International Journal of Science, Engineering and TechnologyVoice Search Optimization As A Digital Marketing Strategy: Implications For Online Retail Performance
Authors: Dr. Rajidi Rammohan Reddy, Dr. Sayantani Chakraborty
Abstract: Voice search optimization has become one of the revolutionary approaches in digital marketing which has major consequences for the efficiency of online retailing. Given the rising significance of voice-activated purchases, which is likely to reach the market value of 0.3 billion in 2025, retailers need to implement innovative solutions to improve their digital marketing strategies. This paper presents the impact of VSO implementation on e-commerce performance indicators through a combination of quantitative and qualitative analyses of 2,847 websites of online retailers. It was discovered that application of AI-based technologies for voice search optimization cuts down the costs of creating content by 81%, increases the organic traffic by 156%, and raises the conversion rate by up to 34% in comparison with standard search traffic.
DOI: http://doi.org/10.5281/zenodo.22768724
Cyber Security: Protecting Digital Information in the Modern Era
Authors: Assistant Professor Dr. A. Akil Nivetha, A. Indhu Pavithra, M.Sabina, G.Gowri, Fathima Mirsha
Abstract: Cyber security has become one of the most important fields in the digital era due to the rapid growth of information technology and internet-based services. Individuals, businesses, educational institutions, healthcare organizations, and government agencies rely heavily on digital platforms to store, process, and transmit sensitive information. As the use of online services continues to increase, cyber criminals are developing more sophisticated methods to exploit security vulnerabilities. Cyberattacks such as malware, phishing, ransomware, data breaches, denial-of-service attacks, and identity theft have become major threats that affect millions of users worldwide. The primary objective of this research paper is to study the importance of cyber security, identify common cyber threats, analyze their impact, and examine modern security technologies that help protect digital information. The paper discusses various security mechanisms, including firewalls, antivirus software, encryption, intrusion detection systems (IDS), intrusion prevention systems (IPS), multi-factor authentication (MFA), and secure network protocols. These technologies play a significant role in protecting the confidentiality, integrity, and availability of information. This research also focuses on the role of Artificial Intelligence (AI) and Machine Learning (ML) in improving cyber security. AI-based systems can analyze network traffic, detect abnormal activities, identify malware, and respond to threats in real time. The study explains how intelligent security systems improve cyber defense and reduce the response time to security incidents. A layered cyber security approach is proposed in this paper to provide comprehensive protection against various cyberattacks. The proposed framework integrates authentication mechanisms, encryption techniques, firewall protection, intrusion detection systems, continuous monitoring, and user awareness programs. This integrated approach significantly reduces security risks and enhances digital trust. The findings of this research indicate that no single security technology is sufficient to prevent all cyberattacks. Organizations should adopt a combination of advanced technologies, regular software updates, security awareness training, and effective cyber security policies to build a secure digital environment. As digital transformation continues to grow, cyber security will remain a critical requirement for protecting information assets and ensuring safe online communication.
AI-Based Employee Healthcare and Well-Being Monitoring System with Personalized Break Recommendations
Authors: Assistant Professor R.Baby, S.Subashini, T.Kalpana, S.Jothipriyan, K.A.Athila, R.Arul Stenin
Abstract: Sitting for long periods without moving at computer workstations is now a common part of office jobs, and it is closely linked to higher stress, tiredness, muscle and bone problems, and increased risks for heart and metabolic health over time. Traditional workplace wellness programs often respond only after problems arise, are the same for everyone, and don’t consider how each employee actually works day to day, which makes them less effective. This paper introduces an AI-powered system that monitors employees’ health and well-being by tracking how they use their work systems, like how long they stay active, when they take breaks, and how they interact with tools. It also includes optional data from wearable devices and information employees share about how they feel, to help identify signs of tiredness and stress. The system uses these signs to create custom pop-up messages on the employee’s screen. These messages remind them to do easy health activities, like standing and stretching after sitting for a certain amount of time, or drinking water at set times for staying hydrated. A system that combines simple rules with machine learning adjusts when and how long breaks are given based on each worker’s usual habits, instead of using the same schedule for everyone. A feedback loop lets the system learn from suggestions that are accepted, skipped, or rejected, helping it decrease alert fatigue over time and make the alerts more relevant. The early check of the design shows that this system can help cut down on long periods of sitting, promote better drinking habits, and make people feel more positive, all without affecting how much work they can get done. The paper also talks about the system design, the method used in the algorithm, how it is better than current solutions, and what could be done next, like keeping data private while processing it on the device and making better use of wearable devices.
Eco-Adaptive Bioring for Flood and Drought Management in Agriculture: A Sustainable Root-Zone Water Management Approach
Authors: Assistant Professor R.Muthu Lakshmi, C.Janaki, R.Anitha, S.Arockiya Anusri, M.Muthumari, A.Prince Joshva
Abstract: Agriculture is increasingly affected by irregular rainfall patterns, prolonged droughts, intense rainfall events, surface runoff, and temporary flooding. Conventional agricultural water-management practices generally address excess water and water scarcity as separate problems, requiring drainage during wet conditions and irrigation during dry conditions. This study proposes an Eco-Adaptive BioRing as a sustainable, decentralized root-zone water management approach designed to address both conditions using a single localized system. The proposed BioRing is positioned around the crop root zone and is designed to facilitate controlled water collection, temporary retention, gradual infiltration, and excess-water drainage. During heavy rainfall, the system is intended to reduce prolonged water accumulation around the crop by directing surplus water through a controlled overflow pathway while allowing a useful portion of the incoming water to infiltrate into the surrounding soil. During dry periods, the retained soil moisture can gradually become available to the crop roots, potentially extending the duration of root-zone moisture availability and reducing unnecessary irrigation demand. The methodology includes sustainable material selection, BioRing fabrication, field installation, soil-moisture monitoring, flood-condition evaluation, drought-condition evaluation, water-balance assessment, and comparison with conventional root-zone management. The proposed approach integrates water conservation, localized infiltration, controlled drainage, and sustainable material utilization within a single agricultural intervention. The system has potential applications in rainfed farming, irrigated agriculture, drought-prone regions, and areas experiencing irregular rainfall and temporary flooding. The proposed Eco-Adaptive BioRing provides a promising low-complexity approach for improving root-zone water regulation and increasing the effective utilization of available agricultural water resources.
