VOLUME 12, ISSUE 12, DECEMBER 2025

Abstract: Air pollution remains one of the most pressing environmental challenges of the 21st century, it comes with severe consequences for both public health and ecological balance. Prolonged exposure to pollutants such as particulate matter, carbon dioxide, and volatile organic compounds has been linked to respiratory illnesses, cardiovascular diseases, and even premature mortality. Despite these risks, conventional air quality monitoring systems are often limited by high costs, fixed infrastructures, and restricted accessibility, leaving large populations without adequate real-time information. To address this gap, this study presents Vento Aureo, an IoT and Artificial Intelligence (AI)-based framework designed for real-time air quality monitoring and forecasting. The system leverages portable IoT sensors to collect pollutant data, which is later transmitted to the cloud for analysis. Machine learning algorithms are employed to identify patterns and predict short-term air quality trends, enabling proactive responses to hazardous conditions. Data visualization and user interaction are facilitated through a mobile application that delivers live readings and predictions directly to end-users, supporting informed decision-making in daily life. Furthermore, the framework holds potential to aid policymakers and urban planners by providing accessible, large-scale insights into pollution dynamics. By integrating portability, affordability, and predictive intelligence, Vento Aureo offers a practical step toward mitigating the harmful effects of poor air quality and promoting healthier urban environments.

Keywords: Air Quality Monitoring, Internet of Things (IoT), Machine Learning, Noise Pollution, Real-Time Data, Cloud Integration, Smart Environment.

Abstract: Current tracing pad solutions often present several problems which may only operate if the device was plugged in all the time from the source of electricity. Traditional light boxes are bulky and require a constant power source, limiting portability and convenience. A conventional transparency is removably mounted in an opening in front of a box containing a translucent shade, which is mounted to pivot about its upper edge toward and away from the transparency, and to closed and open positions, respectively. A first lamp adjacent the lower edge of the shade illuminates the transparency when the shade is in its open position, and a second lamp rearward of the shade may be energized to superimpose an image of the shade onto the transparency, when the shade is closed (William A. Heindl, Jr., 1970) [1]. This study addresses the gap by developing an LED tracing pad mounted to a removable case and adjustable stand, integrated with a 235mm x 350mm solar panel compartment and a partitioned for drawing instruments, offering artists a portable, sustainable, and user comfort device for detailed tracing and design projects. The device allows LED tracing pad to use rechargeable that is generated by booth solar and alternating current (AC) in charging processes. The LED tracing pad with a rechargeable module supplied with power source from both solar and alternating current (AC) would be a versatile tool for drafting students, architecture students, artists, illustrators and designers in tracing designs, drawings or patterns in a maximum size of 345mm x 475mm. It combines the benefits of a traditional tracing pad with the added convenience of portability and sustainable charging options which the device doesn't need to be plugged in all the time. A rigorous evaluation, involving 50 experts and end-users, assessed the device's performance using a five-point Likert scale. These evaluators were selected through purposive sampling based on their expertise. The testing and evaluation took place at Capiz State University-Main Campus, while the development work was done at the researcher's residence. Testing revealed that the rechargeable module using 5V rechargeable battery has sufficient capacity to provide several hours of continuous use for 4 hours with a charging indicator to show charging status and battery level. The device's aesthetic, industrial-grade design prioritizes both user-friendliness and safety. Overall, the device received a "Very Acceptable" rating, demonstrating its effectiveness of its design, technical features, operating performance and composition that integrates a rechargeable module with safe wiring, ensuring continuous, sustainable power for use.

Keywords: LED, Rechargeable Module and Tracing Pad

Abstract: Mobile communication has evolved from basic voice services to advanced, high-speed networks that drive global connectivity. The introduction of fifth-generation (5G) technology, which provides ultra-fast speeds, low latency, high reliability, and mass device connectivity, is a major leap forward. This is accelerating the growth of the Internet of Things (IoT), enabling seamless communication between billions of devices, and supporting applications across a wide range of sectors. This paper examines the current state of mobile communication, focusing on the technological advancements brought by 5G and its integration with IoT. It highlights key benefits such as increased network capacity and real-time data processing while addressing challenges such as infrastructure costs. The study further explores the transformative impact on healthcare, education, entertainment, and industrial automation. By analysing industry trends and emerging research, this work provides insight into the evolving mobile ecosystem and outlines the opportunities and challenges shaping the transition to 6G and future networks.

Keywords: mobile communication, IoT,2G,3G,4G,5G, AR, VR.

Abstract: The classic marketing mix (4Ps → 7Ps) is no longer sufficient in an AI-first, platform-dominated economy. This article proposes the 10Ps framework - extending the traditional 7Ps with three digitally-native pillars: Personalization, Platforms and Performance Analytics, as the strategic foundation for modern marketing. Growth hacking, with its relentless experimentation, A/B testing, and data loops, serves as the operational engine that activates these new Ps. Using India's $5B+ quick-commerce war (Zepto, Blinkit vs. Swiggy, Flipkart) and IndoAI's edge-camera marketplace as live case studies, the paper demonstrates how the 10Ps create compounding growth flywheels once the "minimum viable data threshold" (~50-100k engaged users) is crossed. It also exposes growth hacking's cold-start paradox and introduces an 11th meta-principle, Data Threshold Awareness, to prevent romanticising a methodology that remains structurally inaccessible to most early-stage ventures.

Keywords: Marketing Mix 3.0, 7Ps, 4Ps, Personalization, Platform, Performance Analytics, Growth Hacking, AI-Driven Marketing, Q-Commerce, Indoai, AI Camera

Abstract: The 3I-ATLAS is part of NASA's effort to transition from legacy systems to cognitive data systems for its science and mission operations; this cognitive data system will enable the use of Artificial Intelligence (AI) and Big Data to operate in an Integrated Environment. The 3I-ATLAS is designed to ingest, integrate and interpret multiple domain sources of data automatically; through the use of deep learning-based pipelines, knowledge graphs, and cloud-native orchestration the 3I-ATLAS enables real-time analytics, semantic reasoning and predictive maintenance. This paper reviews the design of the 3I-ATLAS with emphasis on the AI aspects as well as the governance aspects of the 3I-ATLAS, with an eye to how these concepts are advancing NASA's Digital Transformation efforts toward developing Autonomous, Trustworthy, and Interoperable Space Data Ecosystems.

Keywords: NASA 3I ATLAS, Artificial Intelligence, Big Data, Semantic Integration, Cognitive Computing, Interoperability, AI Governance.

Abstract: In order to retain profitability and guarantee long-term sustainability, organizations must get adequate commercial value from the use of business analytics. In the rapidly evolving telecommunications industry, businesses based in Bangladesh are having to deal with the issue of improving their performance in order to remain competitive. Nevertheless, when it comes to the effect that business analytics has on the performance of an organization in this particular context, there is a dearth of research that has been conducted. Recently, business intelligence and analytics have emerged as a strategic approach that may be used in management tasks, providing opportunities to improve the effectiveness of operations. Despite the fact that there is a growing interest in the evaluation of examples of analytics adoption, to the best of our understanding, few studies have been conducted to investigate the effects of a data-driven culture and the repercussions of the adoption of business analytics, particularly with regard to how it impacts the performance of managers in their jobs. Every company is affected by the changes that occur in a dynamic corporate environment throughout time. Two essential variables that have a significant impact on the success of a firm are the implementation of technological improvements and the diversity of products. As a result, pre-selected superior strategies are often inadequate, which makes it necessary to identify and develop new strategies in order to increase the competitiveness of the organization. This demonstrates the importance of the abilities and resources that firms must develop in order to achieve a competitive edge from a resource-based strategy approach.

Keywords: Business Analytics, Process Agility, Competitive Advantage, SEM

Abstract: The developed system, "Nurturing Plant with Smart Water and AI Care for Sustainable Growth," introduces a lightweight smart-agriculture framework that integrates IoT sensors and AI-based intelligence for real-time plant monitoring and automated care. Soil moisture, temperature, humidity, and water level are continuously measured using embedded sensors to provide precise environmental insights. A predictive machine learning-based irrigation mechanism analyzes current readings and historical patterns to supply the optimal amount of water, preventing both over-irrigation and water stress. In parallel, an AI image-processing module identifies plant diseases and nutrient deficiencies at an early stage, enabling timely intervention and reducing crop losses. The combined automation significantly minimizes manual effort, enhances water efficiency, and supports sustainable plant growth. With its modular and scalable design, the system is suitable for home gardening, greenhouse setups, and large-scale agricultural environments, demonstrating how IoT and AI can together improve plant health and resource management.

Keywords: Smart Agriculture, IoT, Artificial Intelligence, Deep Learning, YOLOv8, Leaf Disease Detection, Soil Monitoring, Smart Irrigation, ESP8266, RS485, Predictive Water Management, Real-Time Monitoring, Sustainable Farming, Machine Learning, Web Interface, Precision Agriculture, NPK Sensor, Automation System, Plant Health Analysis.

