contributing to the model’s decision, supporting clinicians with clearer insight into AI-assisted diagnoses.

Keywords: Deep Learning, Diabetic Retinopathy, Convolutional Neural Networks, Fundus Image Analysis, Grad- CAM, Medical Imaging.

Abstract: The evolution of digital gaming platforms has introduced significantly different interaction experiences, particularly between traditional desktop gaming and emerging Virtual Reality (VR) environments. This study compares the visual immersion, user interaction, performance demands, and cognitive engagement associated with VR gaming and desktop gaming. Additionally, the growing role of Artificial Intelligence (AI) in enhancing gameplay, adaptive difficulty, realism, and personalized user experiences is examined across both platforms. Findings indicate that VR offers higher sensory immersion and presence due to its 3D spatial environment and natural motion controls, while desktop gaming provides greater accessibility, comfort, and precision in competitive gameplay. AI contributes to both platforms by improving game physics, character behaviors, and user-tailored responses.

Keywords: Virtual reality, VR game, 3D game, Oculus Rift, Unity.

Abstract: The Eco-Smart Waste-to-Electricity System is an innovative sustainable-energy solution that integrates waste management, renewable electricity generation, sensor-based automation, and GSM-enabled remote monitoring. The core objective of the project is to convert different types of waste—such as plastics, paper, and organic materials—into usable electrical energy while ensuring operational safety and real-time supervision. In this system, waste materials are collected and fed into a Zaar Box (Heating/Burning Chamber). When the waste begins to burn, the heat generated is transferred to a heating panel, which converts the thermal energy into electrical energy. This generated electricity is stored in a 12V, 1.2Ah lead-acid battery. The stored energy is later used to power various loads such as LED bulbs or small DC devices. A heating sensor/temperature sensor plays a crucial role—during the electricity generation and charging phase, the output supply remains OFF for safety. Once charging reaches the required level and the temperature stabilizes, the sensor triggers the system to turn ON the output power supply, allowing the bulb to glow and demonstrating live energy generation from waste.

Keywords: Arduino uno, GSM (Globel System for Mobile Communication), Heating panel, Temperature Sensor, Voltage Sensor.

Abstract: Clean water is a basic need for all living beings. Without water survival in the Earth is not possible. Water covers about 70% of the Earth’s surface among that only 3% of that is pure water. Water gets polluted due to any reasons like industry waste, sewage waste, garbage waste. Hence it is important to maintain cleanliness and hygiene of water. We considered this water pollution as a serious issue and start to work on the project. We decided to incorporate technology to get the work done effectively and efficiently. Our project design is in such a way that it collects the waste which floats on water bodies. In present time almost all the people are familiar with robots. We are going to design a very interesting robot that is RF controlled Robot. It is important to monitor the pH of a water body. The Wireless Aquatic Waste Management Boat is an innovative solution designed to address water pollution and monitor aquatic ecosystems. This autonomous boat integrates cutting-edge technologies such as IoT, advanced sensors, and wireless communication. It efficiently removes floating waste while continuously monitoring critical water quality parameters such as pH, temperature, turbidity, and dissolved oxygen. Equipped with real-time data transmission capabilities, the boat provides stakeholders with actionable insights through cloud-based platforms and user-friendly interfaces, promotes sustainability and reduces human intervention. This project aims to conserve aquatic ecosystems.

Keywords: Sewage, garbage, Remote Controlled Robot, Radio Frequency, potential hydrogen, Wireless, aquatic waste management, autonomous boat, pH detection, environmental monitoring, IoT, real-time data, renewable energy, water pollution, sustainability, ecosystem conservation.

Abstract: The BioEx project, short for Biodiversity Explorer, is a web-based application designed to promote interactive learning and exploration of biological species. It integrates Next.js, React.js, and Tailwind CSS to deliver a fast, responsive, and user-friendly interface. The platform provides a range of educational features such as the “Species of the Day”, Feature Highlights, and an interactive database of organisms that helps users explore biodiversity in an engaging manner. BioEx aims to bridge the gap between traditional biology education and modern digital learning by combining technology with environmental awareness.