Design and Pso-Based Optimization of an Intelligent Autonomous Mobile Robot for Environmental Risk Detection and Disinfection
Authors: Okika Stephen Sunday, Kingsley I. Chibueze, C.O. Ugwoke, Emeka Augustine Chinachi, Onyeabo Uzoamaka Agatha, Malachy O. Ugwuoke
Abstract: The increasing threat of viral disease transmission in healthcare and public environments has created a need for automated, contactless and efficient environmental disinfection technologies. This study presents the design and optimization of an intelligent mobile robot for autonomous environmental-risk detection and disinfection, with particular consideration of resource-constrained environments in Nigeria. The proposed system integrates ultrasonic, infrared, gas and thermal sensors with autonomous navigation, proportional-integral-derivative (PID) motor control, sensor-fusion-based environmental assessment and a UV-C disinfection unit. Particle Swarm Optimization (PSO) was incorporated to minimize robot travel distance, disinfection time and energy consumption while maximizing environmental coverage. The robot was evaluated in a controlled 5 m × 5 m indoor test environment containing representative obstacles. Performance was assessed using coverage efficiency, disinfection time, energy consumption and environmental-risk detection accuracy. The PSO-optimized configuration achieved 92% coverage efficiency, compared with 72% for the non-optimized configuration. Disinfection time decreased from 35 to 25 min, representing a 28.6% reduction, while energy consumption decreased from 120 to 96 Wh, corresponding to a 20.0% reduction. Sensor fusion improved environmental-risk classification accuracy from 78% for a single-sensor configuration to 91%, representing a 13-percentage-point improvement. The findings demonstrate that combining sensor fusion, autonomous navigation and PSO-based path optimization can substantially improve the operational efficiency of mobile robotic disinfection. The proposed architecture provides a practical foundation for contactless environmental disinfection in hospitals, schools and other high-risk facilities, although biological validation and direct pathogen sensing remain necessary before clinical deployment.
DOI: http://doi.org/10.5281/zenodo.22793418
Comparative Analysis of Progressive Collapse Resisiting Capacity of Different Structural Systems
Authors: Muhammad Awais, Engr. Danish Saeed, Engr. Zaheer Ahmed
Abstract: Progressive collapse is a disproportionate structural failure in which the localized loss of a critical load-carrying element can cause redistribution of forces, excessive deformation, and subsequent failure of adjacent structural members. An 11-storey regular reinforced-concrete commercial building was developed and analysed using SAP2000 under identical geometric, material, loading, and boundary conditions to ensure a consistent comparison. The Alternate Path Method (APM), following the GSA and UFC approaches adopted in the thesis, was employed by considering three sudden column-removal scenarios: corner-column removal, edge/mid-column removal, and interior-column removal. The structural response was evaluated in terms of vertical displacement, inter-storey drift, axial-force redistribution, bending and shear response, and overall structural stability. The results demonstrated a clear difference among the three structural systems. For the MRF, vertical displacement increased with storey height and reached approximately 0.16 in, with the second column-removal scenario producing the highest displacement response. In comparison, the Diagrid and Outrigger systems showed substantially lower displacement ranges, with their maximum responses occurring in Scenario 3. The MRF also exhibited the highest inter-storey drift, with the plotted response reaching approximately 8 in, whereas the Diagrid system showed a substantially lower drift response of approximately 2.5 in, and the Outrigger system exhibited the lowest drift response, reaching approximately 1.4 in. Axial-force redistribution increased from the upper storeys toward the lower storeys in all three systems. The MRF and Diagrid responses approached approximately 600 kip at the lower levels, whereas the Outrigger system showed a comparatively lower maximum axial-force response of approximately 400 kip. Among the removal cases, Scenario 3 generally produced the comparatively higher drift and axial-force response. Overall, the Diagrid System demonstrated the most favorable progressive-collapse performance, primarily because its triangulated perimeter configuration provides greater stiffness, redundancy, and efficient alternate load paths. The Outrigger System also demonstrated strong performance, particularly in controlling displacement and inter-storey drift through effective load transfer between the central core and perimeter columns. In contrast, the MRF exhibited the least favorable response, characterized by relatively higher deformation and drift.
Embedded IoT-Based Smart Assistive Glove System For Real-Time Gesture Recognition And Emergency Alerts
Authors: Aneesh Jantikar, Amrutha M, Aishwarya M C, Harshith C
Abstract: Communication is a daily struggle for people with speech and hearing impairments whenever the person across from them does not know sign language, and that struggle turns serious in an emergency, when getting help depends on being understood right away. Vision-based recognition, wearable sensing, and sensor-fusion methods have each moved sign lan-guage interpretation forward, but every one runs into the same problems: heavy computation, poor portability, slow response, or privacy loss. This paper works through these gesture recognition approaches, weighing what each does well against where it falls short, and looks at how embedded machine learning and edge computing are changing what a wearable device can do. One gap keeps showing up across the literature: almost none of these communication-focused systems build in real emergency support, and the few that try rarely fit how a nonverbal or hearing-impaired person would want to trigger one. Building on that gap, this paper proposes EchoHand, an IoT-enabled glove pairing flex sensors, a six-axis IMU, and an ESP32 microcontroller to recognize gestures and produce speech or text output, while reusing that pipeline to send a location-aware alert when needed. It closes by covering the practical hurdles this design faces and where it goes next.