Abstract: This project presents an advanced real-time feedback system designed to elevate public speaking skills by performing an integrated audio-visual analysis through webcam input. The system intelligently interprets key non-verbal cues such as posture alignment, gesture consistency, facial orientation, and eye-contact patterns while simultaneously assessing crucial speech metrics including filler-word frequency, speaking speed, articulation clarity, and vocal modulation. By providing immediate, data-driven feedback and structured progress summaries, users can steadily refine their communication style and presentation effectiveness. The platform is developed using Streamlit for a smooth and interactive interface, supported by a robust backend that integrates Convolutional Neural Networks (CNNs) for body-language assessment, Hugging Face NLP models for speech interpretation, and Librosa for comprehensive audio feature extraction. Trained on a diverse collection of annotated public speaking recordings, the system delivers reliable and context-aware insights while upholding strict standards of data privacy and ethical compliance. Extensive evaluations confirm its accuracy, responsiveness, and adaptability. With continuous enhancements guided by real user feedback, this AI-powered solution makes professional grade public speaking training more accessible, scalable, and personalized for learners across all backgrounds.

Keywords: Public speaking, real-time feedback, body language, speech analysis, CNN, Hugging Face, Librosa, NLP, audio-visual processing, feature extraction, user interface, Streamlit, Tkinter, machine learning, deep learning, emotion detection, posture, gestures, eye contact, and filler word.

Abstract: In today's world, safety and security of homes have become one of the primary concerns due to increasing incidents of theft, fire hazards, and environmental threats. To address this issue, our project proposes a smart Home Security and Automation System using Internet of Things (IoT) technology. The system integrates five sensors - PIR sensor for motion detection, MQ2 sensor for gas leakage detection, DHT11 sensor for temperature and humidity monitoring, Fire sensor for flame detection, and an Ultrasonic sensor for automatic water tank level detection and filling. These sensors are connected to a microcontroller, which continuously monitors the surrounding environment. In the event of any abnormal activity such as intrusion, gas leakage, sudden fire, or critical temperature rise, the system immediately triggers an alert. The system is integrated with the Blynk IoT platform, which enables real-time notifications to be sent via email to the homeowner, ensuring instant awareness of the situation. Additionally, the ultrasonic sensor automates the water filling process, thus reducing manual effort and ensuring water availability. This project not only enhances security bualso adds automation for convenience and energy efficiency. The system is lowcost, easy to install, and scalable for future improvements like AI-based detection and voicecontrolled automation. Hence, the proposed solution provides a reliable, userfriendly, and effective approach for securing homes while offering intelligent automation features.

Keywords: IOT, Home Automation, Security System, Smart Sensors, Blynk IOT

Abstract: The pharmaceutical sector in India is undergoing a paradigm shift, driven by rapid digitization and increasing internet penetration. E-pharmacies online platforms facilitating the sale of medicines have emerged as a significant disruptor, promising accessibility, affordability, and convenience. However, this burgeoning sector operates within a complex ecosystem fraught with challenges. This paper explores the multifaceted hurdles facing e-pharmacies in India, categorized into regulatory ambiguity, operational logistics, patient safety concerns, and the socio-economic conflict with traditional brick-and-mortar retailers. Through an analysis of the current legal framework and market dynamics, this study argues that while e-pharmacies hold immense potential for public health, a robust and clear regulatory mechanism is a prerequisite for sustainable growth.

Abstract: In the development of neurodegenerative diseases including Alzheimer's disease (AD), Parkinson's disease (PD), multiple sclerosis (MS), amyotrophic lateral sclerosis (ALS), and Huntington's disease, neuroinflammation has been recognized as a major role player. It is characterized by persistent microglia and astrocyte activation, excess release of cytokines, and oxidative stress leading to neuronal homeostasis disturbance and neurodegeneration. In this extended review, we provide a snap-shot of some of the key players and molecular constituents involved in chronic neuroinflammation. It also examines pharmacological modulation therapies via cytokine signaling, glial activation and oxidative stress pathways for the reduction of neuronal loss. Emerging data emphasizes the influence of novel therapeutic drugs like inflammasome antagonists, kinase inhibitors and neuroprotective nutraceuticals on modulation of inflammation. The review also considers challenges of translating preclinical observations to clinical benefit, such as restrictions associated with the blood-brain barrier, drug toxicity and heterogeneity in disease. Elucidating neuroinflammatory mechanisms may lead to safer and more effective pharmacological treatments for neurodegenerative diseases.

Keywords: Neuroinflammation; Neurodegeneration; Microglia; Astrocytes; Cytokines; Alzheimer's disease; Parkinson's disease; Multiple sclerosis; Pharmacological modulation; Neuroprotection.

Abstract: Viral diseases present on one of the most severe threats to global agriculture, frequently results in drastic yield losses and major economic setbacks for farmers. A crucial factor enabling the persistence and rapid transmission of plant viruses is the presence of weeds, which serve as alternate hosts and reservoirs for viral pathogens. Weeds not only harbor these viruses but also support insect vectors, such as aphids, whiteflies, leafhoppers, and thrips, that facilitate the swift spread of the diseases to nearby crops. Examples of this epidemiological link include Cynodon dactylon, which carries Rice Tungro Virus, and Parthenium hysterophorus, which is linked with Tobacco Streak Virus. This study aims in understanding the pivotal role of weeds in the epidemiology of crop viral diseases and evaluate integrated strategies designed to mitigate their impact. Effective virus management relies heavily on the framework of Integrated Weed Management (IWM), which strategically combines cultural practices, mechanical removal, biological agents, and selective chemical methods. The primary goal of IWM in this context is to reduce viral inoculum sources and effectively break the disease cycle. It is essential that crop protection prioritize for removal of harmful, virus-hosting weeds rather than indiscriminate eradication, as some weeds contribute positively to ecological balance by supporting pollinators and soil health. Research supports these sustainable efforts through the study of plant-based extracts, like neem, which act as natural antivirals and biopesticides.

Keywords: Crop protection, Weeds, Viral diseases, Virus reservoirs, Integrated Weed Management, Plant health.

Abstract: Smart health and nutrition management are crucial because of the rapid increase in lifestyle-related diseases like obesity, diabetes, and hypertension. This paper introduces Smart Health Diet Planner, an intelligent AI-driven system that creates personalized diet plans by analyzing health data and using nutrition-based machine learning. By using predictive algorithms and customizable food databases, the system offers accurate calorie estimates, balances macronutrients and micronutrients, and provides diet recommendations tailored to specific conditions. The planner combines user health profiles, activity levels, and dietary preferences to create optimized daily meal plans. Results show that the system greatly enhances adherence to healthy eating habits and leads to better health outcomes. The proposed diet planner emphasizes the significant role of artificial intelligence in personalized nutrition and preventive healthcare.

Keywords: AI-Driven Diet Planning, Personalized Nutrition, Machine Learning in Healthcare, Micro Front-End Architecture, Scalable Web Application, Health Data Analysis, Preventive Healthcare

Abstract: Timely medical transportation determines survival outcomes in emergencies, yet many individuals continue to face delays due to traffic congestion, ambulance unavailability, or fragmented communication systems. VitaLink is a community- driven digital platform designed to connect patients in urgent need of transport with verified local service providers, including two-wheelers, four-wheelers, and autorickshaws. The platform prioritizes accessibility, real-time coordination, and flexible payment options-ensuring that financial limitations do not prevent patients from reaching healthcare facilities. This paper presents the complete design, methodology, implementation strategy, system behavior, and expected societal impact of VitaLink. By leveraging community participation and modern mobile technologies, VitaLink aims to create a reliable emergency mobility ecosystem suitable for urban, semi-urban, and rural environments.

Keywords: Emergency transportation, mobile healthcare, patient mobility, real-time systems, community networks.

Abstract: Although intrusion detection systems are still essential to defense, they frequently lack transparency and integrated rehearsal. This project provides a web-based intrusion detection system dashboard that integrates multi-vector attack simulation, interactive visualization, and real-time monitoring into a single workflow. With ten configurable attack families (SQL injection, DDoS, brute force, port scan, XSS, CSRF, MITM, phishing, buffer overflow, privilege escalation), synthetic, production-shaped telemetry feeds shared contexts that drive topology and infrastructure views while exercising the same detection path. Context-driven state propagation, deterministic heuristics, and integrated alerting maintain alignment between response and visualization. Experiments demonstrate that >98% accuracy with

Abstract: Oral cancer is a common and dangerous disease around the world, and how well someone survives often depends on how early it's found. Right now, doctors mostly rely on their own eyes and experience to spot it, which can take a lot of time, be influenced by personal judgment, and sometimes lead to mistakes. NeuroVision is working on solving these problems by creating a smart, automatic system that can find oral cancer early using deep learning and transfer learning methods.In this project, we use already trained convolutional neural network (CNN) models to look closely at images of mouth lesions, helping to tell the difference between cancerous and healthy tissues.Transfer learning helps the system learn faster and perform better, even when there aren't many medical images to work with.

