Journal of Science Engineering Technology and Management Science

Automatic washing machines have become an essential household appliance because they simplify the clothes washing process, reduce manual effort, minimize water consumption, and improve washing efficiency. Conventional washing machines generally use microcontroller-based controllers or electromechanical timers to manage washing operations such as water filling, washing, rinsing, and spinning. However, these traditional controllers often experience limited processing speed, reduced flexibility, increased maintenance complexity, and slower response during sequential control operations. Recent advancements in Digital System Design, Field Programmable Gate Arrays (FPGA), Verilog Hardware Description Language (HDL), and Finite State Machine (FSM) architectures have enabled the development of high-speed hardware controllers capable of performing automatic washing operations with greater reliability and deterministic execution. This project presents the Design of an Automatic Washing Control System Using Verilog HDL. The proposed controller is implemented using Verilog HDL on an FPGA platform, integrating a water level sensing module, door lock controller, wash cycle controller, rinse controller, spin controller, timer module, motor control unit, LCD display interface, and Finite State Machine (FSM) into a unified hardware architecture. Initially, the controller receives the selected washing mode and continuously monitors the water level and door status before initiating the washing process. The FSM sequentially controls water filling, detergent mixing, washing, rinsing, spinning, and cycle completion while ensuring safe operation throughout the washing cycle. The system continuously supervises water level, motor operation, and timer status to maintain reliable washing performance. Experimental evaluation demonstrates high washing cycle accuracy, reliable state transitions, low processing delay, efficient FPGA resource utilization, and stable hardware operation under different washing conditions. The proposed FPGA-based controller significantly improves washing automation, reduces controller complexity, minimizes execution time, enhances operational reliability, and provides a scalable and costeffective solution for next-generation intelligent washing machine control systems.

Keywords : Verilog HDL, FPGA, Automatic Washing Machine, Finite State Machine (FSM), Digital Controller, Embedded Systems, Hardware Description Language, Washing Control System, Motor Control, Digital Logic.

Author : 1ARELLA SANJAY, 2Mr.M.ANIL, 3DESHAVANTH MADHU, 4KONGA PAVAN SUDHAN, 5BOMMAGONI SWETHA

Title : FPGA-BASED AUTOMATIC WASHING MACHINE CONTROL SYSTEM USING VERILOG HDL

Volume/Issue : 2026;03(08)

Page No : 587-595

The increasing demand for sustainable road construction and the rapid accumulation of plastic and rubber waste have become major environmental challenges worldwide. Conventional bituminous pavements require large quantities of natural materials and petroleum-based binders, resulting in increased construction costs and environmental concerns. The utilization of waste plastic and waste rubber as modifiers in bitumen offers an effective solution for improving pavement performance while simultaneously addressing waste disposal problems. Plastic and rubber modified bitumen enhances the engineering properties of conventional bitumen by increasing its stability, flexibility, resistance to deformation, and durability under varying traffic and climatic conditions. This project presents the Design of Bitumen Modified with Plastic/Rubber Road by investigating the suitability of waste plastic and crumb rubber as bitumen modifiers for flexible pavement construction. The study focuses on evaluating the physical and mechanical properties of modified bitumen using laboratory tests such as penetration test, softening point test, ductility test, viscosity test, specific gravity test, and Marshall Stability test. Different percentages of waste plastic and crumb rubber are blended with conventional bitumen to determine the optimum modifier content that provides maximum pavement performance. The experimental results demonstrate that plastic/rubber modified bitumen significantly improves Marshall Stability, load-bearing capacity, rutting resistance, fatigue resistance, moisture resistance, and service life compared with conventional bitumen. The modified pavement also exhibits lower maintenance requirements and improved resistance to high-temperature deformation and cracking. The proposed methodology provides highway engineers with an economical and environmentally sustainable approach for road construction by converting waste plastic and rubber materials into valuable pavement resources while reducing environmental pollution and improving the durability of flexible pavements.

Keywords : Plastic Modified Bitumen, Rubber Modified Bitumen, Waste Plastic, Crumb Rubber, Marshall Stability, Flexible Pavement, Sustainable Road Construction, Bituminous Mix, Highway Engineering, Pavement Performance, Recycling Technology, Green Roads.

Author : 1VASHNAVI, 2DR. P.ROHIT, 3P. RAKESH, 4S .AJAY

Title : DESIGN AND PERFORMANCE EVALUATION OF PLASTIC AND RUBBER MODIFIED BITUMINOUS PAVEMENT

Volume/Issue : 2026;03(08)