Comparative Assessment of Two Mix Proportions for Interlocking Concrete Tiles Targeting 30 MPa Characteristic Strength
Authors: C.O. Ugwoke, Emeka Augustine Chinachi, Onyeabo Uzoamaka Agatha, Okika Stephen Sunday, Kingsley I. Chibueze
Abstract: The performance of interlocking concrete tiles depends strongly on the proportions of cement, water, fine aggregate, and coarse aggregate used in their production. This study investigated two concrete mix proportions developed for interlocking tiles with a target characteristic compressive strength of 30 MPa. River sand in Zone 2 and 10 mm granite chippings were used as fine and coarse aggregates, respectively. The reported specific gravities of the fine and coarse aggregates were 2.66 and 2.71. Two mixes were designed using SCALE 149 software. Mix 1 had a reported proportion of 320:190:760:1195 and a water-to-cement ratio of 0.594, while Mix 2 had a proportion of 275:160:790:1240 and a water-to-cement ratio of 0.58, for cement, water, fine aggregate, and coarse aggregate, respectively. Five bone-shaped specimens were produced for each mix, demoulded after 24 h, cured for 28 days, and tested for compressive strength using a 200 kN compression testing machine. The reported compressive strengths for Mix 1 were 34.0, 37.7, 40.2, 39.6, and 30.7 MPa, giving a mean strength of 36.44 MPa. Recalculation from the five individual results gives a sample standard deviation of 4.02 MPa. Mix 2 produced strengths of 43.7, 34.1, 36.8, 40.2, and 37.4 MPa, with a mean of 38.44 MPa and a sample standard deviation of 3.65 MPa. Using the characteristic-strength approach reported in the source study, fck = f̄ − Ks with K = 1.64, the calculated characteristic strengths are 29.85 MPa for Mix 1 and 32.45 MPa for Mix 2. Thus, Mix 2 was the preferred mix among the two investigated proportions because it exceeded the 30 MPa target while using less cement than Mix 1. The findings provide a practical basis for selecting a mix proportion for interlocking concrete tiles, although the limited number of specimens warrants further validation with larger experimental datasets and additional durability tests.
CFD-Based Optimization of Nozzle Arrangement for Enhanced Evaporative Cooling Performance of a Mist Fan
Authors: Emeka Augustine Chinachi, Onyeabo Uzoamaka Agatha, Kingsley I. Chibueze, Malachy. O. Ugwuoke, Okika Stephen Sunday, C.O. Ugwoke
Abstract: The research investigated the cooling efficiency of a mist fan by redesigning its nozzle arrangement and determining the effects of nozzle arrangement on cooling performance, with emphasis on temperature distribution, velocity distribution and water-mass-fraction distribution. Five different nozzle arrangements were comparatively analyzed to determine the arrangement with the highest efficiency. A three-dimensional computational fluid dynamics (CFD) model based on a Lagrangian–Eulerian approach was used to simulate the interaction between the air and dispersed water droplets. The computational domain represented an enclosed room geometry measuring 7200 mm × 3600 mm × 3000 mm. A grid-sensitivity analysis was performed using three grids, and the reported temperature differences between successive grids were small. The five nozzle arrangements were compared at an air velocity of 5.96 m/s. The reported cooling efficiencies for arrangements 1, 2, 3, 4 and 5 were 66%, 67%, 68%, 70% and 63%, respectively, while the corresponding average temperatures were 304.2, 303.7, 303.6, 303.4 and 303.8 K. Nozzle arrangement 4 produced the highest reported cooling efficiency and the lowest average temperature. The temperature contours also indicated a more uniform distribution for arrangement 4. The further compared the CFD predictions with experimental measurements and reported strong agreement between the two approaches.
Spectrophotometric Analysis of Amino Acids and Proteins
Authors: Kenjaev A.Q., Kenjayeva S.N.
Abstract: In this work, the spectrophotometric properties of amino acids and proteins are scientifically studied. First, general concepts regarding the biological significance of amino acids and proteins, as well as the main methods used in their analysis, are provided. The theoretical foundations of spectrophotometry—the interaction between electromagnetic radiation and molecules, the Bouguer-Lambert-Beer law, the molar absorption coefficient, and the UV (ultraviolet) absorption properties of amino acids—are highlighted. Furthermore, the spectral characteristics of aromatic amino acids (tryptophan, tyrosine, phenylalanine), the spectrophotometric analysis of proteins, and methods for determining their concentrations are analyzed. In the practical section, Olympiad problems, along with the capabilities, challenges, and limitations of the spectrophotometric method, are presented. At the end of the work, general conclusions and recommendations for future research are provided.