Keywords: Oral Cancer, Early Detection, DeepLearning, Transfer Learning, NeuroVision, Convolutional Neural Network (CNN), Medical Image Analysis, Image Preprocessing, Classification, Diagnostic Support.

Abstract: Ragging remains a persistent problem in most colleges and a large number of students still show hesitation in reporting whatever they experience or witness [6], [7]. Most of the existing channels for reporting, such as helplines or complaint boxes or even emails, either ensure no confidentiality or are not easy to use, which leads to many incidents going unnoticed. This lack of a safe and dependable system creates fear among students and limits the institutional ability to act in time. In order to tackle such challenges, the Campus Shield portal is designed as a secure and user-friendly digital place where complaints can be submitted by students on a confidential or anonymous basis. Evidence could be uploaded in the form of images or videos, besides tracking the complaint progress. The included administrative dashboard will, in turn, help the authorities to review the complaints and act upon them as quickly as possible. Strong data protection along with a structured workflow forms part of the portal for ease of reporting, building trust, and aiding quicker intervention[1].

Keywords: Anonymous Reporting, Student Safety, Secure Complaint System, Anti Ragging Portal, Campus Security.

Abstract: Fraud detection has emerged as a vital area of research in the era of digitalization, where financial transactions and online services have become increasingly vulnerable to fraudulent activities. With the expansion of e-commerce, online banking, insurance claims, and telecommunication services, identifying and preventing fraud has become a major challenge for organizations. Traditional rule-based systems, while effective for structured and historical data, often fail to detect new and adaptive fraud patterns. As a result, modern fraud detection systems increasingly rely on advanced data-driven approaches such as Data Mining, Machine Learning, Deep Learning and Artificial Intelligence to recognize suspicious behavior and anomalies in real time. A comprehensive review of current fraud detection techniques, including supervised learning, unsupervised learning, and hybrid models combine these approaches. It highlights widely used algorithms such as Decision Trees, Random Forests, Neural Networks, Support Vector Machines, and anomaly detection frameworks. Furthermore, it discusses key performance metrics like Precision, Recall, and ROC-AUC, which are essential for evaluating detection efficiency. It also addresses major challenges such as data imbalance, privacy concerns, lack of labeled datasets, and the dynamic nature of fraud schemes. Finally, it outlines emerging research trends focused on explainable Artificial Intelligence (AI), Graph-based Detection, and adaptive learning systems, offering insights into future pathway for building more accurate and resilient fraud detection mechanisms.

Keywords: Fraud Detection, Deep Learning, Graph Neural Network (GNN), Real-Time Credit Card Fraud, Banking Security, Artificial Intelligence.

Abstract: Job satisfaction and job stress among nurses are critical factors that impact turnover rates. Analyzing the job satisfaction levels of female nurses in specific hospitals within Vijayapura City is important for several reasons. Firstly, the healthcare sector relies heavily on nurses to provide essential care and support to patients. Examining the various factors influencing their job satisfaction and its correlates in this study helps further to improve the working conditions of nursing staff.

Keywords: Health Profession, job satisfaction, Job stress, Private Hospital

Abstract: COVID-19 is just spreading like a wildfire in the world that brought enormous changes, and one of the unbelievable changes that happened is Online Education (The Economic Times). The rapid expansion of digital technology and the sudden transition to online learning during the global pandemic have reshaped the educational experience for students in ways that continue to influence their academic performance and overall wellbeing. Recent times the online education have also taken many changes and many easy escapes of attending the same online education classes. The benefits of online education are limited in certain population. In the same way the study talks about online education impact on student health and performance. In this study we take different questions related to the variable online education, student health and student performance. The study mainly focused on students and a questionnaire is shared with different students. Using SPSS and tests like Anova, Correlation and T-test using the variables we got to know that the relation between student academic performance in online and offline. There is also a relation between different factors of student health and performance. The study focusses on students of different age group and finding the difference between the groups. So, this study helps in knowing the importance of online learning and the relation between student academic performance online and offline.

Keywords: Online education, student health, student academic performance, traditional education.

Abstract: The increasing deployment of surveillance systems has created a need for intelligent monitoring solutions that can automatically interpret visual data instead of relying solely on manual observation [1], [4]. DeepSecure - Suspicious Human Activity Recognition from Surveillance Videos addresses this challenge by integrating deep learning techniques with computer vision to detect abnormal or potentially dangerous activities in real time [1], [10]. The system is designed to identify suspicious human behaviours such as violent actions, panic movements, and unauthorized gatherings, while simultaneously detecting environmental hazards including fire and smoke [5], [6]. Using convolutional neural networks and YOLO-based object detection, DeepSecure effectively analyses both spatial and contextual information from live and recorded video streams [1], [10]. A Flask-based web application enables users to interact with the system through a browser interface, supporting both live camera feeds and uploaded surveillance footage [4], [10]. OpenCV is employed for efficient video processing, and a MySQL-backed authentication mechanism ensures secure access control [4]. The modular design of the system allows flexible deployment across public and industrial environments such as airports, educational institutions, and smart-city infrastructures.

Abstract: This paper investigates the dynamics of algal biodiversity and water quality of Rushula Cheruvu, which is a historically important, rain-fed freshwater Rushula Cheruvu (a traditional freshwater water body) of Kakatiya origin and is found in the Nagarkurnool District of Telangana, India. Five representative stations were sampled monthly and a single annual cycle was followed (December 2024-November 2025) to assess the phytoplankton composition and significant physico-chemical parameters. Eighty-seven algal species (Bacillariophyceae, Chlorophyceae, Cyanophyceae, Euglenophyceae, and Dinophyceae) were identified. The seasonal changes displayed more diversity in winter and post monsoon seasons than in summer and monsoon seasons, whereby there were increased prevalence of cyanobacteria. The multivariate statistical tests, including Principle Component Analysis, Cluster Analysis and Multiple Linear Regression, included the nutrient availability, temperature, and light penetration as the key factors in the control of the algal community structure, where phosphorus played a significant role in the determination of algal abundance. The Rushula Cheruvu (a traditional freshwater water body) was determined as meso-eutrophic with seasonal trends towards the eutrophic conditions in the seasons of environmental stress. The Rushula Cheruvu, especially, is not given water except through rainfall and runoff of forest catchment water, and no farming or fertilizers are used, it is a vital source of drinking water to tribal populations and wildlife. The results show the ecological importance of this almost pristine freshwater system and the necessity to maintain its natural water quality and biodiversity by strict conservation and long-term monitoring.

Keywords: Algal Biodiversity, Water Quality, Multivariate Analysis, Eutrophication, Phytoplankton Dynamics

Abstract: This project presents an IoT-based crowd density mapping system designed to enhance safety during emergency evacuations. Using YOLOv5 for real-time people detection, the system monitors crowd levels and triggers alerts when thresholds are exceeded. An ESP32 microcontroller controls indicators, doors, and notifications based on the detected density. The solution enables faster decision-making and effective crowd management in public spaces.

Keywords: Crowd Detection, YOLOv5, IoT, Emergency Evacuation

Abstract: This work introduces an explainable deep learning system that is designed for road scene feature classification based on a convolutional deep learning network VGG16 in combination with SHAP for explanation. The system processes the input road images by resizing the images to 160 x 160 x 3, normalized the pixel values and made predictions from six possible classes: HV, LMV, Pedestrian, Road Damages, Speed Bump and UnsurfacedRoad. A filtering mechanism based on confidence is included, whereby if the predictions are below a threshold the results will be labeled Plain Road to avoid uncertain outcomes. In order to produce constant explanations, a balancing of data set using TensorFlow's ImageDataGenerator is created from test images and stored for SHAP initialization of background data set. SHAP is then used with an Image masker with inpaint_telea in order to generate pixel-wise attributions to indicate the most influential areas in the input image. The whole system is implemented using Streamlit interface including image uploading, real-time prediction, class probability visualization, and explanation heatmaps rendered using Matplotlib. All technical components such as preprocessing, caching, background generation and explainer setup are directly derived from the project's code which has been implemented. The resulting framework helps to add transparency in road feature classification with the combination of deep learning performance and interpretable visual outputs to help users understand what the model is reasoning about.

Keywords: RoadFeatureClassification DeepLearning VGG16 SHAP ExplainableAI Streamlit ImageProcessing ComputerVision RoadSafety FeatureVisualization.