Keywords: Interactive learning, Next.js, React.js, Tailwind CSS, Biology education

Abstract: In an increasingly globalized economy, real-time access to accurate currency exchange information has become essential for international trade, travel, and digital financial transactions. This research presents the design and development of a Real-Time Currency Converter capable of delivering instant and reliable exchange rate calculations using live data APIs. The system integrates automated data fetching, server-side validation, and a user-friendly interface to ensure both accuracy and convenience. Modern technologies such as RESTful APIs, cloud-based services, and mobile application frameworks are employed to enhance system performance and scalability. Experimental results show that the proposed application achieves high efficiency, low latency, and consistent accuracy under various operational conditions. This study contributes to financial technology solutions by providing a robust model for real-time currency conversion that can be expanded to include predictive analytics and AI-driven financial insights in the future.

Keywords: Real-time exchange rates, currency converter app, financial technology, API integration, mobile application, data validation, cloud computing, fintech, live currency data.

Abstract: Blood group detection using fingerprint patterns is an emerging biometric approach that aims to identify a person’s blood group without invasive procedures. Conventional methods require blood sampling, laboratory equipment, and skilled technicians, which may be time-consuming and uncomfortable. This proposed method focuses on analyzing fingerprint ridge characteristics and applying machine learning or pattern-matching techniques to establish a correlation between fingerprint patterns and specific blood groups. By extracting features such as loops, whorls, and arches, and mapping them to biological datasets, blood groups can be detected efficiently. This approach has the potential to provide fast, cost-effective, portable, and non-invasive blood group identification. It can be highly useful in medical emergencies, blood banks, forensic science, and remote healthcare systems. The technology offers scope for automation and can significantly enhance medical record management and identity verification. With further research and large dataset analysis, this method can become a reliable alternative to traditional blood group testing.

Keywords: Blood Group Detection, Fingerprint Recognition, Biometrics, Machine Learning, Non-Invasive Testing, Pattern Analysis, Medical Identification, Ridge Characteristics, Healthcare Technology, Forensics.

Abstract: AI-Powered Voice and Chatbot Ordering System presents an AI-driven grocery ordering system that enables users to place orders through voice commands and chatbot-based interactions. The system integrates Natural Language Processing (NLP), speech recognition, and a web-based interface to simplify product search, order placement, and user communication. A Flask backend manages authentication, product information, inventory updates, and order processing, while SQLite is used as the database layer. The platform reduces manual effort, supports hands-free operation, and offers a more accessible alternative to traditional grocery ordering applications. The results demonstrate improved usability and efficient processing of voice and text-based queries.

Keywords: Voice Ordering, Chatbot Application, NLP, Speech Recognition, Flask Framework, AI-Based Ordering System

Abstract: Effective communication between the deaf community and the hearing population remains a significant challenge due to the limited prevalence of sign language proficiency. This project introduces a multi-sensor fusion-based gesture recognition system aimed at enhancing interaction for deaf individuals. The system integrates data from various sensors, such as), flex sensors, and electromyography (EMG) sensors, to accurately capture and interpret hand gestures. By employing advanced machine learning algorithms, including deep learning models, the system translates complex sign language gestures into textual or auditory outputs in real- time. This approach not only improves recognition accuracy but also ensures robustness against environmental variations, offering a reliable solution for seamless communication. The proposed system holds promise for applications in assistive technologies, facilitating better integration of deaf individuals into diverse social and professional settings.

Keywords: Multi-sensor fusion, gesture recognition, sign language interpretation, deaf communication.