Response Surface Optimization of Fermentation Conditions for Mesquite Seed (Prosopis Africana) Based on Nitrogen Evolution
Authors: Onyeabo Uzoamaka Agatha, Malachy. O. Ugwuoke, Okika Stephen Sunday, Kingsley I. Chibueze, C.O. Ugwoke, Emeka Augustine Chinachi
Abstract: Mesquite seed (Prosopis africana) is an underutilized tropical seed with nutritional and economic potential and is traditionally processed into fermented condiments. Because traditional fermentation is carried out under different local conditions, the progress and outcome of the process can vary. This study examined fermentation time, temperature and particle size as controllable factors and used response surface methodology (RSM) to relate them to nitrogen evolution. Matured mesquite seeds obtained from Afor-Opi, Nsukka Local Government Area, Enugu State, Nigeria, were boiled, decarped, oven-dried at 35°C, grated and sieved into particle sizes of 250–1200 µm. Fifty-gram portions were fermented at 25–45°C for up to 120 h, and nitrogen evolution was monitored at 24-h intervals. The RSM design produced 20 randomized experimental combinations for the mesquite seed. The resulting quadratic model was significant (F = 153.12, p < 0.0001) and accounted for 99.28% of the variation in nitrogen evolution (R² = 0.9928). The adjusted R² and predicted R² were 0.9863 and 0.9456, respectively, while adequate precision was 52.011. Fermentation time, temperature, particle size, the time–temperature interaction, the time–particle-size interaction and the quadratic time term were significant. Numerical optimization gave 113.6 h, 44.33°C and 313.33 µm as the best combination within the experimental region, with a predicted nitrogen evolution of 0.915%. The result indicates that the three operating variables can be used together to control the fermentation response. The predicted condition should, however, be confirmed experimentally before it is considered for scale-up.
High-Speed Low-Noise Dynamic Comparator for ADC Applications: Design and Simulation in 50 nm CMOS
Authors: Murari Kumar, Associate Professor Dr. Deepti Shinghal
Abstract: High-speed and low-noise comparators are essential building blocks of modern analog-to-digital converters (ADCs), where propagation delay, noise, offset voltage, power consumption, and circuit complexity directly influence the overall conversion performance. This paper presents the design and simulation of a high-speed, low-noise dynamic CMOS comparator intended for high-performance analog and mixed-signal applications. A comparative analysis of conventional dynamic latch-based, double-tail latch-based sense-amplifier, uncalibrated dynamic, and reconfigurable dynamic comparator architectures is performed using key performance parameters including propagation delay, operating speed, RMS noise, offset voltage, slew rate, and transistor count. The proposed architecture incorporates a differential amplifier stage within the regenerative latch and employs optimized clocking to improve signal integrity and decision speed. The design is evaluated using LTspice IV with a 50 nm CMOS technology model. Simulation results indicate a minimum propagation delay of 0.650 ns, corresponding to a maximum operating speed of 1.538 GHz. The proposed comparator also achieves an RMS noise of 704.38 µV while using 13 transistors. The results demonstrate a balanced trade-off between speed, noise performance, and circuit complexity. The proposed architecture can therefore serve as a suitable comparator core for high-speed ADCs and other low-power mixed-signal systems.
Smart Crop Prediction and Fertilizer Recommendation Based on ML and IoT
Authors: Associate Professor Dr.Manjula K, Ashwini M, Ramavathcharansainaik, Ramya S, Rakshitha S
Abstract: Agriculture depends on many factors such as soil nutrients, temperature, humidity, rainfall, soil type and crop characteristics. Selecting a suitable crop and applying the right fertilizer are important for obtaining good production. Traditionally, farmers mainly depend on previous experience, local knowledge and manual soil testing. These methods can be useful, but they may not provide the same result for different soil and weather conditions. In recent years, machine learning and Internet of Things technologies have been increasingly used to support agricultural decision-making. This literature review studies existing research related to crop prediction, fertilizer recommendation, machine learning, IoT-based monitoring and intelligent agricultural decision-support systems. Different machine learning algorithms such as Random Forest, Decision Tree, K-Nearest Neighbour, Support Vector Machine, XGBoost, CatBoost, Logistic Regression and ensemble methods have been applied in previous studies. IoT systems have also been developed using sensors and microcontrollers to collect soil and environmental information in real time. The reviewed studies show that machine learning can provide useful predictions when sufficient and relevant agricultural data are available, while IoT can help overcome the limitation of relying only on static datasets. However, several limitations are still present in existing systems. Some studies focus only on crop prediction, while others concentrate only on fertilizer recommendation. Many systems are evaluated using small or region-specific datasets, and real-time sensor integration is not always included. Explainability, fertilizer quantity optimization, climate variation and practical deployment also require further attention. Based on these observations, this review identifies the need for an integrated system that combines crop prediction, fertilizer recommendation, real-time IoT data and comparative machine learning evaluation. The proposed research direction is therefore focused on developing a practical smart farming system that can use soil and environmental parameters to predict a suitable crop and recommend an appropriate fertilizer.
Designing An Intelligent Chatbot: A Natural Language Processing Approach Using Deep Learning Models
Authors: Assistant Professor K.Petchiammal, M. Mathumathi, S. Ramesh, S.Karthikeyen, S.M. Nusra Fathima, K. Murugeshwari
Abstract: The growing demand for automated, human-like conversational systems has positioned Natural Language Processing (NLP) as a core enabler of intelligent chatbot design. This study presents a deep learning-based framework for building an intelligent chatbot capable of intent recognition, entity extraction, and context-aware response generation. Three NLP architectures are compared for the intent-classification and response-generation pipeline: a Bidirectional LSTM with attention, a fine-tuned BERT-based intent classifier, and a Transformer-based sequence-to-sequence model. A labelled conversational dataset is preprocessed through tokenization, stop-word handling, lemmatization, and word-embedding generation before model training. The models are evaluated using accuracy, precision, recall, F1-score, and BLEU score to assess both classification quality and generated-response fluency. Experimental results show that the BERT-based intent classifier achieves the highest intent-recognition accuracy, while the Transformer sequence-to-sequence model produces the most fluent and contextually relevant responses. The comparative study offers practical guidance for selecting NLP architectures suited to customer support, virtual assistance, and other conversational-AI applications, and contributes toward building chatbots that are more accurate, context-aware, and reliable.