Abstract: Greenhouse farming requires continuous monitoring of environmental parameters to ensure optimal crop growth and yield. Traditional greenhouse management systems rely heavily on manual supervision, which is time- consuming, error-prone, and inefficient. This paper presents an IoT-based automated greenhouse monitoring and control system integrated with security management to address these challenges. The proposed system continuously monitors critical parameters such as temperature, humidity, soil moisture sensors connected to an ESP32 microcontroller. The collected data is transmitted securely to a cloud platform for real-time visualization and analysis. Automated control mechanisms are implemented to regulate irrigation, ventilation, and lighting based on predefined threshold values. Additionally, secure data transmission, user authentication, and remote access features are incorporated to protect sensitive agricultural data. Experimental results demonstrate improved resource utilization, reduced human intervention, and enhanced crop growth conditions, making the proposed system suitable for modern smart agriculture applications.

Keywords: Internet of Things (IoT), Smart Greenhouse, ESP32, Environmental Monitoring, Cloud Computing, Secure Agriculture.

Abstract: The rapid advancements and integration of Artificial intelligence (AI) its impact across industries has extended to Human Resource Management (HRM) redefining workforce management. The potential applications of artificial intelligence (AI) in human resource management (HRM) are all explored in this paper. AI-driven tools and techniques optimize the recruitment, performance management and employee engagement. The AI promises to enhance decision-making efficiency and accuracy as conventional HRM practices are time consuming and biased. The paper studies the role of AI as an important to carry out the various functions of human resource practices, where AI can handle recruitment, hiring, performance appraisal, allocating the Jobs, reducing workload at workplace and increasing workplace efficiency. The new applications for AI in HRM, sentiment analysis, predictive analytics, intelligent decision support and personalized employee experiences are reviewed. The integration of AI into HRM poses a number of difficulties like bias, privacy issues, and transparency which are some of the attributes having ethical and legal ramifications of using AI in decision-making processes that are pointed in this study. To effectively deal with this change, strategies including work role redefinition, employee skill development and having a collaborative atmosphere between humans and AI are suggested. The possible advantages and breakthroughs that AI might bring to HRM practices are highlighted as the future perspectives of AI in HRM are examined. AI furnishes from selected studies, revealing adoption rates, prevalent techniques, and sector specific implementations and also gives brief understanding of the future goal of artificial intelligence.

Keywords: Artificial Intelligence, Human Resource Management, data governance, work culture.

Abstract: Helmets are essential safety equipment worn by motorcycle riders, athletes, traffic police officers, and industrial workers to provide critical head protection against traumatic impacts and injuries, yet conventional designs create significant thermal discomfort that undermines their protective benefits and reduces user compliance. The internal microclimate created within traditional helmets results in elevated temperatures, excessive perspiration, and diminished cognitive performance, particularly in tropical climates or during high-intensity activities, creating a critical gap in helmet technology that demands innovative solutions. To address this thermal challenge, we propose an advanced Active Thermoelectric Cooling Helmet System that integrates semiconductor cooling technology with renewable energy sources by utilizing the Peltier Effect-a principle where electric current flowing through a junction of dissimilar materials (copper and bismuth thermocouple elements) activates a heat-pumping mechanism that generates cooling on one side and heat rejection on the opposite side. The system architecture comprises four primary components: a Peltier module positioned within the helmet structure, a primary inlet fan that draws cool air directly from the Peltier's cold side into the helmet interior, a secondary exhaust fan that efficiently removes heated air and excess moisture from the helmet cavity, and a sophisticated Battery Management System (BMS) that regulates and distributes electrical power while optimizing energy efficiency. Power supply and energy independence are achieved through a solar photovoltaic (PV) array installed on the helmet's exterior surface with a custom-designed supporting frame that provides continuous renewable energy generation during daylight hours, complemented by a rechargeable battery storage system serving as an energy buffer to ensure uninterrupted cooling operation during low-light conditions and variable solar irradiance. This innovative design delivers multiple significant operational advantages including enhanced thermal comfort enabling prolonged comfortable use, improved user compliance through superior comfort levels, maintained cognitive function and alertness, autonomous renewable energy operation eliminating grid dependency, extended operational efficiency across diverse user groups, and versatile applicability for motorcycle riders, athletes, traffic enforcement personnel, rescue workers, and industrial workforce in hazardous environments. Preliminary testing reveals temperature reductions of 8-12°C compared to conventional helmets with battery performance demonstrating 6-8 hours of continuous operation under standard sunlight conditions, validating the thermoelectric design approach and establishing practical viability. Current limitations include increased helmet weight due to solar panel integration and initial manufacturing costs exceeding conventional helmets, with cooling performance varying based on ambient temperature and solar irradiance; however, future iterations will focus on lighter materials, improved battery capacity, and enhanced solar cell efficiency to optimize performance. The thermoelectric cooling helmet system represents a significant advancement in personal protective equipment engineering, successfully addressing the critical thermal discomfort challenge that limits conventional helmet use by combining semiconductor cooling technology, renewable energy systems, and intelligent power management into a cohesive protective solution with substantial potential to improve thermal comfort, enhance wearer safety, extend operational efficiency, and establish new standards for next-generation helmet technology applicable across law enforcement, occupational safety, motorsports and emergency response sectors worldwide.

Keywords: Helmet, Peltier module kit, Battery, Battery management system and Solar panels.

Abstract: Gamified learning integrates game-like elements, such as points, levels, and challenges, into educational content. This approach addresses the critical problem that current programming education often lacks the mechanisms necessary to sustain student motivation, provide immediate feedback, or reward incremental progress, leading to high dropout rates and ineffective learning. In programming, students often struggle with abstract concepts, complexity, engagement, and motivation. By integrating game mechanics, gamification transforms routine coding tasks into interactive and enjoyable experiences, thereby making learning more accessible and enhancing retention through reward-based progress and instant feedback. The primary goal of this type of project is to design a gamified learning platform specifically aimed at making programming education more engaging for students of all levels. Specific objectives include enhancing engagement by turning passive learners into active participants, improving skill mastery through structured debugging exercises and challenges, and creating motivation through competition and collaboration. For implementation, the platform often uses a first-person 3D environment built in Unity, where players interact with gameplay elements, such as fighting enemies or progressing through levels, by correctly solving programming challenges. Docker containers are utilized in the architecture to securely execute player-submitted code, provide real-time results, and facilitate an iterative learning loop. This strategy cultivates resilience and a growth mindset, confirming that gamification significantly improves learning outcomes and student satisfaction.

Keywords: gamification, motivation, engagement, game-based learning, educational technology, learning outcomes.

Abstract: Many cardiac rhythm abnormalities develop gradually and may not produce immediate symptoms, making continuous monitoring essential for early identification. This work describes the implementation of a real-time cardiac monitoring system that integrates sensor-based data acquisition, wireless communication, and automated signal analysis. An AD8232 ECG sensor is used to capture cardiac electrical activity, which is processed by an ESP32 controller and transmitted to a cloud platform for storage and visualization. To support ECG analysis, additional parameters such as pulse rate and body temperature are also recorded. A web application developed using the Flask framework retrieves the stored data and applies machine learning models to identify irregular cardiac patterns. Experimental evaluation confirms stable signal acquisition, reliable heart rate computation, and accurate differentiation between normal and abnormal rhythms. The developed system offers a compact, low-cost, and scalable solution suitable for continuous cardiac monitoring in home and remote healthcare environments.

Keywords: ECG analysis, Arrhythmia detection, IoT healthcare, ESP32, AD8232, Cloud-based monitoring, Machine learning.

Abstract: Continuous pursuit of predictive and ethical drug testing systems has driven convergence between the fields of biotechnology and engineering. Emerging Modalities The preclinical research field has been significantly shaped by 3D bioprinting and various organoid platforms. These technologies pretend the native tissue architecture, cell interactions and physiological microenvironments to a much greater extent than the traditional two-dimensional (2D) cultures or animal models. 3D bioprinting allows for the layer-by-layer assembly of living tissues using bioinks that contain cells and biomaterials, whereas organoids, stem cell-derived self-organized mini-organs, can recreate a range of human physiological functions. Together, they have made amazing improvements in the accuracy of predicting drug activity, toxicity and metabolism, pushing us toward personalized therapeutics. This paper focuses on the development, rationale, and pharmaceutical use of bio printed organoid systems as well as provides overview of current challenges, ethical considerations and regulatory viewpoints. The fusion of AI, microfluidics and omics technologies is envisioned to evolve these platforms into autonomous patient-specific drug discovery ecosystems for the next era of precision medicine.