Abstract: This research article analyzes the current state of agile software development methodologies and how they are being adopted across various software industries. As companies persist in adapting to digital transformation, agile methodologies have progressed past their conventional structures to include new technologies, mixed models, and expanded implementations. This research examines prevailing trends such as the incorporation of artificial intelligence into agile methodologies, the merging of DevOps and DevSecOps with agile frameworks, and the obstacles organizations encounter in the process of agile transformation. This study identifies crucial success factors, prevalent challenges, and new trends in agile implementation through a thorough review of literature and analysis of industry practices. The results show that although agile adoption is on the rise, organizations are progressively tailoring frameworks to fit particular contexts, resulting in hybrid methodologies. The study emphasizes the increasing significance of metrics-based strategies, ongoing education, and cultural change as essential factors for effective agile implementation. This thorough examination offers important perspectives for software development companies aiming to enhance their agile methodologies in a changing technological environment.

Keywords: Agile Software Development, DevOps, Scrum, Kanban, Scaled Agile, Software Engineering, Digital Transformation, Continuous Integration

Abstract: Social media addiction has become a growing concern in recent years, especially among teenagers and young adults. This study explores the major causes and negative effects of excessive social media use. The key causes include the need for social approval, fear of missing out (FOMO), instant rewards through likes and comments, and the widespread availability of smartphones and internet access. These psychological and technological factors make users spend more time online than intended. The effects of social media addiction are seen in emotional, academic, and health-related areas. It often leads to anxiety, low self-esteem, sleep disturbances, lack of concentration, and reduced physical activity. In severe cases, it affects relationships and overall mental well-being. The study highlights the need for awareness, digital discipline, and healthy usage habits to reduce the impact of social media addiction. Understanding both the causes and effects is important for developing better prevention and intervention strategies.

Keywords: Social Media Addiction, FOMO, Social Approval, Mental Health, Sleep Disturbance, Academic Performance, Screen Time

Abstract: By creating an AI-powered system that forecasts churn using the Telecom Customer Churn Dataset and increases ROI through customized discounts, this project addresses the problem of customer retention in the telecom sector. To determine eligibility for a discount, a Logistic Regression model examines consumer data. A Python Flask backend is connected to the web interface (HTML, CSS, JavaScript), and Firebase allows real-time updates to an Android app for UPI payments and notifications. The solution increases client loyalty and spurs growth by combining machine learning, analytics, and cloud technology.

Keywords: AI-based churn insights ROI-optimized ML system Intelligent retention framework Predictive churn modelling Profit maximization using AI End-to-end churn prediction pipeline Hybrid ML approach Customer retention optimization.

Abstract: The rapid evolution of autonomous driving technology has sparked significant interest in developing systems that can operate safely and efficiently in dynamic real-world environments. A critical component of any self-driving car system is its ability to perceive and detect objects in its surroundings. Accurate and real-time object detection ensures that the vehicle can make informed decisions to avoid obstacles, follow traffic rules, and interact with other vehicles and pedestrians safely. This paper explores the implementation of real-time object detection for a self-driving car using the Haar Cascade Classifier, a widely recognized machine learning technique for object detection. The Haar Cascade Classifier is based on the concept of using Haar-like features, which are simple rectangular patterns, to detect objects by analyzing the differences in pixel intensity within those regions. These features, when combined in a cascade of increasingly complex classifiers, allow for the rapid detection of objects such as pedestrians, vehicles, traffic signs, and other critical obstacles. In this implementation, we utilize a pre-trained Haar Cascade model to detect objects from the camera feed of a self-driving car, processing the video in real time. To achieve real-time performance, we optimize the algorithm by carefully adjusting parameters such as the detection scale and window size, as well as reducing false positives through post-processing techniques. The system is tested under different environmental conditions, such as varying lighting, weather, and object sizes, to assess the robustness and effectiveness of the approach in practical scenarios. The results demonstrate that while the Haar Cascade Classifier can achieve reasonable accuracy and speed for detecting simple objects in structured environments, challenges remain in more complex and dynamic situations, such as heavy traffic or poorly lit roads.

