Authors: Mr.C.Radhakrishnan, Nijuram

 

Abstract: Cognitive Sleep Modulation through Generative AI and Real-Time Multi-Sensor Fusion introduces an intelligent, adaptive framework designed to improve sleep quality using advanced artificial intelligence techniques. The system gathers multi-modal physiological data—including electroencephalography (EEG), heart rate variability (HRV), respiratory signals, and body movement—from wearable and IoT-enabled devices. A real-time sensor fusion mechanism integrates these heterogeneous data streams and applies deep learning models to accurately classify sleep stages and detect disruptions. Based on the identified physiological state, generative AI algorithms produce personalized audio guidance, calming soundscapes, and cognitive relaxation prompts tailored to individual neural patterns. The framework dynamically adjusts environmental conditions such as lighting, sound, and temperature to facilitate smooth transitions across sleep cycles. Reinforcement learning strategies continuously optimize interventions by learning from long-term sleep efficiency metrics and user feedback. Experimental evaluations indicate reduced sleep onset latency, prolonged deep sleep phases, and improved sleep consistency. This intelligent, non-invasive solution demonstrates strong potential for personalized sleep enhancement and contributes to advancements in digital healthcare, cognitive science, and AI-driven wellness systems.

DOI: https://doi.org/10.5281/zenodo.19184531

 

Published by: vikaspatanker

Authors: P. Sunil, G. Swapna

Abstract: Quantum machine learning has emerged as a promising approach to enhance computational efficiency by leveraging the principles of quantum computing. However, the limitations of current Noisy Intermediate-Scale Quantum (NISQ) devices, such as noise, limited qubit availability, and circuit depth constraints, restrict the implementation of fully quantum models. To address these challenges, this study focuses on the design, implementation, and performance evaluation of hybrid quantum-classical machine learning (HQML) models. The proposed approach integrates parameterized quantum circuits with classical optimization techniques to enable efficient learning within NISQ environments. The study employs standard benchmark datasets, including Iris, Breast Cancer, and MNIST, to evaluate the performance of the hybrid model. The results indicate that the HQML model achieves competitive accuracy on small and medium-sized datasets while maintaining balanced precision, recall, and F1-score. However, performance declines for complex datasets due to hardware limitations and noise effects. Additionally, the hybrid model demonstrates a lower number of parameters compared to classical deep learning models but requires higher training time due to iterative quantum-classical optimization. The findings highlight that hybrid quantum-classical models provide a practical and scalable approach for utilizing quantum computing in the current technological landscape. Although challenges related to noise, scalability, and computational overhead persist, advancements in quantum hardware and algorithm design are expected to improve performance. This study contributes to the growing field of quantum machine learning by providing a systematic framework for evaluating hybrid models on NISQ devices and identifying key areas for future research.

Published by: vikaspatanker

Authors: Danish Tiwari, Mr. Rheetham Menon

Abstract: The evolution of cloud computing has significantly influenced modern software architecture, driving a shift from traditional monolithic systems to microservices-based designs. Monolithic architectures, while simpler to develop initially, often face challenges related to scalability, maintainability, and deployment flexibility. In contrast, microservices architecture enables the decomposition of applications into loosely coupled, independently deployable services, enhancing scalability, resilience, and continuous delivery capabilities. This study explores the key trends, benefits, and challenges associated with migrating from monolithic systems to microservices in the cloud era. It examines architectural patterns, containerization technologies, and orchestration tools that facilitate this transition. Additionally, the research highlights critical considerations such as service communication, data management, security, and DevOps integration. Real-world industry practices and case-based insights are analyzed to understand the practical implications of migration strategies. The findings suggest that while microservices offer significant advantages in terms of agility and scalability, successful adoption requires careful planning, robust infrastructure, and organizational readiness. The study concludes that microservices, when effectively implemented in cloud environments, play a crucial role in enabling digital transformation and supporting modern, scalable applications.