Page No : 596-604

Road transportation plays a significant role in the economic development of a country by facilitating the efficient movement of people and goods. The increasing traffic volume, higher axle loads, and rapid urbanization have created the need for durable and cost-effective pavement systems capable of providing long-term serviceability under varying environmental and traffic conditions. Flexible pavements are widely adopted due to their ease of construction, lower initial cost, and efficient load distribution characteristics. However, improper pavement thickness design may result in premature failures such as rutting, fatigue cracking, surface deformation, and excessive maintenance costs. Therefore, accurate pavement design using advanced analytical tools has become essential for ensuring pavement performance and durability. This project presents the Design of a 4-Lane Flexible Pavement Using IIT PAVE Software and IRC 37. The study focuses on designing a flexible pavement structure based on the guidelines specified in IRC:37-2018 by considering design traffic, axle load repetitions, subgrade strength, climatic conditions, and pavement material properties. The pavement structure is analysed using IIT PAVE, a mechanistic–empirical pavement analysis software developed by the Indian Institute of Technology. The software computes critical pavement responses including horizontal tensile strain at the bottom of the bituminous layer, vertical compressive strain on the top of the subgrade, surface deflection, and stress distribution under wheel loading. These responses are evaluated to determine the required thickness of bituminous layers, granular base, granular sub-base, and subgrade. The results demonstrate that the designed pavement satisfies the fatigue and rutting criteria recommended by IRC standards while ensuring structural stability, durability, and long service life. The proposed methodology provides highway engineers with an accurate and economical approach for designing four-lane flexible pavements capable of accommodating increasing traffic demand while minimizing maintenance requirements and improving pavement performance.

Keywords : Flexible Pavement, IIT PAVE, IRC 37:2018, Mechanistic-Empirical Design, Four-Lane Highway, Pavement Thickness Design, Bituminous Pavement, Fatigue Analysis, Rutting Analysis, Highway Engineering, Pavement Performance, Transportation Engineering.

Author : 1NANDHINI, 2Mrs.V.KAVITHA, 3K. ARUN NAYAK, 4K. BALAJI

Title : DESIGN AND STRUCTURAL ANALYSIS OF A FOUR-LANE FLEXIBLE PAVEMENT USING IITPAVE AND IRC 37

Volume/Issue : 2026;03(08)

Page No : 605-613

Efficient thermal management is a critical aspect of modern aircraft engine design, as excessive temperatures can reduce engine efficiency, accelerate component degradation, and compromise operational safety. This project presents the modelling and Computational Fluid Dynamics (CFD) analysis of aircraft engine fan cooling to investigate airflow characteristics and heat dissipation within the fan section. A three-dimensional model of the engine fan assembly is developed using computer-aided design (CAD) software and imported into a CFD environment for numerical simulation. The analysis evaluates airflow velocity, pressure distribution, turbulence intensity, and temperature variations under different operating conditions. Appropriate boundary conditions, material properties, and turbulence models are applied to replicate realistic engine cooling scenarios. The simulation identifies regions of high thermal concentration and airflow recirculation that may reduce cooling efficiency. Based on the obtained results, design modifications such as optimized blade geometry, improved airflow channels, and enhanced cooling passage configurations are assessed to improve thermal performance. The CFD outcomes demonstrate that optimized airflow significantly enhances heat transfer, minimizes temperature gradients, and promotes uniform cooling across critical engine components. The study also highlights the importance of numerical simulation in reducing prototype development time, lowering experimental costs, and supporting design optimization during the early stages of aircraft engine development. Overall, the proposed CFD-based approach provides valuable insights into the thermal behaviour of aircraft engine fans and offers an effective methodology for improving cooling efficiency, engine reliability, and operational lifespan. The findings contribute to the advancement of aerospace thermal management systems by enabling engineers to design safer, lighter, and more energy-efficient aircraft propulsion systems while maintaining high performance under demanding flight conditions.

Keywords : Aircraft Engine, Engine Fan Cooling, Computational Fluid Dynamics (CFD), Thermal Analysis, Heat Transfer, Airflow Distribution, Temperature Distribution, Turbulence Modelling, Cooling Performance, Aerospace Engineering.

Author : 1N. SWETHAK, 2Dr.A.ANITHA, 3B. DINESH, 4T. VIGNESHWAR , 5D. MAHESH, 6B. ANJANEYULU

Title : DESIGN, MODELING, AND CFD ANALYSIS OF AN AIRCRAFT ENGINE COOLING FAN

Volume/Issue : 2026;03(08)

Page No : 614-622

The camshaft is one of the most critical components of an internal combustion engine, as it controls the opening and closing of intake and exhaust valves, directly influencing engine performance, fuel efficiency, and emission characteristics. This project focuses on the modelling and manufacturing of a six-cylinder engine camshaft by integrating modern Computer-Aided Design (CAD) techniques with precision manufacturing processes. Initially, a three-dimensional (3D) model of the camshaft is developed using CAD software based on standard engine specifications, considering parameters such as cam profile, shaft dimensions, lobe spacing, journal diameter, and timing sequence. The digital model is evaluated to ensure dimensional accuracy and proper alignment of all cam lobes required for a six-cylinder engine configuration. Following the modelling phase, suitable engineering materials are selected based on mechanical strength, wear resistance, hardness, and fatigue performance. The manufacturing process includes machining operations such as turning, milling, grinding, and precision finishing to produce the required camshaft geometry. Heat treatment and surface finishing techniques are also incorporated to improve hardness, durability, and service life under high-speed operating conditions. Dimensional inspection and quality verification are performed to ensure compliance with design specifications and manufacturing standards. The developed camshaft is expected to provide accurate valve timing, reduced friction, improved mechanical reliability, and enhanced engine efficiency. The proposed methodology demonstrates the importance of integrating CAD modelling with advanced manufacturing technologies to produce high-quality engine components while reducing production time and manufacturing cost. This study contributes to the development of reliable and precisionengineered camshafts suitable for modern six-cylinder automotive and industrial engine applications.