A Study On Consumer Perception Towards Second-Hand Cars in The Indian Market
Authors: Dr. P. Manochithra
Abstract: The Indian second-hand car market has become an important segment of the automobile industry, influenced by affordability, brand reputation, vehicle condition, maintenance concerns, and changing consumer access to digital information. This study examines consumer perceptions towards second-hand cars in the Indian market, with particular attention to the reasons for purchasing used cars, the importance attached to brand reputation, concerns regarding repairs, satisfaction with second-hand cars, and experience of learning to drive using second-hand cars. The study uses the survey data reported in the original manuscript and presents the descriptive results in a consolidated and journal-oriented format. The reported responses indicate that vehicle condition, affordability, and brand reputation are important considerations in second-hand car purchase decisions. The findings also indicate that consumers attach importance to brand reputation and display varying levels of concern regarding repairs and satisfaction. The study further highlights the potential role of transparency, inspection and certification, digital platforms, and word-of-mouth communication in shaping consumer confidence. The results provide a basis for understanding selected dimensions of consumer perception in the Indian used-car market. However, the sampling procedure, study period, geographical coverage, questionnaire design, and inferential statistical procedures require further documentation before the manuscript can be considered a fully reproducible empirical study.
DOI: http://doi.org/10.5281/zenodo.22872088
A Role of Biotechnology in Climate Change Mitigation
Authors: Assistant Professor M. Kalavathi, I.Najitha, R.Lakshmi, G.Sivaram, V. Madhumitha, D. Weslin Immanuvel
Abstract: Climate change is one of the major global challenges, requiring innovative and sustainable approaches to reduce greenhouse gas emissions and protect the environment. Biotechnology plays an important role in climate change mitigation by using biological systems and processes to develop eco-friendly solutions. It contributes to carbon sequestration through the use of microorganisms, algae, and plants, while also supporting the production of biofuels such as bioethanol, biodiesel, and biogas as alternatives to fossil fuels. Biotechnology can improve agricultural productivity through climate-resilient crops, reduce the use of chemical fertilizers and pesticides, and promote sustainable waste management through biodegradation and bioconversion. Advances in microbial biotechnology and synthetic biology also offer opportunities for carbon dioxide capture and conversion into valuable products. Thus, biotechnology provides diverse strategies for reducing greenhouse gas emissions, conserving natural resources, and promoting a low-carbon and sustainable economy. Integrating biotechnology with renewable energy, agriculture, and environmental management can significantly contribute to global efforts to mitigate climate change.
Influence of Sound Frequencies on Plant Growth and Physiological Development
Authors: R. Baby, J. Jerlin, M. Veni, S. Divya, M. Arunatharan, A. Mohamed Esmail
Abstract: Plant growth and physiological development are influenced by a wide range of environmental stimuli, and recent research has increasingly focused on the role of acoustic energy as a non-invasive growth-modulating factor. Conventional agricultural practices rely primarily on light, water, nutrients, and temperature control, while the potential of sound frequency exposure remains largely underexplored in mainstream cultivation systems. This paper presents a systematic experimental framework for studying the Influence of Sound Frequencies on Plant Growth and Physiological Development, which examines how controlled acoustic stimuli across different frequency ranges affect germination rate, stem elongation, leaf area, chlorophyll content, and overall biomass accumulation. The proposed framework integrates a calibrated sound frequency generator, a controlled plant exposure chamber, and a network of growth-parameter sensors to capture physiological responses under repeatable experimental conditions. Statistical and machine-learning-based correlation analysis is applied to the collected data to identify frequency ranges that produce measurable and consistent effects on plant development. In addition, an intelligent recommendation module suggests optimal frequency and exposure duration for specific plant species based on observed growth trends. The framework is designed to minimize experimental variability, ensure reproducibility across trials, and support data-driven insights for sustainable and technology-assisted agriculture. Experimental evaluation demonstrates measurable variation in physiological parameters across frequency treatments, improved understanding of acoustic-plant interaction, and a practical pathway toward sound-assisted cultivation techniques, making the proposed framework a valuable contribution to smart and precision agriculture research.
DOI: http://doi.org/10.5281/zenodo.22873563
AI-Based Fake News Detection: A Review
Authors: R. Baby, S. Harini Anandhi, M. Jackulin, S. Kalpana, A. Raja Dharsan, A. Jegatheswaran
Abstract: The rapid growth of digital platforms and social media has transformed the way people create, access, and share information. Although online platforms provide fast access to news and public information, they have also contributed to the rapid spread of misleading and fabricated content. Fake news can influence public opinion, create confusion, damage reputations, and reduce trust in legitimate information sources. Traditional fact-checking methods often require significant human effort and may not be sufficient to process the large volume of information generated online. Artificial Intelligence (AI), Machine Learning (ML), Natural Language Processing (NLP), and Deep Learning techniques have therefore emerged as important approaches for automated fake news detection. This review examines the major AI-based techniques used for identifying fake news, with particular emphasis on text preprocessing, feature extraction, machine learning algorithms, deep learning models, and natural language processing. Traditional machine learning approaches such as Naive Bayes, Logistic Regression, Support Vector Machines, and Random Forest have been widely explored, while deep learning architectures such as Convolutional Neural Networks, Recurrent Neural Networks, Long Short-Term Memory networks, and transformer-based models provide more advanced methods for capturing contextual and linguistic patterns. The review also discusses commonly used datasets, evaluation metrics, challenges related to misinformation, dataset quality, language diversity, evolving news patterns, and model generalization. Finally, future research directions including multilingual detection, explainable AI, multimodal analysis, and real-time misinformation detection are discussed. This review highlights the potential of AI-based approaches to support automated and scalable fake news detection systems while emphasizing the need for reliable datasets, continuous model updating, and human fact-checking.