Keywords: Real-Time Object Detection, Self-Driving Car, Haar Cascade Classifier, Autonomous Vehicles, Machine Learning, Computer Vision, OpenCV, Image Processing, Traffic Sign Detection, Pedestrian Detection, Vehicle Detection, Artificial Intelligence (AI), Feature Extraction

Abstract: Potholes continue to pose serious and recurring challenges to transportation safety, vehicle durability, and overall roadway efficiency. Their presence results in increased accident risk, higher fuel consumption, and substantial damage to vehicle suspension systems. Traditional detection approaches-such as manual inspection, complaint-based reporting, and periodic municipal surveys-fail to provide real-time, scalable, and accurate results, especially in dynamic traffic environments. This research proposes an advanced, AI-driven pothole detection framework that integrates YOLO-based deep learning with multi-sensor fusion to significantly enhance detection reliability. The system utilizes RGB cameras for visual analysis, radar and LIDAR for 3D surface profiling, IMU sensors for vibration-based anomaly confirmation, ultrasonic sensors for depth estimation, and GPS modules for precise geo-tagging. A dedicated sensor-fusion layer ensures robust performance by validating detections across diverse environmental conditions including low-light scenarios, rain, uneven illumination, and partial occlusions. Furthermore, the system incorporates V2V communication to broadcast real-time alerts to nearby vehicles and uploads validated detections to a cloud-based analytical dashboard for predictive maintenance and road-health monitoring. Experimental evaluation across varied terrains demonstrated a detection accuracy above 93%, with significantly reduced false positives compared to camera-only models. The results confirm that the proposed multi-sensor, deep-learning-driven architecture is highly suitable for integration into intelligent transportation systems, enabling safer mobility and smarter roadway infrastructure management.

Keywords: Pothole Detection, Deep Learning, Multi-Sensor Fusion, YOLO Algorithm, LIDAR, Radar Profiling, IMU Sensors, Ultrasonic Depth Measurement, GPS Geotagging, V2V Communication, Intelligent Transportation Systems (ITS), Cloud Analytics, Smart Road Maintenance

Abstract: Virtual Reality (VR) is rapidly redefining creative and technological practices within the Indian film industry. By enabling immersive 360-degree environments and interactive cinematic experiences, VR allows audiences to actively engage with film narratives rather than simply observe them. Indian filmmakers and production houses are gradually adopting VR for various stages of filmmaking, including pre-production visualization, virtual set design, advanced VFX integration, and interactive promotional campaigns. The use of VR significantly enhances audience engagement by offering realistic simulations, first-person perspectives, and highly immersive storytelling formats. In addition, VR is strengthening the growth of India’s animation, gaming, and digital content sectors by creating new employment opportunities for VR designers, VFX artists, and immersive media developers. The Indian film industry—especially Bollywood and southern regional cinema—has begun experimenting with VR experiences such as VR trailers, immersive action sequences, and virtual tours of film sets. These innovations support cost reduction, efficient planning, and improved creative control. VR also presents new opportunities for regional cinema to showcase cultural stories through immersive narratives. Despite these advantages, VR adoption in India faces challenges such as high technological cost, limited access to VR devices among general audiences, a lack of skilled VR professionals, and insufficient VR theatre infrastructure. Nevertheless, with increasing digital transformation, falling hardware costs, and rising interest in immersive entertainment, VR is poised to become a significant catalyst for innovation in Indian cinema. Overall, VR has the potential to revolutionize the filmmaking workflow, audience experience, and future of cinematic storytelling in India.

Keywords: Virtual Reality, Indian Films, Immersive Cinema, 360° Storytelling, Digital Filmmaking

Abstract: This project presents a real-time, AI-powered ransomware defense application designed to safeguard critical data against evolving cyber threats, with targeted deployment for academic institutions, individual users, government organisations, and data centers. The proposed system continuously monitors file and process activity, employing a combination of heuristic rules and advanced AI algorithms to detect and mitigate suspicious behavior. Upon ransomware detection, the application automatically blocks or terminates malicious processes, quarantines affected files, and securely preserves private encryption keys to enable safe data recovery without payment. Additionally, a rollback mechanism facilitates rapid restoration of recent files, while real-time notifications and a user-friendly dashboard support continuous monitoring and management. Lightweight, cost-effective, and cross platform, this solution advances indigenous cybersecurity capabilities in alignment with Atmanirbhar Bharat and smart automation initiatives, promoting resilience and self-reliance in digital security.