Keywords : Camshaft, Six-Cylinder Engine, Computer-Aided Design (CAD), Manufacturing, Valve Timing, Precision Machining, Heat Treatment, Surface Finishing, Mechanical Design, Internal Combustion Engine.

Author : 1M.RAHUL, 2S.VENKAT LAKSHMAN, 3A.UMESH CHANDRA, 4K.VINAY RAM KUMAR, 5G.MALLIKARJUNA , 6M. VENKATA SAI RUTHVIK

Title : DESIGN, MODELING, AND MANUFACTURING OF A SIX-CYLINDER ENGINE CAMSHAFT

Volume/Issue : 2026;03(08)

Page No : 623-632

Digital identity is now an integral part of the digital ecosystem and is essential for safe access to the Internet, communication, financial systems, health, and e-government services. As digital technology has expanded, so have the opportunities for identity theft, fake accounts, phishing, deepfakes, impersonation and other forms of identity-based cyber-attacks and cannot prevent these new types of attacks with traditional security methods and require intelligent and adaptive detection. Artificial Intelligence (AI) has come to the rescue with the possibility of automated identity verification, behavioral analysis, biometric authentication, anomaly detection and real-time threat identification. This study presents a comprehensive literature analysis on intelligent detection of harmful digital identities using artificial intelligence. Topics covered include digital identity basics, several kinds of harmful digital identities, methods for verifying digital identities, deep learning and machine learning, large language models, generative AI, and intrusion detection systems based on artificial intelligence. According to the reviewed literature, AI has a major influence on improving detection accuracy, automation, scalability, and resilience to sophisticated cyberattacks. The results underscore the importance of AI in creating trustworthy, secure, and privacy-protecting digital identity management systems, as well as potential avenues for further research and development that could improve intelligent cybersecurity solutions

Keywords : Artificial Intelligence, Malicious Digital Identities, Machine Learning, Deep Learning, Identity Fraud Detection

Author : Dr. Sunita Dixit

Title : Intelligent Detection of Malicious Digital Identities Using Artificial intelligence (Ai)-Driven Approach

Volume/Issue : 2026;03(08)

Page No : 633-639

The creation of future intelligent applications relies on distributed, community-driven, and decentralized resource sharing, made possible by open edge networks, a new paradigm in computing. With edge computing, resources like processing, storage, and networking are brought closer to end devices. This improves scalability, real-time responsiveness, and decreases latency and bandwidth usage, all while decreasing reliance on centralized cloud infrastructures. With an emphasis on their function in constructing collaborative, open, and resilient edge ecosystems, this survey offers a thorough evaluation of community-driven and distributed computing designs. Architectural models and enabling technologies such as artificial intelligence, the Internet of Things, software-defined networking, network function virtualization, and multi-access edge computing are covered in the article, which serves as an introduction to the basics of edge computing. Open edge networks are further investigated in the study by looking at their uses in smart cities, healthcare, transportation, robotics, smart homes, robots in agriculture, smart roads, and disaster management. In addition, it reviews key security, privacy, and trust mechanisms required for secure distributed environments and highlights challenges related to interoperability, heterogeneous resource management, scalability, and cyber resilience. The study concludes by outlining future objectives and identifying existing research needs in the pursuit of intelligent, autonomous, secure, and sustainable open edge network architectures. These designs should be able to serve varied real-time applications across heterogeneous computing environments.

Keywords : Open Edge Networks, Distributed Computing Architectures, Artificial Intelligence (AI), Internet of Things (IoT)

Author : Dr Chintal Kumar Patel

Title : Toward Open Edge Networks: A Survey of Community-Driven and Distributed Computing Architectures

Volume/Issue : 2026;03(08)

Page No : 640-647

This study presents a comparative analysis of working capital management practices in five selected banks in India — State Bank of India, HDFC Bank, ICICI Bank, Axis Bank, and Federal Bank — over the five-year period 2021–2025. The study is descriptive and analytical in nature and is based on secondary data drawn from the published annual reports, profit and loss accounts, balance sheets, and other secondary financial sources of the selected banks. Key indicators — Earnings per Share, Book Value per Share, Net Profit Margin, Return on Equity, and Return on Assets — are used to evaluate the liquidity, profitability, and overall financial performance of the selected banks for the terminal year of the study period, and the banks are ranked on the basis of their overall comparative performance. The results show that ICICI Bank recorded the strongest profitability in the terminal year, HDFC Bank recorded the highest absolute earnings and net worth per share together with the most stable growth pattern, and State Bank of India recorded the sharpest overall improvement in profitability and efficiency during the period, while Axis Bank displayed comparatively higher earnings volatility despite healthy growth in book value. The study concludes that a combination of profitability, consistency, and growth indicators provides a more complete basis for evaluating the working capital management performance of Indian banks than any single ratio considered in isolation, and that continued attention to cost efficiency, income diversification, and liquidity discipline remains central to sustaining financial performance in the banking sector.