Keywords: 3D bioprinting, organoids, drug testing, pharmacology, tissue engineering, personalized medicine, regenerative pharmaceutics

Abstract: This research work is aimed to develop Castor oil/kerosene blend as alternative to petro-diesel. Castor oil was blended with kerosene and ethanol at different proportions according to extreme vertices mixture design. Density, Cloud point, Color, Aniline point, API gravity, Diesel index, Cetane number, Sulphur content, kinematic viscosity, flash point and fire point of the blends were determined and then statistically analyzed and optimized. The results obtained showed that the flash points of Castor blends 61.18oF, which implies that, the value of Castor blend is close to the ASTM D 93 standard value for flash point on diesel. The kinematic viscosity for Castor blends was found to be 27.20cst, which indicated that the blends have values near the approved ASTM value for kinematic viscosity. The sulphur content for Castor blends (0.023) were within the accepted range for diesel oil for sulphur content. The values obtained for the cetane number of Castor (30.39) are below the accepted value set by ASTM for diesel oil. When measurements on API gravity of castor blends (34.22) were made, they fell within the accepted range set by motor vehicle manufacturers. The aniline value for Castor blends was found to be 83.05oF, which are considered to be acceptable. It can therefore be concluded that in this research work it was found that the Castor blends have diesel characteristics.

Keywords: Castor oil, Petro-diesel, Density, Diesel Index, Biofuel

Abstract: This study uses a quantitative survey-based approach to investigate consumer awareness, perception, trust and purchase intent towards AI-generated influencers. A structured online questionnaire circulated online collected 101 responses, covering demographic details, familiarity with virtual influencers and attitudes measured through categorical and Likert-scale items. The findings suggest that most respondents are aware of AI-generated influencers having encountered their content but still perceive these virtual personalities as less authentic and trustworthy compared to human influencers. AI-generated influencers appear moderately appealing visually but lack trust and reliability resulting in low purchase intent for products recommended by AI-generated influencers. Human influencers continue to have stronger persuasive power, with higher credibility and emotional relatability. Overall, the study highlights that AI-generated influencers contribute to product discovery and digital engagement but their impact on consumer behaviour is limited. These insights provide a base for understanding the evolving role of AI-driven creators in the field of influencer marketing.

Keywords: AI-generated influencers, consumer perception, trust and credibility, purchase intent, influencer marketing

Abstract: Geopolitical events such as wars, trade conflicts, and diplomatic tensions have become major sources of uncertainty in global financial markets. These shocks influence macro-financial conditions and individual investment behaviour. The development of a Geopolitical Risk (GPR) index by Caldara & Iacoviello (2022) has enabled a systematic measurement of geopolitical uncertainty and its effect on risk pricing. Rising GPR has been shown to increase market volatility, reduce returns, and negatively affect investor confidence, especially in emerging economies (Salisu et al., 2022; Zhang et al., 2023). With the growing participation of retail investors and digital access to financial markets, it is crucial to understand how individuals respond to geopolitical shocks. Recent studies indicate that heightened GPR alters household participation choices, asset allocation, and hedging behaviours, often driving investors toward safe havens such as cash and gold or toward conflict-related sectors like defense equities (Agarwal et al., 2022; Cai et al., 2024; Klein, 2024; Lee, 2023). Additionally, geopolitical anxiety amplifies sentiment-driven responses and encourages herding behaviour, which may weaken rational investment decision-making (Guo & Shi, 2024; Medhioub, 2025). Despite growing evidence, research remains limited regarding long-term household wealth effects and how investor traits such as financial literacy or digital trading exposure mediate behavioural changes. This study seeks to address these gaps and contribute to a deeper understanding of how geopolitical risk shapes individual investment strategy.

Keywords: Geopolitical Risk, Investors, Market Volatility, Portfolio Choice, Safe-Haven Assets, Behavioural Finance

Abstract: The rapid penetration of electric vehicles (EVs) presents significant challenges and opportunities for modern smart grids. Uncoordinated EV charging can lead to increased peak demand, voltage deviations, and accelerated aging of distribution infrastructure, whereas coordinated charge scheduling can enhance grid reliability, reduce operational costs, and facilitate renewable energy integration. This paper presents a comprehensive review of EV charge scheduling strategies within smart grid environments. Various control architectures, including centralized, decentralized, and hierarchical approaches, are examined along with their corresponding optimization objectives such as cost minimization, peak load reduction, loss mitigation, and user comfort maximization. The review covers mathematical formulations based on deterministic, stochastic, and robust optimization, as well as emerging data-driven and reinforcement learning-based techniques for real-time scheduling under uncertainty. Bidirectional vehicle-to-grid (V2G) operations and their role in providing ancillary services are also discussed. Furthermore, commonly used datasets, simulation tools, and performance metrics for evaluating charging strategies are summarized. Finally, key challenges related to scalability, user behavior modeling, uncertainty management, and cyber security are highlighted, and future research directions toward intelligent, flexible, and user-centric EV charging frameworks are identified.

Keywords: Electric Vehicles (EVs), Smart Grid, Charge Scheduling, Smart Charging, Vehicle-to-Grid (V2G), Optimization Techniques, Reinforcement Learning, Demand Response, Renewable Energy Integration, Distribution Networks.

Abstract: The rapid growth of electronic devices has led to an alarming rise in electronic waste (e-waste), posing significant environmental and health hazards while also resulting in the loss of recoverable valuable materials. To address this challenge, SmartBinX introduces an intelligent, AI-driven framework that integrates Internet of Things (IoT) sensing, computer vision, and Generative Artificial Intelligence (GenAI) for efficient e-waste assessment, classification, and reuse optimization. The proposed system employs smart sensors and machine learning algorithms to measure and evaluate the material composition of discarded electronic products such as laptops, smartphones, and circuit boards. Generative AI models are further utilized to generate disassembly instructions, predict potential reuse pathways, and optimize recycling processes based on material recovery value.

Keywords: Electronic Waste (E-waste), Smart Waste Management, SmartBinX, Internet of Things (IoT), Computer Vision, Generative Artificial Intelligence (GenAI), Machine Learning, Material Composition Analysis, E-waste Classification.

Abstract: Industrial material movement in small and medium scale industries in India commonly depends on either low-cost Automated Guided Vehicles (AGVs) that follow fixed paths, or expensive Autonomous Mobile Robots (AMRs) with unnecessary sensing and processing overhead. AGVs lack path adaptability and halt when path conditions change, while AMRs are prohibitively costly for widespread deployment. This paper presents BHEEMA, a semi-autonomous navigation robot positioned between AGVs and AMRs. BHEEMA uses a modified region-focused Dijkstra algorithm running on a low-cost ESP32, along with a 5-sensor IR junction detection array for local navigation and Wi-Fi based coordination through a local server. The robot intelligently selects a subsection of the full map and performs local shortest-path planning, reducing computation while allowing dynamic routing. The system supports multi-robot coordination, collision prediction, and deadlock reduction inspired by the Braess Paradox. The design is highly cost-effective, scalable and suitable for Indian warehouse and cottage industry environments.

Keywords: AGV, AMR, Path Planning, ESP32, Dijkstra Algorithm, Swarm Robotics, Industrial Automation

Abstract: The rapid expansion of digital communication has significantly increased the amount of spam across platforms such as SMS, email, URLs, and social media. These spam messages often contain phishing links, fake offers, and harmful advertisements that threaten user security. To address this issue, the project proposes a Machine Learning-based Spam Classification System that can automatically detect and filter spam from various communication sources. The system begins by cleaning and preprocessing text data through steps like tokenization, stop-word removal, and normalization. It then uses TF-IDF to convert textual information into numerical features. Multiple machine learning models are trained to accurately distinguish between spam and legitimate messages. The system learns underlying patterns that help identify spam more effectively. Its performance is evaluated using metrics such as accuracy, precision, recall, and F1-score. The proposed solution minimizes false detections and improves reliability. It is capable of processing large volumes of data and can be easily integrated into real-time applications. Overall, the system strengthens communication security and builds user trust.

Keywords: Spam Classification, Machine Learning, SMS Spam Detection, Email Spam Filtering, URL Analysis, Social Media Spam, Text Classification, TF-IDF, Cyber Security.

Abstract: The rapid adoption of electric vehicles (EVs) has intensified the demand for efficient, safe, and user-friendly charging technologies. Conventional plug-in charging methods introduce limitations such as long charging times, physical connector wear, and range anxiety. This paper presents an advanced dynamic wireless charging system for electric vehicles based on inductive power transfer principles. The proposed system enables contactless energy transfer between road-embedded transmitter coils and vehicle-mounted receiver coils while the vehicle is stationary or in motion. The system integrates power electronics, resonant coupling, sensing units, microcontroller-based control, and IoT monitoring to enhance charging efficiency and automation. Experimental evaluation demonstrates reliable wireless power transfer, effective vehicle detection, and real-time system monitoring. The proposed approach supports the development of sustainable and intelligent transportation infrastructure.