Keywords: RansomShield AI, Swadeshi Secure, DataShield India, BharatSafe Defender, CyberRaksha AI.

Abstract: This project presents the development of a multifunctional smart vending machine capable of dispensing beverages and items using voice commands, coin validation and UPI-based digital payments. The system is built around an Arduino Mega controller and integrates an HC-05 Bluetooth module for wireless voice operation, relay-driven motor pumps for dispensing hot milk, cold milk, and decoction, and a DC motor with a spring-based mechanism for item dispensing. A webcam-driven coin detection system enables the machine to accept only ₹10 coins while rejecting all invalid denominations. The vending machine supports dual operating modes: voice-controlled operation via a mobile application and manual operation through four dedicated buttons for hot coffee, cold milk, item dispensing, and Wi-Fi activation. A Wi-Fi dongle provides 330 seconds of internet access per recharge, ensuring reliable connectivity for UPI transaction verification. The machine dispenses beverages or items only after confirming successful payment; otherwise, it returns the coin and aborts the process. A 16×2 LCD display provides real-time updates, including payment status, beverage selection, and Wi-Fi activation timing. Overall, the proposed system delivers a cost-effective, automated, and user-friendly vending solution suitable for cafeterias, hostels, offices, and public environments, offering enhanced convenience through hybrid payment support and smart control functionality.

Keywords: Smart Vending Machine, Voice Control, UPI Payment, Coin Detection, Arduino Mega, HC-06 Bluetooth, Motor Pump, Wi-Fi Automation.

Abstract: Our research presents a cutting-edge AI-driven plat form, Wealth Wizard, designed to empower users with advanced tools for financial management and planning. The system features automated expense categorization, budget forecasting, personalized recommendations, and real-time alerts. Leveraging machine learning models such as Random Forest Classifiers and ARIMA, the platform identifies spending patterns and forecasts future expenses, enhancing financial literacy and decision-making. An interactive chatbot ensures user engagement, addressing queries and guiding platform use. This research highlights the innovative use of AI and ML in revolutionizing personal finance, paving the way for accessible and efficient financial solutions. Index Terms—finance technology, expense categorization system, machine learning, budget prediction.

Keywords: AI, ML, KNN (K-Nearest Neighbours), ARIMA (Auto Regressive Integrated Moving Average), Word Vectorizer (Word2Vec).

Abstract: The rapid expansion of digital banking has increased the number of customers maintaining multiple bank accounts, resulting in inconvenience and security challenges associated with carrying numerous ATM cards and remembering multiple PINs. This research proposes the development of a unified, intelligent, and secure ATM card capable of managing multiple bank accounts through a single authentication framework. Using RFID, fingerprint biometrics, Arduino-based embedded systems, and GSM-enabled OTP verification, the proposed solution enhances both convenience and security. This paper discusses the architecture, methodology, hardware integration, and performance assessment of the system that consolidates multiple bank accounts into one secure card.

Keywords: RFID, Biometrics, ATM Security, Multi-Account ATM Card, Embedded Systems, Arduino, OTP Verification.

Abstract: Federated Learning (FL) has emerged as a powerful paradigm for privacy-preserving medical image analysis, enabling collaborative training across multiple pathology centers without sharing raw patient data. In breast cancer diagnostics, both histopathological image classification and segmentation are essential for identifying malignant regions and supporting early clinical decision-making. However, the development of high-performing deep learning models is challenged by institutional data silos, staining and scanner variability, annotation inconsistencies, and severe non-IID data distributions across clinical sites. This literature review synthesizes recent advances in FL for histopathology - including encrypted aggregation, differential privacy mechanisms, attention-guided architectures, parameter-efficient modality adapters, and segmentation-driven frameworks for whole-slide imaging. Special emphasis is placed on FedImp, an impurity-based optimization method that adaptively weights client updates using entropy-driven data-quality measures. While originally evaluated in classification scenarios, FedImp directly addresses critical limitations affecting segmentation tasks, such as uneven label availability, morphological heterogeneity, and client imbalance. By prioritizing informative updates and suppressing noisy or skewed contributions, FedImp enhances convergence stability, improves generalization, and reduces communication overhead in multi-center FL settings. Through a comparative review of ten influential studies, this work highlights existing methodological gaps and positions FedImp as a compelling foundation for future federated histopathological breast cancer segmentation pipelines, integrating privacy, robustness, and clinical scalability.