Keywords : Working Capital Management, Liquidity, Profitability, Indian Banks, Ratio Analysis, Net Profit Margin, Return on Assets.

Author : Roopa.BK, Syed Salaudin

Title : A STUDY ON WORKING CAPITAL MANAGEMENT ON SELECTED BANKS IN INDIA

Volume/Issue : 2026;03(08)

Page No : 648-659

This study presents a comparative analysis of the risk and return performance of ten selected NIFTY 50 companies — Reliance Industries, HDFC Bank, Bharti Airtel, ICICI Bank, State Bank of India, Tata Consultancy Services, Infosys, Kotak Mahindra Bank, Hindustan Unilever Limited, and Larsen & Toubro — over the five-year period 2021–2026. The study is descriptive and comparative in nature and is based on secondary data drawn from monthly stock price data obtained from NSE India, Money control, and Yahoo Finance. Key indicators — Average Monthly Return, Total Holding Period Return, Annualized Return, Monthly Risk (Standard Deviation), and Annualized Risk — are used to evaluate the return-generating capacity and volatility of the selected stocks, and the companies are ranked on the basis of their overall comparative risk-return performance. The results show that Bharti Airtel, Larsen & Toubro, and Tata Consultancy Services recorded the strongest overall performance, combining high annualized returns with a favourable return-to-risk ratio, while Infosys and HDFC Bank recorded negative annualized returns despite bearing considerable price volatility. The study concludes that a combination of average return, annualized return, and standard deviation provides a more complete basis for evaluating equity performance than any single indicator considered in isolation, and that investors should assess both return potential and risk exposure before making portfolio decisions in the Indian equity market.

Keywords : Risk, Return, NIFTY 50, Standard Deviation, Annualized Return, Equity Investment, Indian Stock Market, Portfolio Management.

Author : B Sindhu, Dr, Mallesha L

Title : RISK AND RETURN ANALYSIS OF NIFTY 50

Volume/Issue : 2026;03(08)

Page No : 660-675

This study explores the impact of Supplier Relationship Management (SRM) on achieving and sustaining competitive advantage at Viswas Foods, a reputed food processing enterprise in South India.In an industrywhereoperationalexcellenceandsupplierreliabilityarecrucial, SRM practices determine the firm’s agility and resilience. A mixed-methods approach was adopted to gather both quantitative and qualitative data from 30 suppliers through structured surveys and semi-structured interviews. The research applied weighted sum analysis, factor analysis, and Pearson correlation techniques to examine the effect of SRM components on competitive advantage. Key findingsindicatethatlong-term, trust-based supplier relationships (Weighted Sum:130, Grade A) are the most influential in enhancing competitive advantage, showing a strong positive correlation (r = 0.72, p < 0.01). Transparent communication (Weighted Sum: 128, Grade B) and effective performance monitoring (Weighted Sum:126, Grade B)arealsovital contributors. On the other hand, supplier contract management and risk mitigation strategies received a Grade C, indicating areas of weakness that need urgent attention. Factor analysis extracted three core SRM dimensions—Trust (explaining 41.86% of variance), Communication (12.00%), and Performance Monitoring (10.79%). Demographic insights reveal that 36.7% of respondents are aged 36–45, a segment with potentially strong business maturity. However, only 13.4% of suppliers had long-term relationships exceeding six years, reflecting gaps in supplier retention efforts. The study proposes strategies such as loyalty incentives, clarity in contractual terms, and the integration of digital collaboration tools to improve SRM and ensure long-term competitiveness.

Keywords : Supplier Relationship Management, Competitive Advantage, Trust, Communication, Supply Chain Stability

Author : Ashin M Shaji, Dr.Suma Mathew

Title : IMPROVING SUPPLIER RELATIONSHIP AS A COMPETITIVE EDGE IN MANUFACTURING: A STUDY ON VISWAS FOODS

Volume/Issue : 2026;03(08)

Page No : 676-680

Maintaining the security and dependability of rail transportation systems depends on the prompt and precise detection of foreign items on railway lines. By combining a two-stage architecture based on YOLOv8 and Overhaul Knowledge Distillation (OKD), this study offers an improved foreign object intrusion detection framework that solves the drawbacks of current approaches, specifically low efficiency and suboptimal accuracy. To lessen the computational load on detection models, a lightweight image classification model quickly filters railway photos to find those that might include foreign items. The YOLOv8 object detector precisely locates and identifies the foreign objects in images that have been marked as suspicious. YOLOv8 provides notable improvements over its predecessors in terms of inference speed and detection accuracy. The Overhaul Knowledge Distillation technique is used to further improve the classification stage's performance, enabling the lightweight classifier to learn from a more intricate teacher network and attain competitive accuracy with increased efficiency. The suggested method establishes a new state-of-the-art in railway foreign item detection by outperforming current solutions in both speed and robustness, according to experimental evaluations.