Keywords: Electric vehicles, wireless power transfer, inductive charging, dynamic charging, IoT-based monitoring

Abstract: Effective coordination of the repair activities and inventory management is important towards increasing service efficiency. The methodical repair activities performed by a typical repair center are only record-based and lack integration or automation. This results in a significant expenditure of time by the technical and managerial personnel to manually follow up on the status of the ongoing repairs, along with inventory discrepancies that cause setbacks. This article proposes the design of the Repair Orders Inventory Agent software system. The software provides a modular structure made up of a database service and a central service. The central service has a repair orders service and a repair service. The software employs a service-oriented architecture. It uses the SQLite database. The software has been developed in Python. It is a software platform used in the repair of goods. It provides automated validation as well as repair lifecycle management. It has capabilities of real-time inventory update. In fact, experimental analysis proves that the Repair Orders Inventory Agent contributes towards increased accuracy, error reduction, as well as increased transparency levels of repair and inventory management. One of the significant areas that this proposed system showcases is related to real application uses of backend modularity in servicing con Keywords Repair Order Management, Inventory Synchronization, Service Automation, Python Backend Systems, SQLite Database, Intelligent Repair Systems

Abstract: With the rapid growth of digital infrastructure and network-based services, organizations are increasingly vulnerable to cyber threats such as unauthorized access, malware injection, and denial-of-service attacks. Traditional security mechanisms often fail to provide real-time detection and continuous monitoring of network activities. This paper presents CyberFlux, a intrusion detection and monitoring system designed to identify malicious network behavior and provide timely alerts to system administrators. The proposed system analyzes network traffic patterns and system logs to detect anomalies and classify potential intrusions. CyberFlux integrates a centralized monitoring dashboard that enables real-time visualization of security events and intrusion reports. The system is implemented using a scalable backend architecture and evaluated under simulated attack scenarios to validate its effectiveness. Experimental results demonstrate improved detection accuracy and faster response times compared to conventional rule-based systems. The proposed solution is suitable for deployment in small- and medium-scale organizational environments to enhance cybersecurity resilience.

Keywords: Intrusion Detection, Cybersecurity, Machine Learning, Deep Learning, Network Security, Monitoring System.

Abstract: Academic performance monitoring plays a vital role in improving learning outcomes and supporting informed instructional decisions. Conventional grade management systems, such as spreadsheets and static digital registers, primarily focus on data storage and provide limited analytical insight or interactive feedback. As a result, educators spend considerable time interpreting academic records manually, while students receive minimal understanding of their performance trends. This paper presents GradeBoardAgent, an AI-driven academic performance monitoring system that integrates structured data management with conversational intelligence. The system enables educators and students to interact with academic records using both a visual dashboard and natural language queries. Built using FastAPI for backend services, SQLite for persistent storage, and a generative AI model for conversational interaction, the platform supports automated insights, performance visualization, and real-time academic queries. Experimental evaluation demonstrates that GradeBoardAgent improves accessibility, reduces manual workload, and enhances transparency in academic monitoring. The proposed system highlights the practical application of conversational AI in education and illustrates how intelligent dashboards can transform static academic data into actionable insights.

Keywords: Academic Performance Monitoring, Conversational AI, Learning Analytics, FastAPI, SQLite, Intelligent Education Systems

Abstract: Traditional car management system models are field-based, with complex interfaces that necessitate extensive training to master. The proposed project aims to design an intelligent car management system that integrates agentic artificial intelligence via Google's Gemini 2.0 Flash approach, FastAPI technology, and SQLite database services. Thus, this proposed model will allow natural language processing to undertake full CRUD tasks to implement commands such as "show available cars" autonomous execution as database queries without human interference Performance analysis indicates a data-entry time reduction of up to 60%, response times of less than a second, and accuracy of 95% with respect to interpretation of user inputs, sufficient to handle up to 50 concurrent users. The underlying technology, Agentic, relies on pattern recognition software that uses FC calls to manage carry-over effects in interpretation guaranteeing data integrity.

Keywords: Agentic AI, Natural language processing, Fleet Management, Conversational AI, Database Automation, Function-Calling

Abstract: Effective management of daily schedules has become increasingly challenging due to the growing number of personal, academic, and professional responsibilities. Traditional calendar applications depend heavily on manual input and provide limited intelligent assistance. This paper presents a Personal Calendar Agent, an AI assisted conversational scheduling system that enables users to manage calendar events using natural language interaction. The proposed system allows users to create, update, delete, and retrieve events through simple conversational commands. A Gemini 2.0 AI model is used for intent recognition and information extraction, while a Fast API-based backend performs scheduling logic and event processing. Event data is persistently stored using an SQLite database. Additional features such as automated reminders, conflict detection, and category-based filtering improve scheduling efficiency. Experimental evaluation shows that the system reduces manual effort, improves usability, and enhances accuracy compared to traditional calendar tools.

Keywords: Personal Calendar Agent, Conversational AI, Intelligent Scheduling, Natural Language Processing, FastAPI, SQLite.

Abstract: Managing properties is a job. It means taking care of property records and talking to tenants. You also have to deal with maintenance requests and keep track of rent payments. People usually do these things by hand using phone calls or simple computer programs. This can cause problems, like delays and mistakes in the records. It also gives property owners and managers a lot work to do. Property management needs a system. This system should be easy to use and make things simpler for everyone involved in property management. Property management tasks should not be so hard. Property management can be made easier, with a system. This paper is, about a Property Service Agent that uses Agentic Artificial Intelligence. It helps people manage things related to their property by talking to it. The system lets tenants and property owners ask about their property check if it is available tell someone about problems that need to be fixed and update information using language. The Property Service Agent knows what the user wants. Does what is needed behind the scenes in a safe and controlled way. The Property Service Agent makes it easy for users to get things done. The system is designed using a modular architecture that includes a conversational interface, an agentic AI layer, backend services, and a database for storing property information. Experimental results show that the system improves ease of use, reduces manual effort, and provides timely responses compared to traditional property management approaches. The proposed solution demonstrates how conversational and agent-based AI can be effectively used to enhance property management systems.

Keywords: Property Service Agent, Agentic Artificial Intelligence, Property Management System, Conversational Interface, Intelligent Automation.

Abstract: Effective communication of daily academic content remains a major challenge in many schools despite the growing adoption of digital tools. Homework, notes, and announcements are still largely shared through verbal instructions, handwritten boards, or unstructured messaging platforms, leading to information loss, inconsistency, and increased workload for teachers. This paper presents the Digi School Agent, an intelligent, conversational AI-based academic content management system designed to address these challenges. The proposed system provides a centralized platform where teachers can upload, update, retrieve, and manage school content in a structured manner. Unlike traditional systems, Digi School Agent integrates a natural language conversational interface that allows users to interact with the system using simple commands. The platform also offers real-time analytics and auditing features to monitor academic activity, teacher engagement, and content distribution across classes and departments. Developed using FastAPI, SQLite, HTML, CSS, JavaScript, and an AI conversational framework, the system demonstrates improved accessibility, automation, and transparency in school communication. Experimental usage and testing confirm that the system effectively simplifies academic workflows while supporting data-driven decision-making in educational institutions.

Keywords: Conversational AI, School Content Management, Educational Automation, Academic Analytics, Intelligent Agent, Digital Education

Abstract: The increasing volume of global visa applications has made manual assessment slow, inconsistent, and prone to human error, highlighting the need for intelligent and automated decision-support systems. This project presents a scalable Visa Approval Status Prediction System built using machine learning and integrated with a complete MLOps pipeline for continuous training, automated deployment, real- time inference, and monitoring. The system analyzes applicant features such as demographics, academic history, financial stability, work experience, and documentation quality using multiple ML algorithms including Logistic Regression, Random Forest, XGBoost, and SVM. The MLOps workflow incorporates GitHub Actions for CI/CD automation, Docker for containerized deployment, and AWS (EC2, S3) for cloud hosting and model registry. After preprocessing steps such as feature engineering, data balancing, and normalization, the final model delivers high accuracy with strong precision-recall performance, making it suitable for real-world visa decision support. The system ensures reliability, scalability, and adaptability through continuous monitoring and automated retraining, demonstrating an efficient and production-ready approach to modernizing visa evaluation processes.

Keywords: Visa Approval Prediction, Machine Learning, MLOps, CI/CD, Docker, AWS, XGBoost

Abstract: In today's fast-paced digital world, people often miss relevant events due to information overload, poor discovery mechanisms, and complex registration processes. Traditional event management platforms require users to manually search, filter, and book events, which can be time-consuming and inefficient. To address these challenges, this paper presents People Meet Agent, an intelligent event discovery and management system powered by Agentic Artificial Intelligence. The proposed system acts as a smart assistant that understands user preferences through conversational interaction and autonomously performs tasks such as event discovery, recommendation, and booking. By using agentic AI principles, the system can take decisions, interact with backend services, and complete event-related tasks with minimal user effort. The platform supports personalized event suggestions, simplified registration, and real-time responses through a conversational interface. People Meet Agent improves user engagement by reducing manual steps and providing a seamless event experience. The system is especially useful for academic, professional, and social events, making event participation more accessible and efficient.