DOI: https://doi.org/10.5281/zenodo.19183798

Published by: vikaspatanker

Authors: Vinay Kumar Reddy Vangoor

 

Abstract: Enterprise Linux environments face persistent challenges in configuration management at scale, including configuration drift, compliance violations, and the high manual overhead required to maintain consistent system states across large server fleets. Traditional Infrastructure-as-Code (IaC) tools such as Ansible, Puppet, and Chef provide automation frameworks but demand significant human expertise to author, validate, and evolve configuration playbooks, creating a critical bottleneck in operational efficiency. This paper presents the AI-Assisted Configuration Management Framework (AICMF), an end-to-end system that integrates large language models (LLMs) with existing IaC pipelines to automate playbook generation, semantic policy validation, and continuous configuration enforcement across enterprise Linux environments. The framework employs a fine-tuned transformer-based model augmented with retrieval-augmented generation (RAG) to interpret natural language configuration intents and produce syntactically correct, policy-compliant IaC artefacts. A continuous drift detection module performs real-time state reconciliation against defined baselines, triggering automated self-healing pipelines with tiered human-in-the-loop approval gates for risk-proportionate oversight. Experimental evaluation across a 1,000-node heterogeneous Linux testbed comprising RHEL 9, Ubuntu 22.04 LTS, and CentOS Stream 9 over a 12-week period demonstrates a playbook accuracy rate of 94.3%, a 78.6% reduction in mean-time-to-remediate compared to manual baselines, a drift detection latency of 47 seconds with a 3.8% false positive rate, and a CIS Level 2 benchmark compliance rate of 91.3%. These results establish that AI-assisted IaC substantially reduces operational overhead while improving system reliability, security posture, and auditability in enterprise Linux deployments.

DOI: https://doi.org/10.5281/zenodo.19183576

 

Published by: vikaspatanker

Authors: Mr. Pratik Kalukhe, Mr. Shriyash Korade, Mr. Ankit Kapure

Abstract: The sustainability and profitability of modern agriculture hinge critically on selecting the optimal crop for specific geographical and environmental conditions. Traditional crop selection methods often rely on generalized historical data or farmer intuition, failing to account for the maximum achievable yield potential. This limitation frequently leads to suboptimal land use and reduced profitability. he optimization of agricultural output requires selecting not just a suitable crop, but the highest-yielding crop for specific environmental conditions. Traditional methods of crop selection often lack the scientific depth to accurately forecast crop productivity, leading to suboptimal yields and resource mismanagement. This research proposes a Yield-Aware Crop Selection System Leveraging Machine Learning (ML) to address this gap. The system utilizes a robust classification model to perform the initial recommendation based on key soil parameters (N, P, K, pH) and climatic factors (temperature, humidity, rainfall). Comparative evaluation showed that the Random Forest algorithm delivered the highest accuracy for crop suitability, achieving 98.8%. This system is architecturally designed to integrate a subsequent yield prediction model (using regression analysis) to provide the expected output, thus enabling farmers to make a final, yield-optimized decision. The highly accurate selection phase lays a reliable foundation for maximizing profitability, promoting sustainable farming, and modernizing agricultural practices through data-driven insights. By integrating robust classification with precise yield regression, this system transforms crop selection from a suitability problem into an optimization problem. This approach offers farmers an effective tool for boosting agricultural output, improving resource efficiency, and enhancing economic viability.

DOI: https://doi.org/10.5281/zenodo.19183397

Published by: vikaspatanker

Authors: Mukesh Brijanand Yadav, Prof. Ankush Dhamal

Abstract: Maintaining an optimal body weight is a fundamental aspect of personal healthcare management, as it significantly influences overall well-being, disease prevention, and quality of life. However, many individuals face confusion due to contradictory information available online, lack of personalized guidance, and the limitations of generic weight charts and traditional formulas that fail to account for individual variations and complex interactions between demographic factors. This research proposes an AI-Powered Ideal Weight Prediction System Using Multivariate Regression Analysis designed to assist individuals in identifying their ideal body weight based on key anthropometric parameters including height, age, and gender. The proposed system utilizes machine learning algorithms to analyze user data collected through interactive input interfaces. Features such as height measurements (in centimeters), age demographics (18-100 years), and gender classifications (Male/Female) are used as input parameters for multivariate regression analysis. Multiple regression algorithms including Random Forest Regressor, Decision Tree Regressor, Support Vector Regression, and Linear Regression were implemented and compared to identify the optimal model for weight prediction. The system is trained and evaluated using a comprehensive synthetically generated dataset (n=2000 samples) incorporating realistic biological variations and age-based metabolic adjustments, with ideal weight values calculated using modified Devine formulas enhanced through multivariate analysis techniques. The performance of the models is assessed using standard evaluation metrics including R-squared (R²), Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE) . Experimental results demonstrate that the Random Forest Regressor with 100 estimators achieves superior prediction accuracy compared to other algorithms, effectively capturing complex non-linear relationships between demographic features and ideal weight that conventional univariate methods cannot represent. The multivariate regression approach enables the model to simultaneously analyze interactions between all three input parameters, resulting in more nuanced and personalized predictions.