DOI: http://doi.org/10.5281/zenodo.22874412
Quantifying the Self: Disciplinary Power, Performativity, and Numerical Identity in Laurie Halse Anderson’s Wintergirls
Authors: Associate Professor Dr. Balamayuranathan B, Assistant Professor Ms. Subitsha S
Abstract: This article investigates the formation of feminine subjectivity in Wintergirls through the theoretical frameworks of Michel Foucault and Judith Butler. Existing scholarship has predominantly interpreted anorexia in the novel as a manifestation of psychological disorder, cultural pressure, or bodily control. However, such approaches treat numerical practices—calorie counting, weight monitoring, and bodily calculation—as symptomatic rather than constitutive. This study argues that these practices function as structuring mechanisms of identity. Using qualitative close reading, the analysis demonstrates that Lia’s subjectivity is produced through repeated acts of quantification that displace narrative and experiential self-understanding. Drawing on Foucault’s concept of disciplinary power and technologies of the self, the article shows how surveillance is internalized, rendering the subject both observer and observed. Butler’s notion of performativity further explains how iterative acts of measurement stabilize identity through repetition. The article introduces the concept of “quantified subjectivity” to describe a mode of selfhood constituted through numerical reaffirmation. In this framework, identity is not dissolved but hyper-produced through metric regulation, where numbers acquire epistemic authority over bodily perception and narrative coherence. By foregrounding quantification as a central organizing principle, the study extends existing readings of Wintergirls and contributes to broader discussions on embodiment, discipline, and contemporary forms of self-construction.
Real-Time Cloud Analytics And Machine Learning For Demand Forecasting And Stock-Out Prevention
Authors: Dr. Pankaj Malik, Heer Wadhwani, Rigved Bhondve, Mehar Sahajpal, Liva Jain, Mahi Garg
Abstract: Stock-out events and inaccurate demand estimation are major challenges in modern inventory management, particularly in environments where sales and inventory conditions change rapidly. This paper proposes a Real-Time Cloud Analytics and Machine Learning framework for Demand Forecasting and Stock-Out Prevention that integrates cloud-based data processing, feature engineering, machine learning-based demand forecasting, inventory monitoring, and real-time stock-out risk assessment. The framework utilizes historical sales, inventory levels, price and promotion information, lead time, rolling demand, and temporal features to generate short-term demand forecasts. The predicted demand is combined with safety stock and reorder-point estimation to identify potential stock-out conditions and trigger replenishment decisions. A comparative evaluation of Random Forest, XGBoost, LightGBM, and the proposed cloud-ML framework is conducted using MAE, RMSE, and MAPE, together with stock-out and inventory-management measures. In the illustrative experimental results, the proposed framework achieves an MAE of 13.47, RMSE of 19.84, and MAPE of 7.12%, compared with 15.82, 22.76, and 8.35%, respectively, for LightGBM. The proposed framework also reduces stock-out events to 9, corresponding to a stock-out rate of 2.14%, while achieving a 97.86% service level and an average inventory level of 405 units. These results indicate the potential of integrating real-time cloud analytics with machine learning to improve demand forecasting, inventory visibility, and proactive stock-out prevention.
DOI: http://doi.org/10.5281/zenodo.22890767
Distributional Safe Reinforcement Learning Using Implicit Quantile Networks For Liquidity Stress Management
Authors: Dr. Pankaj Malik, Tejasva Sharma, Tasneem Dewaswala, Viyaylaxmi Sharma, Akshat Jain, Vaibhav Sharma
Abstract: Liquidity stress management requires financial institutions to make sequential funding and asset-allocation decisions under uncertainty, where rare but severe shortfalls carry disproportionate cost. Conventional reinforcement learning (RL) agents optimize expected return and are therefore poorly suited to this setting, since they are indifferent to the shape of the return distribution and provide no explicit safety guarantee against tail losses. This paper proposes DSRL-IQN, a distributional safe reinforcement learning framework that combines an Implicit Quantile Network (IQN) with a Conditional Value-at-Risk (CVaR) risk estimator and an explicit safety layer for liquidity stress management. Rather than learning a single expected return, DSRL-IQN learns the full return distribution conditioned on state and action, from which a CVaR estimate of tail risk is derived at each decision step. A constraint-projection safety layer then restricts the agent to the subset of actions whose estimated CVaR remains within an institution-defined liquidity buffer, and selects the safe action with the highest expected return. The framework is evaluated on a liquidity-stress simulation environment constructed from balance-sheet and market-based liquidity indicators, under four synthetically generated stress regimes of increasing severity (Normal, Moderate, Severe, Extreme). DSRL-IQN is compared against DQN, Double DQN, PPO, QR-DQN and a plain IQN agent using return, Sharpe ratio, maximum drawdown, CVaR at the 5% level, and the rate of safety-constraint violations. Experimental results show that DSRL-IQN attains the highest risk-adjusted return and the lowest tail risk and violation rate across all stress regimes, while an ablation study confirms that the CVaR estimator and the safety layer contribute complementary and largely additive improvements. A sensitivity analysis further characterizes the trade-off between profitability and safety as the CVaR confidence level is varied. The results indicate that distributional, risk-constrained reinforcement learning is a practical and effective approach for liquidity stress management under uncertainty.