Keywords: Federated Learning, Histopathological Image Analysis, Breast Cancer Detection, Important Deep Neural Network Layers, Non-IID Data,FedImp.

Abstract: India’s rural population continues to experience major barriers to quality healthcare, such as shortages of doctors and nurses, poor infrastructure, long travel distances, and limited health awareness. Recent advancements in Artificial Intelligence (AI), Natural Language Processing (NLP), and mobile connectivity have led to the emergence of AI-driven health chatbot as a optimum solution. These chatbots provide basic health support through symptom guidance, health education, appointment scheduling, medication reminders, and mental-health assistance. This paper explores how AI-based health chatbots improve healthcare access in rural India, especially in tribal areas. It reviews current deployments, evaluates their impact on healthcare accessibility and health-seeking behavior, and discusses technological, ethical, and infrastructural challenges. Findings show that when integrated with government health systems and community health workers, AI chatbots can strengthen first-level care, improve health information delivery, and support public health programs. However, issues such as low digital literacy, diverse local languages, poor internet connectivity, and the need for proper clinical supervision limit their effectiveness. The paper concludes with recommendations to scale chatbot-based healthcare solutions across rural India.

Keywords: AI in healthcare, health chatbots, virtual health assistants, rural health, India, telemedicine, digital health services.

Abstract: The recognition and categorization of bone fractures hold great significance in urgent medical care. It determines how practitioners make critical decisions and aids in preventing delays that might jeopardize patient safety. Manually interpreting X-rays, CT scans, and MRIs can be time-consuming, and honestly, human reviewers can overlook crucial details, especially when it comes to tiny or intricate fractures. However, deep learning has revolutionized this landscape. Automated systems now rapidly identify fracture patterns with remarkable precision. In this research, we examine how convolutional neural networks (CNNs), transfer learning, and hybrid deep learning frameworks can elevate our ability to detect fractures. We train and evaluate these models using medical images that we have pre-processed think data augmentation, image enhancement, and feature extraction. This helps the models generalize more effectively and identify fractures that may be less evident. The objective is to classify fractures based on type, severity, and location, enabling physicians to initiate appropriate treatment promptly. Our findings reveal that deep learning models surpass traditional machine learning methods, achieving higher sensitivity and specificity across diverse datasets. AI-driven tools can significantly boost radiologists’ efficiency, alleviate their workload, and facilitate quicker, better care for patients. Looking forward, there's an opportunity to incorporate additional data types, create real-time systems, and enhance understanding of AI, Deep Learning, and Machine Learning decisions, thus making these tools even more dependable for everyday clinical applications.

Keywords: Bone fracture identification, Deep learning, Convolutional neural networks (CNN), Medical image categorization, X-ray evaluation, Transfer learning, Computer-aided diagnosis (CAD), Medical imaging.

Abstract: This project introduces a smart home automation system built on Internet of Things (IoT) technology. It connects everyday household devices through sensors and microcontrollers, enabling them to exchange data and operate intelligently. The system is developed using the Arduino IDE for collecting and displaying data, as well as for remotely controlling appliances. A NodeMCU ESP8266 module serves as the main controller, handling both sensing operations and wireless communication. The goal of this project is to design an efficient and user friendly platform that automates home appliances, reduces manual effort, and promotes energy conservation while enhancing overall comfort and convenience

Keywords: Microcontrollers, Arduino IDE, Wireless communication, Energy conservation