Keywords :

Author : Kulsum Begum, Dr. C. Berin Jones

Title : EFFICIENT RAILWAY FOREIGN OBJECT DETECTION USING YOLOV8

Volume/Issue : 2026;03(08)

Page No : 681-686

In this work, we extend the recently proposed Quantum Vision (QV) theory in deep learning for object recognition by integrating it with the Xception architecture, forming a novel Heavy QV-Xception model. The QV theory, inspired by the particle-wave duality in quantum physics, treats objects as information waves rather than static images, enabling deep neural networks to capture richer representations. Building on this concept, our Heavy QVXception model leverages a robust QV block to transform conventional images into wave-function representations and processes them through the depthwise separable convolutional layers of Xception for enhanced feature extraction. This hybrid approach benefits from both the quantum-inspired information representation and the efficient, high-performance architecture of Xception. Extensive experiments on multiple benchmark datasets demonstrate that the Heavy QV-Xception model consistently outperforms standard Xception and other conventional CNNs, highlighting the effectiveness of combining QV theory with advanced deep learning architectures for improved object recognition accuracy

Keywords :

Author : Sabiya Begum, Lubna Nausheen, Ruqaiya Fatima

Title : COMBINING DEEP LEARNING AND QUANTUM VISION THEORY TO IMPROVE OBJECT RECOGNITION

Volume/Issue : 2026;03(08)

Page No : 687-693

For prompt management and better patient outcomes, early and precise diagnosis of Alzheimer's disease (AD), especially the shift from Cognitively Normal (CN) to Mild Cognitive Impairment (MCI), is essential. This study creates an improved model based on the Xception architecture, building on previous deep learning techniques that use convolutional neural networks (CNNs) with channel attention mechanisms. Both local and global neuroimaging features are captured by the suggested system using sophisticated feature extraction techniques, which are then combined via a learning fusion process. With a binary classification accuracy of 99% for CN versus MCI individuals, our model performs noticeably better than earlier approaches. This enhancement highlights the Xception network's capacity to identify subtle imaging indicators linked to early cognitive deterioration. The system seeks to offer a dependable, accessible diagnostic tool to assist medical professionals in the early detection of Alzheimer's disease, enabling prompt and focused treatment approaches.

Keywords :

Author : Maryam Misba, Afshan Fatima, Ruqiya Fatima

Title : NEUROIMAGING-BASED MCI PREDICTION USING DEEP LEARNING

Volume/Issue : 2026;03(08)

Page No : 694-701

The crucial need for effective thyroid nodule categorisation and early diagnosis is highlighted by the rising incidence of thyroid cancer. By speeding up the diagnostic process, automated systems can greatly benefit doctors. However, the difficulty of feature extraction and the scarcity of medical imaging databases make this objective difficult to achieve. By focusing on the extraction of significant traits that are crucial for tumour diagnosis, this work tackles these issues. By incorporating cutting-edge feature extraction techniques, the suggested method improves the capacity to identify thyroid nodules in ultrasound pictures. The classification framework identifies several suspicious classifications in addition to differentiating between benign and malignant nodules. The combined classifiers offer a thorough description of thyroid nodules and show encouraging accuracy in initial assessments. Thyroid nodule categorisation techniques have advanced significantly as a result of these findings. This study offers a novel strategy that may provide useful assistance in clinical settings, enabling a quicker and more precise identification of thyroid cancer.

Keywords :

Author : Kurma Varalaxmi, Dr. C. Berin Jones

Title : AN OPTIMIZED XCEPTION-BASED FRAMEWORK FOR COMPUTER-AIDED THYROID NODULE CLASSIFICATION USING ULTRASOUND IMAGES

Volume/Issue : 2026;03(08)

Page No : 702-707

The increasing frequency and sophistication of cyber-attacks have highlighted the need for accurate and efficient network intrusion detection systems. Conventional signature-based detection techniques often struggle to identify evolving and previously unseen attacks, making machine learning a promising solution for modern cybersecurity applications. This paper presents a CatBoost-based network intrusion detection framework for classifying network traffic into normal and malicious categories. The proposed approach utilizes the KDD Cup 1999 dataset, where data preprocessing and feature selection techniques are applied to improve model efficiency and reduce computational complexity. The performance of the proposed model is evaluated against widely used machine learning algorithms, including Decision Tree and Random Forest. Experimental results demonstrate that the CatBoost classifier provides superior classification performance due to its ordered boosting strategy, efficient handling of categorical features, and improved generalization capability. To demonstrate the practical applicability of the proposed framework, a Flaskbased web application is developed for user authentication, network traffic prediction, and performance visualization. The proposed framework offers an efficient, scalable, and reliable solution for network intrusion detection and can contribute to strengthening cybersecurity through intelligent machine learning techniques.