 

 

Published by: vikaspatanker

Authors: Mr. Sanket P. Datir, Mr. Swaraj A. Kale, Mr. Sumit M. Bahakar, Mr. Vipin V. Thorat, Prof. Ravindra R. Solanke

Abstract: Road accidents caused by over-speeding are a major problem, especially in areas like school zones, hospitals, and residential areas. To improve road safety, an automatic vehicle speed control system using Radio Frequency (RF) technology is proposed. In this system, an RF transmitter is installed in restricted zones and an RF receiver is placed in the vehicle. When the vehicle enters the restricted area, the transmitter sends a signal that is received by the vehicle’s receiver. The microcontroller processes this signal and automatically limits the vehicle speed. When the vehicle exits the restricted zone, the system restores the normal speed. This system helps reduce accidents and improves safety in sensitive areas.

Published by: vikaspatanker

Authors: Ayush Chettri, Aakansh Rai, Ashish Sunar, Asish Shakya

Abstract: This paper presents the design and implementation of a web-oriented Learning Management System (LMS) that aims to improve academic management in an institute. The system integrates course management, role-based access control, and real-time attendance tracking using modern web technologies including React.js, Node.js, and PostgreSQL. A modular three-tier architecture is adopted to ensure scalability and maintainability. The system is evaluated through functional testing and user feedback, demonstrating improved efficiency, accuracy, and usability compared to traditional manual methods. The proposed LMS reduces administrative workload, enhances communication, and provides a structured digital learning environment, making it suitable for deployment in academic institutions.

DOI: https://doi.org/10.5281/zenodo.19177834

 

Published by: vikaspatanker

Authors: Bommisetty Srihari, K Balasubrahmanyam, Mareddy Sai Kotireddy, Dr. U. Saravanakumar, Mr. E. Vinoth Kumar

 

Abstract: Recent advances in smartphones and affordable open-source hardware platforms have enabled the development of low-cost architectures for Internet-of-Things (IoT)-enabled home automation and security systems. These systems usually consist of sensing and actuating layer that is made up of sensors such as passive infrared sensors, also known as motion sensors; temperature sensors; smoke sensors, and web cameras for security surveillance. These sensors, smart electrical appliances, and other IoT devices connect to the Internet through a home gateway. This paper lays out an architecture for a cost-effective smart door sensor that will inform a user through an Android application, of door open events in a house or office environment. The proposed architecture uses an Arduino-UNO board along with the API. Several programming languages are used in the implementation and further applications of the door sensor are discussed as well as some of its shortcomings such as possible interference from other radio frequency devices.

DOI:

 

Published by: vikaspatanker

Authors: Pratika Yadav

Abstract: Artificial Intelligence (AI) has become a revolutionary force in credit evaluation and SME (small and medium enterprises) financing in the quickly changing financial ecosystem. The underlying creditworthiness of SMEs is frequently overlooked by conventional credit evaluation techniques, which mostly rely on financial statements and collateral. This study contrasts traditional credit evaluation methods with AI-driven financial tools to see how they affect SME credit choices. For the study, a descriptive and quantitative research design was chosen. A structured questionnaire disseminated via Google Forms was used to gather primary data from 56 respondents. Awareness of AI tools, perceived effectiveness in evaluating credit risk, decision accuracy, transparency, processing speed, and confidence in AI-based lending systems were all evaluated by the questionnaire. Reliability testing, graphical depiction, mean score interpretation, and percentage analysis were used to assess the gathered data. The results show that AI-driven financial tools greatly improve decision consistency, shorten loan processing times, and increase the accuracy of credit risk assessments. However, due to worries about algorithm transparency, data privacy, and technology infrastructure, adoption rates are still moderate. Though it presently serves as a decision-support tool rather than a whole substitute for conventional techniques, AI-based credit evaluation is generally having a favorable impact on SME funding.

 

 

Published by: vikaspatanker