DOI: http://doi.org/10.5281/zenodo.22891885
Mobile Addiction: The Lost Childhood Of Youth And The Growing Concern Of Parents Digital Dependency Among Children And Adolescents: Challenges For Parents, Education, And Society
Authors: Dr. Manju Gupta
Abstract: This study highlights that while mobile phones have become an indispensable part of the digital age, their excessive and uncontrolled use has emerged as a serious concern, particularly for children and adolescents. Mobile addiction is contributing to significant psychological, social, and educational challenges, profoundly affecting the overall development of the younger generation. As a result, childhood is rapidly changing, with children becoming increasingly detached from outdoor activities, social interactions, and family bonding, leading to adverse effects on their mental health, emotional well-being, and personality development. The study reveals that mobile addiction is not merely an individual habit but has evolved into a broader social and familial challenge. Addressing this issue requires the collective efforts of parents, educators, policymakers, and society as a whole. Parents’ concerns are both natural and justified, as they directly observe changes in their children’s behavior, declining academic performance, irritability, reduced attention span, and increasing social isolation. Considering the seriousness of the problem, the study recommends practical measures such as promoting digital detox program, regulating screen time, and encouraging participation in offline recreational, educational, and physical activities. Families must also serve as role models by practicing responsible and balanced digital behavior. Educational institutions should incorporate digital literacy, mental health awareness, and self-regulation strategies into their curricula to help students develop healthy technology-use habits. Furthermore, policymakers should formulate and implement effective policies to ensure a safe, balanced, and child-friendly digital environment. The study concludes that through coordinated efforts by families, schools, communities, and government agencies, it is possible to reduce mobile addiction and restore a healthier, more active, and happier childhood for future generations.
DOI: http://doi.org/10.5281/zenodo.22892130
Cyber Security in Bioinformatics: Threats, Vulnerabilities, and Defense Tactics
Authors: Research Scholar J.Bindhu Bhargavi
Abstract: Large-scale genomic and biological data processing, analysis, and storage are now possible because to the quick development of bioinformatics. However, there are also serious cybersecurity threats associated with the growing digitization of biological data, endangering the availability, confidentiality, and integrity of data. Because they contain individually identifiable and unchangeable information that cannot be changed once disclosed, genomic records are particularly sensitive. Examining vulnerabilities in cloud-based analytic platforms, genetic databases, bioinformatics pipelines, and data-sharing frameworks, this study explores the nascent topic of cyber-bio security. Unauthorized access, ransomware attacks, malicious code injection, insider threats, and data manipulation inside sequencing and analysis workflows are among the common threat vectors reviewed in the study. It also examines security flaws in bioinformatics software development caused by insufficient encryption techniques, shoddy authentication procedures, and a lack of security-by-design principles. The article suggests a multi-layered security framework that includes block chain-based integrity verification, encryption standards, access control policies, secure software engineering techniques, and AI-driven intrusion detection systems to address these issues. There is also discussion of ethical, legal, and regulatory issues pertaining to international data governance and genetic privacy. In order to guarantee robust protection of biological data assets, the results emphasize the critical necessity for interdisciplinary cooperation between cyber security specialists, bio informaticians, healthcare organizations, and legislators.
Next-Generation It Solutions Using Artificial Intelligence and Cloud Computing
Authors: Assistant Professor M.Jeswanth, K.Sai Santhiya, K.Mariammal, K.Surya, G.Ashok, E. Sakthi
Abstract: The rapod development of information technology has transformed the way organizations, businesses, educational institutions, and individuals store, process, and utilize information. Two major technologies driving this transformation are Artificial Intelligence (AI) and Cloud Computing. The combination of AI and cloud computing provides powerful, scalable, intelligent, and cost-effective solutions for next-generation information technology systems. Artificial Intelligence enables machines and software to learn from data, identify patterns, make predictions, automate tasks, and support intelligent decision-making. Cloud computing provides on-demand access to computing resources, storage, applications, databases, and networking services through the internet. Integrating these technologies creates a flexible IT environment capable of handling large volumes of data and delivering intelligent services efficiently. The integration of AI and cloud computing offers several advantages, such as scalability, flexibility, reduced infrastructure costs, improved productivity, faster decision-making, and better resource utilization. Organizations can increase or decrease computing resources according to their requirements while using AI services through cloud platforms. However, several challenges must be considered, including data privacy, cybersecurity, dependence on internet connectivity, ethical concerns, algorithmic bias, service availability, and the complexity of managing AI workloads in cloud environments. Proper security mechanisms, responsible AI practices, data governance, and skilled professionals are therefore essential for successful implementation. The future of IT is expected to move toward highly automated, intelligent, and interconnected systems. Emerging technologies such as edge computing, Internet of Things (IoT), generative AI, serverless computing, and intelligent cybersecurity can further strengthen the capabilities of cloud-based AI solutions. Overall, the integration of Artificial Intelligence and Cloud Computing represents an important foundation for next-generation IT solutions, enabling organizations to build smarter, .
Environmental Pollution And The Harmful Effects Of Pollutants On The Ecosystem
Authors: Odioko Mercy Obianuju, Patrick O. Nwosibe, Paul Nnamdi Ogbu, Olubunmi Abiola Adewusi, Tasie Chimezie Michael, Ibrahim Gadaka Kizito, Onwugbuta Godpower Chukwuemeka, Adamu Musa Dagare, Erienu Obruche Kennedy
Abstract: Environmental pollution is a major global concern arising from increasing industrialization, urbanization, agricultural activities, transportation, petroleum exploration, mining and improper waste disposal. This research work examines environmental pollution and the harmful effects of pollutants on ecosystem components, with emphasis on their sources, environmental fate, transport and ecological consequences. Major forms of pollution, including air, water and soil pollution, are discussed alongside important pollutants such as heavy metals, petroleum hydrocarbons, pesticides, nutrients, plastics and other hazardous substances. This review highlights how pollutants alter soil properties, water quality and atmospheric conditions and affect plants, animals and microorganisms. Particular attention is given to bioaccumulation, biomagnification, biodiversity loss and disruption of ecosystem functions. The potential implications of environmental pollution for human health and ecosystem sustainability are also examined. Furthermore, pollution prevention, monitoring, remediation and ecosystem restoration strategies are considered, including physical, chemical, biological, phytoremediation and emerging nanotechnology-based approaches. This review emphasizes integrated and sustainable pollution-management practices to reduce environmental degradation, protect biodiversity and promote long-term ecosystem health.