Keywords: Agentic AI, Event Management System, Conversational Agent, Event Recommendation, Artificial Intelligence.

Abstract: Festival event management involves complex coordination of multiple stakeholders, real-time information dissemination, and dynamic scheduling across diverse participants. This paper presents Festive Connect Agent, a novel agentic AI system that employs autonomous agents with natural language conversational interfaces to automate festival management workflows. The system leverages tool orchestration through dynamic function calling to manage CRUD operations, integrates performance-optimized microservices architecture for scalable deployment, and implements real-time event automation with social engagement features. The architecture employs the React framework through Google Agent Development Kit (Google ADK) for reasoning-driven task execution, natural language understanding for intent recognition with 87% accuracy, and hybrid recommendation engines achieving 82% precision in personalized suggestions. Evaluation against industry benchmarks (YCSB, SOP-Bench) demonstrates 1000 requests/second throughput. The system addresses the research gap of limited agentic AI applications in event management domains, providing a unified platform combining multi-turn dialogue, intelligent tool selection, audit logging with RBAC, and real-time notifications. This work demonstrates the feasibility and effectiveness of agentic AI architectures for automating complex real-world event management scenarios.

Keywords: Agentic AI, Festival Management, Tool Orchestration, Conversational Interface, Event Automation, Performance Metrics, Real-Time Systems

Abstract: Water conservation are important resources to human survival, development and is a daily responsibility which connects straight to human life, health, dignity etc. To protect environment and natural water resources here it uses the water system model that describes on how to preserve water, and to protect layer-1 of Earth which is called Hydrosphere and to meet future human demands. 97% on earth is of salt water in which 3% is suitable for drinking. The system is mainly introduced to protect or manage freshwater and it even reduces the wastage of water to avoid scarcity. In many parts of the world, people walk long distances to accumulate clean water. Factors like climate changes effects the water resources mainly on agriculture irrigation. It is not simply about saving water it is about valuing life, respecting nature, and taking collective steps to protect one of the planet's most irreplaceable resources. As lakes shrivel, rivers dry up, and groundwater levels fall, it becomes clear that water conservation is not an optional practice it is a crucial necessity for human survival and environmental strength. It is also strongly connected to social well-being. This project aims to survey water conservation not only as a subject but also as a human responsibility.

Keywords: Water conservation, rain water harvesting, agriculture irrigation, precipitation, evaporation Ground water.

Abstract: The increasing demand for refurbished electronic products has created challenges in product discovery, pricing transparency, and efficient management due to variations in condition and availability. Traditional e-commerce systems depend on manual search and filtering, which often leads to increased user effort and reduced clarity. This paper proposes Second Life, an agentic AI-based refurbished goods shopping system that enables conversational interaction for both product exploration and management. The system employs a language model to interpret user intent and invokes validated backend tools to perform database operations such as retrieving product details, checking availability, applying discounts, and updating or removing listings. A structured service-repository architecture is adopted to ensure secure execution and data consistency. The developed system demonstrates improved interaction efficiency, accurate data handling, and reduced manual intervention. The results indicate that agentic AI can effectively support refurbished e-commerce platforms by enhancing usability and automating operational workflows.

Keywords: Agentic AI, Refurbished Electronics, Conversational AI, Tool-Based Execution, ECommerce Automation, Database Management.

Abstract: The increasing complexity and volatility of modern financial markets have amplified the challenges faced by individual investors in making timely and informed trading decisions. While institutional participants benefit from advanced analytical infrastructures, retail investors often rely on fragmented tools that lack real-time intelligence, integration, and interpretability. TradeNexus AI - AI that Thinks Finance addresses this disparity by presenting an AI-driven decision support platform designed to assist investors through unified, explainable, and data-informed market insights. The system focuses on reducing information overload and emotional bias by transforming diverse financial data into structured, actionable guidance. TradeNexus AI integrates three complementary dimensions of market intelligence: technical analysis, fundamental analysis, and news-based sentiment analysis. Technical indicators such as RSI, MACD, and SMA capture short-term market momentum, while fundamental evaluation of financial ratios assesses long-term asset strength. In parallel, natural language processing models analyze financial news to quantify market sentiment. These heterogeneous signals are synthesized through a Weighted Fusion Decision Engine, which generates Buy, Sell, or Hold insights based on consensus logic, thereby enhancing decision reliability without relying on a single analytical perspective. The platform is implemented using a scalable web architecture comprising a responsive frontend, a modular backend, integrated AI processing components, and a PostgreSQL-based data storage layer for secure management of user and portfolio information. By combining real-time analytics, artificial intelligence, and an interactive user experience, TradeNexus AI demonstrates how intelligent decision-support systems can democratize access to advanced financial analysis tools, enabling individual investors to engage with financial markets in a more structured, confident, and informed manner.

Keywords: Artificial Intelligence, Algorithmic Trading Decision Support, Stock Market Analysis, Technical Analysis, Fundamental Analysis, Sentiment Analysis, Machine Learning, Natural Language Processing, Weighted Fusion Decision Engine

Abstract: Lantana is valued for its rich nutritional and powerful healing properties. It contains Beneficial compounds such as triterpenoid, flavonoids, phenolic acids, iridoid Glycosides, alkaloids, saponins, tannins, steroids, and essential oil components. These Compounds provide antioxidant, antimicrobial antifungal, anti-inflammatory Wound healing, and anticancer/cytotoxic activity this review highlights the key Phytochemicals, extraction methods like maceration and sonication improve the quality or bioactive compounds. While adding lantana leaves to the extraction process is Generally safe at moderate doses, very high amounts may cause side effect. Overall, Lantana camara holds great promise as a natural source of nutrition and medicine. With more research, standardization, and safe formulation, it can play a Major role in improving global health.

Keywords: Phytochemistry, Antioxidants Anti- inflammatory, wound healing Insecticidal agent, essential oils.

Abstract: The state of charge (SOC) of a battery is a key parameter for the safe and efficient operation of electric vehicles (EVs), as it directly affects driving range estimation, energy management, and battery protection. Accurate SOC estimation is challenging due to the nonlinear behavior of batteries, variations in operating conditions, temperature effects, and aging phenomena. This paper presents a comprehensive review of SOC monitoring and estimation techniques for electric vehicle applications. First, the fundamental concepts of battery SOC, key battery characteristics, and the role of SOC in battery management systems are discussed. Subsequently, a detailed review of conventional, model-based, and data-driven SOC estimation methods is provided, highlighting their underlying principles and practical applications. The performance of existing approaches is then compared in terms of estimation accuracy, robustness, computational complexity, and suitability under real-world driving and charging conditions. Key challenges and limitations associated with current SOC estimation techniques are identified. Finally, emerging research trends and future directions toward intelligent, adaptive, and real-time SOC estimation frameworks are outlined, followed by concluding remarks on the development of reliable SOC estimation strategies for next-generation electric vehicles.

Keywords: State of Charge (SOC), Electric Vehicles, Battery Management System (BMS), SOC Estimation, Lithium-Ion Batteries, Model-Based Estimation, Data-Driven Methods, Machine Learning, Battery Monitoring.

Abstract: The availability of low-cost genomic sequencing has created vast amounts of genomic data, which presents opportunities and challenges for the interpretation of genomic data in clinical and research environments. In this article, we describe a new software tool for the analysis of genomic data and disease prediction based on machine learning algorithms. The proposed tool applies several supervised learning algorithms to detect patterns between genomic markers and predict disease risk along with confidence measures. The software offers a wide variety of data visualization, model comparison, and feature importance analysis to aid in the interpretation of the results. Tests conducted on example datasets show the software's capacity to effectively identify significant genomic markers and classify disease status with acceptable accuracy. Furthermore, the software has also been implemented as an interactive web application on Google Collab, providing an immediate platform for researchers, educators, and clinicians to apply machine learning to genomic medicine without requiring extensive computational expertise. This research contributes to the emerging area of computational genomics by supplying an open-ended system for hypothesis formation and exploratory analysis in genomic studies.

Keywords: Machine learning, genomics, disease prediction, personalized medicine, feature importance, SNP analysis, bio informatics.

Abstract: Financial fraud in Chhatrapati Sambhajinagar has been evolving rapidly, shaped by digital technologies, shifting economic patterns, and growing urbanization. Traditional scams such as fake investment schemes and identity theft are now joined by cyber fraud, online banking manipulation, and digital payment scams. This changing nature of fraud not only threatens individual savings and business stability but also erodes public trust in financial institutions. The social impacts are significant: families face financial insecurity, small businesses struggle to recover from losses, and vulnerable groups-such as the elderly or less digitally literate-are disproportionately targeted. Beyond monetary damage, fraud creates psychological stress, weakens community confidence, and increases social inequality. Understanding these dynamics is crucial for designing stronger awareness programs, regulatory measures, and community based safeguards that can protect citizens while fostering a culture of financial resilience in the city.