Keywords :

Author : Tahniyat Fatima, Dr. I. Samuel Peter James

Title : ENHANCING CYBERSECURITY THROUGH MACHINE LEARNING: A CATBOOST-BASED FRAMEWORK FOR NETWORK INTRUSION DETECTION

Volume/Issue : 2026;03(08)

Page No : 708-713

Mental health issues including anxiety, stress, and depression are becoming more common among college students, especially those who are under pressure from their studies and careers. Early and precise detection of depressed tendencies can facilitate prompt intervention and psychological assistance. In this work, we use structured survey data to propose a Cat Boost-based machine learning model for predicting depression levels among college students. Age, gender, academic year, CGPA, and university background are among the demographic variables included in the dataset, along with psychological markers obtained from the PHQ-9 survey. A cleaned dataset of 2,028 samples with six categories of depression—No Depression, Minimal, Mild, Moderate, Moderately Severe, and Severe Depression—was used to train and assess the model. Because it can handle categorical variables natively and capture intricate non-linear correlations without requiring a lot of preprocessing, Cat Boost was chosen. The experimental findings showed an accuracy of more than 92%, surpassing conventional machine learning baselines and attaining performance on par with or superior to deep learning systems like FE-BiON. Extensive analyses, such as confusion matrix evaluation and feature importance visualization, verify that PHQ-9-related items have a considerable impact on prediction outcomes. The suggested technique provides a dependable and comprehensible framework for automated depression screening, assisting academic institutions and mental health specialists in fostering the wellbeing of students.

Keywords :

Author : Arshiya Fathima, Dr. C. Berin Jones

Title : MENTAL WELLNESS MONITORING SYSTEM FOR ENTREPRENEURIAL STUDENTS

Volume/Issue : 2026;03(08)

Page No : 714-720

The use of electric vehicles (EVs) is growing quickly, which calls for dependable and effective battery management systems, particularly in battery swapping infrastructures. In order to improve the effectiveness of EV battery swapping systems, this project offers a machine learning-driven web service for real-time battery health estimation. The system forecasts two crucial aspects of battery condition: State of Health (SoH) and remaining charge cycles. It was developed using Flask as the backend web framework and Python for data processing and machine learning. The application uses two potent ensemble learning algorithms, Random Forest Regression and XGBoost, which are trained on historical battery usage data, including charge/discharge current, voltage, temperature, and cycle counts, to produce accurate forecasts. The system allows for quick decision-making at battery changing stations by processing user input in real time and displaying the battery's health condition via an intuitive interface. By lowering the likelihood of premature battery disposal, this solution supports sustainable energy practices in addition to encouraging proactive maintenance and effective use of EV batteries. The application is scalable, effective, and appropriate for integration into actual EV infrastructure thanks to the combination of machine learning and a lightweight Flask-based deployment.

Keywords :

Author : Khadija Maryam, Dr. I. Samuel Peter James

Title : REAL-TIME MACHINE LEARNING FRAMEWORK FOR BATTERY HEALTH ESTIMATION IN EV BATTERY SWAPPING

Volume/Issue : 2026;03(08)

Page No : 721-726

Millions of people use digital platforms like social media to convey their psychological states, making mental health illnesses an increasing global concern. The emotional depth, colloquial language, metaphorical idioms, and contextspecific clues that are frequently included in such posts make it difficult to identify mental illness from textual data. These intricacies are sometimes difficult for general language models and traditional machine learning techniques to grasp, which results in low prediction accuracy. This paper offers a strong and clever framework for multiclass mental disease prediction using cutting-edge deep learning and natural language processing (NLP) approaches to address these issues. The suggested method incorporates domain-adapted transformer models with deep neural networks to enhance comprehension and categorisation of phrases connected to mental health. In particular, rich contextual elements are extracted from psychologically sensitive material using MentalBERT, a BERT version pretrained on mental health corpora. In addition, people with emotional discomfort frequently utilise metaphorical and symbolic language, which is interpreted by MelBERT, a metaphor-aware model. Convolutional Neural Networks (CNNs) for hierarchical feature extraction and a Bidirectional Long Short-Term Memory (BiLSTM) network for capturing longrange semantic dependencies from the sequence's past and future contexts augment these transformer-based models. The system can anticipate several classes thanks to this hybrid architectureof mental illness—such as bipolar disorder, PTSD, anxiety, despair, and more—with increased precision and dependability. A labelled dataset of social media posts was used to train and assess the model, and performance measures showed that it performed better than conventional techniques. The results of this study emphasise the significance of sequential deep learning approaches, figurative language interpretation, and domain-specific modelling in developing an efficient and scalable mental health detection system. By offering a tool that can help professionals with early identification and intervention tactics using digital linguistic analysis, this effort advances the larger objective of mentalhealthawareness.

Keywords :

Author : Ayesha Ahmed, Dr. I. Samuel Peter James

Title : A COMPREHENSIVE FRAMEWORK FOR MULTICLASS MENTAL HEALTH PREDICTION USING LSTM AND NLP

Volume/Issue : 2026;03(08)

Page No : 727-735

Approximate computing is an efficient approach for error-tolerant applications because it can trade off accuracy. Addition is a key fundamental function for these applications. In this project, proposing accuracyconfigurable adder that also maintains a small design area. The proposed adder is based on the approximation of least significant bits, and its configurability of accuracy is realized by with reversible logic gates. Compared with the conventional carry look-ahead adder using maskable half adder the proposed experimental result demonstrates the achievement of the original purpose of optimizing area without reducing the accuracy.