Abstract: Modern non-volatile memories, especially Multi-Level Cell (MLC) Flash and ReRAM, suffer from asymmetric and limited-magnitude errors that significantly degrade reliability. Traditional SEC–DED codes fail to correct multi-level and adjacent magnitude errors efficiently. This paper presents the design and FPGA implementation of Limited-Magnitude Error Correction Codes (LM-ECC) for MLC memory systems. The architecture includes SL-SEC, ML-SEC, and IP-SEC-DAEC designs synthesized on Xilinx FPGA. Comparative results demonstrate that the proposed design achieves lower area, reduced power consumption, and improved error-correction efficiency compared to existing techniques. The implementation is validated using Xilinx Vivado with optimized power, timing, and area metrics.

Keywords: MLC Memory, Error Correction Codes, Limited Magnitude Errors, SEC-DAEC, FPGA Implementation, Xilinx, Digital Design.

Abstract: This smart wheelchair integrates IOT and AI technologies to improve mobility and independent for individuals with disabilities. By combining hand gesture recognition, voice command processing, and intelligent automation, it provides a flexible, user-friendly, and adaptive solution for diverse needs. The IOT framework allows real-time monitoring and communication between the wheelchair and connected devices, enabling remote supervision and diagnostics. Hand gesture control, via motion sensors, ensures intuitive self-navigation, while the voice recognition module, powered by AI, interprets spoken commands for users with limited hand mobility. AI optimizes movement, detects obstacles, and ensures safe navigation. Additional features, such as emergency stop, fall detection, and health monitoring, enhance safety and provide timely alerts to caregivers. This multi-modal control system empowers users with greater independence, safety, and comfort, representing a significant advancement in smart assistive technology, fostering inclusivity and improving quality of life for people with mobility impairments.

Keywords: IOT (Internet of Things), AI (Artificial Intelligence), Hand gesture recognition, Voice command processing, Health Monitoring.

Abstract: Indoor Air Quality (IAQ) greatly affects health, comfort, and sustainability in modern spaces. This project introduces an AI-powered Indoor Air Wellness System that uses Kalman Filter and LSTM algorithms to predict and monitor air quality in real time. Sensors for CO₂, CO, NO₂, temperature, humidity, and light work with an ESP32 microcontroller to process and share data via LCD, ThingSpeak, and Telegram. A solar-powered fan system replaces traditional HVAC, promoting energy efficiency and eco-friendly air circulation. By combining AI, IoT, and renewable energy, the system offers a smart and sustainable way to maintain healthy indoor environments.

Keywords: Indoor Air Quality (IAQ), AI, Kalman Filter, LSTM, ESP32, IoT, Smart Monitoring, Renewable Energy, Energy Efficiency, Sustainability

Abstract: Phishing and smishing attacks have rapidly increased in digital communication platforms, exploiting user trust to steal personal and financial information. Traditional blacklist and rule-based detection systems lack adaptability and fail to detect evolving or zero-day attacks. Although deep learning has shown promise in text-based threat detection, existing research often focuses on a single architecture, lacks real-world datasets, or provides limited benchmarking across models. To address these gaps, this study presents a comparative evaluation of three deep learning models—RNN-LSTM, RNN-GRU, and GloVe-enhanced LSTM—for phishing and smishing text classification. A dataset of 27,000 real-world messages, collected from cybersecurity units and extended with controlled synthetic samples, was preprocessed using tokenization, stemming, padding, and semantic embeddings. Each model underwent structured hyperparameter optimization with dropout, L2 regularization, and early stopping to enhance generalization. Experimental results show that the GloVe-LSTM model achieved the highest performance with 90.07% test accuracy and a 90.16% F1-score, closely followed by tuned LSTM and GRU models. Statistical validation using a McNemar test confirmed no significant difference in model performance (p > 0.05). These findings demonstrate that semantic embeddings significantly improve phishing and smishing detection accuracy, supporting scalable deployment in cybersecurity systems such as email filtering, telecom SMS gateways, and digital fraud prevention platforms.