Keywords: Financial Fraud, Online Fraud, Digital Financial Crimes, Economic Crime

Abstract: Mental health disorders such as depression and anxiety are often underdiagnosed because current assessments rely heavily on self-report questionnaires and clinical interviews, which are subjective, time-consuming, and difficult to scale. Recent studies show that eye movement behaviour- such as fixation patterns, saccade dynamics, and gaze allocation to emotional stimuli-can serve as objective digital biomarkers for mental health conditions [1], [2]. This paper presents VISION AND EMOTION, a real-time mental health assessment system that leverages eye-tracking data captured using a standard camera. The system records gaze trajectories while users interact with carefully designed visual tasks (emotionally valence images, reading tasks, and attention-switching trials) and extracts features such as fixation duration, saccade amplitude, blink rate, and gaze distribution across regions of interest. These features are used to train a machine learning classifier (Support Vector Machine) to distinguish between Normal and At-Risk mental health states. The proposed framework is lightweight, non-invasive, and deployable on commodity hardware without dedicated infrared eye trackers. Experimental evaluation demonstrates that the system can achieve promising classification performance with low latency, enabling near real-time feedback suitable for preliminary mental health screening. By combining eye-tracking analytics with machine learning, the system contributes toward scalable, objective, and cost- effective digital mental health tools that can complement traditional clinical assessments [2].

Keywords: Eye Tracking, Mental Health Assessment, Depression Detection, Gaze Analysis, Digital Biomarkers, Machine Learning, Real-Time Monitoring.

Abstract: Digital transformation has become an essential driver of competitiveness and growth for small businesses in the modern economy. It involves the integration of digital technologies-such as cloud computing, artificial intelligence, e-commerce platforms, and data analytics-into every aspect of business operations. This transformation not only enhances efficiency but also fuels innovation by enabling new business models and customer experiences. However, small businesses often face obstacles such as financial limitations, lack of technical expertise, and resistance to change. This paper examines how small businesses can adopt digital transformation as a strategic tool to foster innovation, improve performance, and remain competitive in an increasingly digital marketplace. The study also explores the barriers to implementation and provides recommendations to overcome them.

Abstract: The rapid digital transformation of business environments has fundamentally altered industry expectations from management graduates. Organizations today demand professionals who are digitally literate, analytically skilled, adaptable, and capable of integrating technology with managerial decision-making. However, a persistent gap exists between industry expectations and the academic curriculum of management education in India. This study aims to examine how digital pedagogy can act as a bridging mechanism between academic management curricula and industry requirements. Using primary data collected from a survey of 235 management students in Rajasthan and early-career professionals, the study investigates perceptions of digital teaching tools, curriculum relevance, skill preparedness, and employability outcomes. A structured questionnaire was employed, and statistical tools such as descriptive statistics, reliability analysis, correlation, and hypothesis testing were applied. The findings indicate a significant positive relationship between digital pedagogy adoption and perceived industry readiness. The paper concludes with industry-oriented suggestions for curriculum redesign, faculty development, and academia-industry collaboration.

Keywords: Digital Pedagogy, Management Education, Industry Expectations, Curriculum Alignment, Employability

Abstract: The rapid proliferation of Internet of Things (IoT) devices has intensified the demand for efficient spectrum utilization, making traditional static spectrum allocation insufficient. Cognitive Radio (CR) technology emerges as a promising solution by enabling dynamic spectrum access through intelligent spectrum sensing and adaptive decision-making. This review paper presents a comprehensive analysis of intelligent spectrum sensing and data fusion techniques in Cognitive Radio-enabled IoT networks. It systematically examines conventional spectrum sensing approaches, including energy detection, matched filtering, and cyclostationary detection, highlighting their limitations in noisy, heterogeneous, and large-scale IoT environments. To address these challenges, the paper explores machine learning and deep learning-based spectrum sensing methods that enhance detection accuracy, robustness, and adaptability. Furthermore, the role of data fusion is critically reviewed, focusing on data-level, feature-level, and decision-level fusion strategies that improve sensing reliability by combining observations from multiple IoT nodes. Intelligent data fusion techniques based on neural networks, fuzzy logic, and reinforcement learning are also discussed, emphasizing their capability to reduce uncertainty and communication overhead. The integration of spectrum sensing and data fusion within edge and fog computing paradigms is analyzed to support real-time and energy-efficient IoT applications. Finally, the paper identifies open research challenges related to scalability, security, latency, and standardization, and outlines future research directions toward 6G-enabled cognitive IoT systems. This review aims to serve as a valuable reference for researchers and practitioners working on intelligent spectrum management in next-generation IoT networks

Keywords: Cognitive Radio, Spectrum Sensing, Data Fusion, Internet of Things, Machine Learning, Cooperative Sensing, Dynamic Spectrum Access.

Abstract: Due to the quick digital transformation of the banking and financial services sector, financial data is now larger, more sensitive, and easier to find. Conventional centralized machine learning solutions pose serious privacy, security, and regulatory issues to businesses since they need gathering user data in one place. Federated Learning (FL) is a novel method that enables users to train models collectively without exchanging raw data. The PCI-DSS, GDPR, and RBI privacy laws are all met by this. In addition to incorporating features like safe aggregation, homomorphic encryption, differential privacy, and blockchain-based auditability, this paper provides a full federated learning system for financial analytics that protects privacy. The study illustrates how FL might reduce the risks associated with central data storage, hence enhancing financial forecasts, risk modelling, fraud detection, and credit rating. A thorough experimental analysis is presented to compare FL to conventional centralized approaches on important performance metrics as computation load, accuracy, privacy protection, and communication efficiency. The results demonstrate that FL maintains model performance competitiveness while significantly improving privacy and regulatory compliance. Additionally, in distributed financial contexts, the suggested blockchain-based auditability layer guarantees the permanence of transparency, verifiability, and recording. The essay also covers potential difficulties, how to apply the concepts, and the most effective methods for handling actual financial systems. Our study concludes by demonstrating the great potential of Federated Learning as a safe, scalable, and legally permissible alternative for next-generation financial analytics.

Keywords: blockchain auditability, risk modelling, homomorphic encryption, secure aggregation, distributed machine learning, federated learning, credit scoring, financial fraud detection, and privacy-preserving analytics.

Abstract: Sleep disruption and circadian misalignment are increasingly recognised as important public health concerns due to their established associations with impaired daytime functioning, metabolic dysregulation, and adverse mental health outcomes. Although these disturbances have been widely investigated in urban and occupational populations, empirical evidence from rural Himalayan settings remains limited. The present community-based cross-sectional study aimed to estimate the prevalence of sleep disruption and circadian misalignment among adults in a rural Himalayan village and to examine their association with lifestyle-related factors, with particular emphasis on evening tea consumption as the principal source of caffeine and pre-sleep electronic device use. The study was conducted among 100 adults (≥18 years) residing in Gamru village, District Kangra, Himachal Pradesh, India, using a structured questionnaire that assessed sleep duration, weekday-weekend variation in sleep timing (social jetlag), electronic device use before bedtime, evening tea intake, and sleep-related disturbances. Descriptive statistics were generated, and associations were examined using chi-square or Fisher's exact tests, with statistical significance set at p

Keywords: Sleep disruption; Circadian misalignment; Social jetlag; Tea consumption; Sleep-wake cycle; Rural health; Lifestyle factors

Abstract: This paper investigates the optimal control of parallel-server queueing systems operating under heavy traffic conditions. The study formulates the system as a Quasi-Birth-Death (QBD) process and applies the Matrix Geometric Method (MGM) to obtain steady-state probabilities and performance measures. Heavy traffic analysis is incorporated to approximate system behavior when utilization approaches unity, where congestion effects dominate and performance deteriorates rapidly. The research embeds control mechanisms including service rate adjustment, admission control, rejection penalties, and server breakdown considerations into the queueing framework. Both scalar and matrix formulations are examined, including breakdown repair dynamics that require solving nonlinear matrix equations numerically. Cost functions incorporating holding, service, and rejection penalties are developed, and numerical results demonstrate significant cost reductions through optimal service rate selection and controlled admission policies. The study highlights that heavy traffic approximations often push optimal solutions toward boundary controls unless nonlinear cost structures are introduced. Overall, the results reveal the economic trade-off between congestion, service capacity, and rejection penalties, providing valuable managerial insights for designing efficient service systems near capacity.

Keywords: Parallel-server, heavy traffic, matrix geometric method, quasi-birth-death process, optimal control, server breakdown, cost optimization