Keywords :

Author : Fahmeena Mahween, Tazeen

Title : DESIGN AND IMPLEMENTATION OF A LOW-POWER HIGH-SPEED CONFIGURABLE ADDER FOR APPROXIMATE COMPUTING

Volume/Issue : 2026;03(08)

Page No : 736-744

Every multi-story building structure that is subjected to ground motion in daily life has to have its behaviour studied since it is a common issue for development. The structure's foundation is trembling due to the earthquake. Buildings oscillate as a result of these vibrations, which might seriously harm the structure. These vibrations generated at the ground level are transmitted to the top of the building due to the bulk of the structure, which creates lateral stresses on the frame and ultimately lowers the moment resistance capacity of building components including columns, beams, and other structural elements. The paper provides an overview of several studies done on multistory buildings while taking various factors into consideration. Reaction spectrum approach is beneficial for determining structure response in specific circumstances of ground vibrations, according to all prior studies, observations, and conclusions. This paper presents the findings of an investigation into the seismic response of the G+15 structure both with and without a damper. Practically all multi-story structures must be studied as threedimensional systems, according to the current edition of the IS: 1893-2002. Buildings' floor plans may be seen as asymmetrical. India's hilly terrain is largely seismically active. In the current work, response spectrum analysis is used to compare friction damper systems, viscous dampers, and mass tuned dampers. Results for the G+15 building are compared in terms of drift, shear, bending, torsion, time period, and frequency values.

Keywords : Earthquakes are natural phenomena, which cause the ground to shake.

Author : KOLLI SAI KUMAR1 GOGULAMUDI MRUDULA2

Title : NON LINEAR TIME HISTORY ANALYSIS OF TALL BUILDING FOR SEISMIC LOAD USING DAMPERS

Volume/Issue : 2026;03(08)

Page No : 745-750

A column is supposed to be a vertical member starting from foundation level and transferring the load to the ground. The term floating column is also a vertical element which (due to architectural design/ site situation) at its lower level (termination Level) rests on a beam which is a horizontal member. The beams in turn transfer the load to other columns below it. There are many projects in which floating columns are adopted, especially above the ground floor, where transfer girders are employed, so that more open space is available in the ground floor. These open spaces may be required for assembly hall or parking purpose. Floating columns are a typical feature adopted in the modern age multi-storey building construction These floating columns reduces the efficiency of the building to resist seismic forces up to large extent, that is why these are the extremely undesirable unit in the seismic prone areas. In this study an attempt was made to analyze and design of G+10 story building by Staad Pro Software. In the present study a comparison was made between floating column and non floating column building. Results like story deflection of building, shear in X direction, Shear in Y direction, Bending in X direction, Bending in Y direction, Area of steel required for the floating column building and non floating column building is studied by using Response spectrum analysis.

Keywords : Floating column, Column, deflection, shear, Bending, Response spectrum analysis.

Author : SAI BHANU PRASAD1 ,M Thiyagarajan2

Title : SEISMIC ANALYSIS OF INTERNAL AND EXTERNAL FLOATING COLUMN BY ETABS SOFTWARE

Volume/Issue : 2026;03(08)

Page No : 751-756

Muscle Blaze fitness supplements, with particular reference to Bengaluru City. The study is descriptive in nature and is based on primary data collected through a structured questionnaire administered to 200 respondents who are active social media users and are aware of, or have purchased, fitness supplements, supplemented by secondary data drawn from company records, published articles, and online sources. The analysis covers the demographic profile of respondents, their social media usage pattern, and the influence of advertisements, personalisation, influencer endorsements, content quality, customer reviews, promotional offers, and brand trust on their purchase decisions towards Muscle Blaze. The results show that the sample is dominated by Gen Z and young Millennial consumers who are daily, Instagram-led social media users, and that social media advertisements, customer reviews, and influencer endorsements are the strongest drivers of awareness, trust, and purchase intention, while promotional offers and personalised marketing further reinforce buying behaviour. The study concludes that social media marketing exerts a consistent and largely positive influence on consumer purchasing behaviour towards Muscle Blaze fitness supplements, and that sustained investment in credible influencer partnerships, authentic customer content, and platform-appropriate promotional campaigns remains central to converting awareness into purchase and advocacy.

Keywords : Social Media Marketing, Consumer Purchasing Behavior, Muscle Blaze, Fitness Supplements, Influencer Marketing, Brand Trust, Consumer Engagement.