Keywords: Phishing Detection; Smishing; Deep Learning; LSTM; GRU; GloVe Embeddings; NLP; Cybersecurity

Abstract: The increasing burden of anxiety and depression requires accessible and effective monitoring tools. This study presents an AI-powered mental health monitoring system that predicts mental states, integrating self-reported text, wearable data such as heart rate variability (HRV), and video-based emotion analyses. The physiological changes detected from wearable devices are evaluated by Random Forest, while facial emotional signs are detected using CNNs to process video inputs. A contextual RNN is employed in the processing of emotions in synthesized text and Cognitive Behavioral Therapy (CBT). In view of the findings, high individual-level stress, emotional patterns, and mental health risk could be identified through the prediction of a moderate to critical level of risk. This integrated model improves the accuracy, accessibility, and real-time tracking of mental health indicators. The accessible monitoring approach underpins its importance and value for improving early diagnosis of mental health and possibly warning clinicians about emerging symptoms. Classification models were developed using an annotated dataset, experimental outcomes, and machine learning techniques.

Keywords: Mental health monitoring, Wearable devices, HRV, Random Forest, Video emotion analysis, Cognitive Behavioral Therapy (CBT), Random Forest, Convolutional Neural Networks (CNNs), LSTM, Contextual AI Systems.

Abstract: Technological evolution has significantly modernized everyday processes, including food service systems within educational institutions. This project, titled “KDK College Canteen Ordering Portal / Web Server”, introduces a QR-enabled digital menu and online ordering platform specially developed for the KDK College canteen. The system’s core aim is to streamline and automate food ordering by integrating web technologies and online payment services. Users can scan a QR code to instantly view the menu, place their orders, make secure transactions, and receive a unique token for order pickup. This approach removes the need for manual order-taking, reduces long queues, and minimizes human error. The portal also features an admin dashboard for managing menu items, tracking sales, and accessing order logs, thereby enhancing overall operational efficiency. The proposed solution supports cashless payments, eco-friendly digital menus, and quicker service delivery—aligning with smart campus initiatives. By digitizing traditional processes, the system enhances user convenience and contributes to a more organized and tech-driven campus environment.

Abstract: Deciding visible and invisible portion of the line and discarding invisible line segments from line is known as Line Clipping.

Keywords: Region Codes, Outcodes.

Abstract: Software testing is a critical and integral phase within the software development lifecycle, aiming to evaluate a software application to identify defects, ensure its adherence to specified requirements, and ultimately enhance its overall quality, reliability, and performance. This systematic process encompasses various techniques and methodologies, including both manual and automated approaches, as well as functional and non-functional testing types. The objective is to proactively detect and mitigate bugs and errors, thereby reducing post-release failures, improving user experience, and minimizing maintenance costs. The increasing complexity of modern software systems necessitates robust testing strategies and the continuous integration of testing practices within agile and DevOps frameworks. Ultimately, effective software testing is essential for delivering high-quality products that meet customer expectations and build trust in the competitive digital landscape. Software testing is the process of identifying the correctness and quality of software program. Software testing is a process with intending to find defects. Software testing is done by software tester. Developer can also make error.

Keywords: Software Testing, errors, bugs, test case, test plan.

Abstract: Lung cancer remains the leading cause of fatalities in all patients with cancer worldwide, thus reflecting highly on the urgent requirement for early detection and diagnostics. This abstract describes a summary of the different databases and methods using an artificial neural network (ANN) algorithm for lung cancer diagnosis. For the deep learning models, we need annotated CT scan images. They are accessible in publicly available datasets such as the Lung Image Database Consortium (LIDC), Lung-PET-CT-Dx, and NSCLC-Radio genomics. These datasets have contributed to the automation of lung cancer diagnosis. Tumor detection and classification are also performed based on X-rays and PET scan imaging data. Several ANN-based methods have been reported for optimal detection of lung cancer. The employed methods included: hybrid learning schemes, data augmentation, feedback in neural network training, multilayer perceptron’s and radiomic feature extraction. These methods aim to enhance the diagnostic accuracy by reducing the false positive rate and help physicians spot malignant nodules early on.