Author : Sai Kiran V R, Dr. Mahesh Kumar C A

Title : IMPACT OF SOCIAL MEDIA MARKETING ON CONSUMER PURCHASING BEHAVIOR TOWARDSMUSCLE BLAZE FITNESS SUPPLEMENTS: AN EMPIRICAL STUDY

Volume/Issue : 2026;03(08)

Page No : 757-772

This study examines the impact of social media marketing on the buying decisions of youth, with particular reference to Ballari city. The study is descriptive in nature and is based on primary data collected through a structured questionnaire administered to 100 young respondents who are active social media users, supplemented by secondary data drawn from journals, industry reports, and other published sources. The analysis covers the demographic profile of respondents, their social media usage pattern, and the influence of advertisements, influencer endorsements, discounts and promotional offers, content type, and peer recommendations on their purchasing decisions. The results show that the sample is dominated by postgraduate students in the 23-26 age group who are daily, Instagram-led social media users, and that advertisements, friends' recommendations, and influencer reviews are among the strongest drivers of attention, trust, and purchase behaviour, while discounts and promotional offers further reinforce buying decisions. The study concludes that social media marketing exerts a strong and largely positive influence on the purchasing decisions and lifestyle of youth, while also revealing a clear demand among young consumers for greater regulation to safeguard their interests, underscoring the need for brands to balance persuasive marketing with transparency and responsible advertising practices.

Keywords : Social Media Marketing, Youth, Buying Decision, Influencer Marketing, Digital Advertising, Consumer Behaviour, Brand Awareness.

Author : Shiva Kumar M, Dr. Mahesh Kumar C A

Title : A STUDY ON SOCIAL MEDIA MARKETING AND ITS IMPACT ON THE BUYING DECISION OF YOUTH

Volume/Issue : 2026;03(08)

Page No : 773-787

Electronic waste (e-waste) is among the fastest-growing waste streams globally, and India's tier-II cities are increasingly significant contributors as smartphone and gadget penetration deepens among young, digitally connected populations. This study sets out to explore the awareness, perceptions, and disposal practices of youth in Nashik, Maharashtra, using the Theory of Planned Behaviour (Ajzen, 1991) as the guiding framework, and to statistically examine whether awareness and perception meaningfully shape actual disposal behaviour. A structured, Likert-scale questionnaire was designed and administered, and the resulting data were planned for analysis in SPSS using reliability testing, Exploratory Factor Analysis, and Multiple Linear Regression, so that the underlying dimensions of youth e-waste behaviour and their relative influence on practice could be identified with statistical rigour rather than assumed from descriptive impressions alone. Going forward, this research is intended to serve several purposes. It is expected to generate a locally grounded, city-specific evidence base for Nashik a tier-II city largely absent from the published e-waste literature that can inform how local civic bodies, educational institutions, and policymakers design awareness and collection interventions. The statistical model developed here is meant to clarify whether interventions should prioritise factual awareness-building or the deeper work of shaping perception and personal relevance, guiding more efficient allocation of limited civic and institutional resources. Beyond its immediate policy relevance, the study is designed to extend the Theory of Planned Behaviour into an underexplored tier-II urban context, opening avenues for comparative future research across other mid-sized Indian cities, longitudinal tracking of behaviour change following awareness interventions, and deeper qualitative exploration of the psychological drivers that convert environmental concern into consistent, formal e-waste disposal practice.

Keywords : e-waste; awareness; perception; disposal practices; factor analysis; regression analysis; Nashik; youth

Author : Swarali Sampada Sachin Kulkarni, Dr. Tanuja Dubey

Title : Exploring Awareness, Perception, and Practices of E-Waste Disposal Among Youth in Nashik

Volume/Issue : 2026;03(08)

Page No : 788-797

Diabetes and its complications, particularly Diabetic Retinopathy (DR) and Diabetic Nephropathy (DN), pose significant challenges to global healthcare systems, requiring accurate predictive models for early diagnosis and intervention. Traditional approaches often underperform due to imbalanced datasets and complex feature interactions. Publicly available datasets, including structured diabetic clinical data and the APTOS 2019 retinal fundus images, were utilized. Preprocessing involved KNN imputer for missing values, outlier detection and handling, MinMax scaling, and SMOTE oversampling to balance the data. Multiple machine learning models were implemented for classification, including variants of Logistic Regression, Random Forest, XGBoost, LightGBM, CatBoost, Multi-Layer Perceptron, as well as hybrid ensemble approaches. Advanced models were further developed, comprising StackingClassifier, Ensemble-ofEnsembles, and image classification architectures including ResNet50, DenseNet121, Xception, NasNetLarge, and an ensemble of Xception + DenseNet121. Evaluation metrics included Accuracy, Precision, Recall, F1-Score, ROC-AUC, RMSE, and LogLoss. DenseNet121 achieved the highest classification performance with 99.6% Accuracy, while StackingClassifier Oversampled reached 99.9% Accuracy for nephropathy, and LightGBM OverSampled attained 99.6% Accuracy for retinopathy. Explainable AI techniques, including LIME, SHAP, and Grad-CAM, provided model interpretability, and a Flask framework-based interface enabled user-friendly prediction deployment, ensuring accurate, transparent, and actionable clinical decision support.

Keywords : Diabetes Mellitus, Diabetic Retinopathy, Diabetic Nephropathy, Machine Learning, Ensemble Models, Predictive Analytics.

Author : M. Thangavel, A.M. Kalpana, S. Ananth

Title : Predicting Diabetic Retinopathy and Nephropathy Complications Using Machine Learning Techniques

Volume/Issue : 2026;03(08)

Page No : 798-807