DOI: http://doi.org/10.5281/zenodo.22933995
Mozambican Financial System: Use Of AI In ATMs With Biometric Authentication And Chatbots
Authors: Yasser Muetirrue Omar
Abstract: The digitalization of financial services has increased the importance of developing banking technologies that are secure, accessible, and user-oriented. This study analyzes the potential integration of Artificial Intelligence (AI), biometric authentication, and conversational chatbots into Automated Teller Machines (ATMs) to address security and accessibility challenges in the Mozambican banking context. The study adopts a qualitative and conceptual research design based on a structured literature review and scenario-based comparative analysis. Relevant literature on AI in banking, biometric authentication, ATM security, banking chatbots, financial inclusion, and digital literacy was analyzed thematically. A conceptual comparison was also developed between conventional PIN-based ATM systems and a proposed AI-biometric ATM model, focusing on authentication security, user assistance, accessibility, and operational usability. The reviewed evidence suggests that biometric authentication can strengthen user verification and reduce vulnerabilities associated with the sharing or observation of PINs, while AI-powered conversational interfaces may improve user autonomy by providing real-time, step-by-step assistance.
DOI: http://doi.org/10.5281/zenodo.22938892
A Comprehensive Review of Edge-Based Information-Centric Networking and Blockchain Integration for Scalable and Secure Vehicular Ad-Hoc Networks
Authors: Mr. Hari Krishna Chilakala, Dr. Boddepalli Kiran Kumar
Abstract: Vehicular Ad-hoc Networks (VANETs), utilized as a part of Intelligent Transportation Systems (ITS), are crucial for transmitting the information in real time between the vehicles as well as between the vehicles and the infrastructure, thereby improving safety, traffic management, as well as driving efficiency. Despite these, VANETs still have some problems such as latency, scalability, and security. This review is of some approaches to tackle such challenges, taking a look into the edge computing, blockchain technology and Information-Centric Networking (ICN) integration. This reduces latency because data is processed closer to the source, which enhances real-time decision making and provides quicker responses in the most critical situations. The main purpose of blockchain is to guarantee secure communication through a decentralized, tamper-proof Ledger that guarantees integrity of data and does not allow access to unauthorized requests. ICN treats data transmission as a problem of optimizing important messages, minimizing latency, and increasing scalability. In the realm of VANETs, the advantages and disadvantages of these technologies are discussed compared to each other in the review. It also points out possible areas for future research such as energy efficiency, privacy preserving techniques and deployment of the solutions in the real world. This paper aims to study the manner in which these technologies can enhance VANET performance and enhance the development of more secure, scalable, and efficient ITSs.
DOI: https://zenodo.org/records/22974242
Investor Perception of Market-Based Financing for India’s Energy Transition: A Comparative Study of Ongc, Ntpc and Coal India
Authors: Vunyala Pravalika, Dr. K. Famitha Begum
Abstract: Aims: To examine investor perception of market-based financing instruments such as green bonds, Qualified Institutional Placements, Offer for Sale disinvestment and ESG-linked bonds used by Oil and Natural Gas Corporation (ONGC), National Thermal Power Corporation (NTPC) and Coal India Limited (CIL) to fund India’s energy transition. Evaluate the influence of government backing and disinvestment transparency on investor confidence. Assess the effect of dividend yield and return expectations on investor perception. Assess the influence of ESG and sustainability disclosures on investor confidence towards the three public sector energy undertakings. Hypothesis: Demographic variables are not significantly associated with investor perception towards market-based financing instruments of ONGC, NTPC and Coal India. Approach: The purpose of the research methodology is to resolve the research problem through a scientific evaluation of the research topic, namely investor perception of market-based financing for India’s energy transition among ONGC, NTPC and Coal India investors. The term “required primary data” refers to data that is original, suitable for questionnaire-based survey procedures, and generates high accuracy through structured discussions with respondents. Secondary data for this study have already been collected for a specific purpose from annual reports, stock exchange filings, business journals, newspapers, websites and prior theses. Examination: The Test of Satisfaction towards market-based financing instruments and the Test of F values for investment amount and age furnished by the perception of investors towards ONGC, NTPC and Coal India. Nevertheless, this demonstrates that investors in the 36-45 years age group and those investing between Rs. 50,001 and Rs. 1,00,000 annually are more likely to be affected by the Test of Satisfaction regarding market-based financing instruments. In summary, market-based financing is becoming the most critical and closely watched avenue through which public sector energy undertakings raise capital for the energy transition. The “Investor Confidence Index” built on the perception of respondents towards green bonds, QIPs, disinvestment and ESG-linked instruments is the most frequently used parameter for assessing the market acceptance of PSU energy financing. Regulators and company managements are obligated to strengthen transparency and disclosure practices that optimise investor confidence. Nevertheless, market-based financing for the energy transition encounters a few obstacles during the structuring and implementation phases. The objective of this paper is to examine these obstacles in order to improve investor confidence and to determine strategies for overcoming them. Furthermore, it examines and compares investor perception across ONGC, NTPC and Coal India.
DOI: https://zenodo.org/records/23010455
IoT Based Automatic Speed Control Using QR Scanner or GPS
Authors: Tejas Nanasaheb Rahane, Sakshi Vikram Shelke, Shravani Sambhaji Hinge
Abstract: Overspeeding is one of the major causes of road accidents, injuries, and loss of lives. The proposed Automatic Speed Control Using QR Scanner and GPS system aims to improve road safety by automatically controlling vehicle speed in speed-restricted areas. QR codes placed on roads contain the permitted speed limit for specific locations such as school zones, hospitals, construction areas, speed breakers and hilly roads. An onboard QR scanner reads the QR code when the vehicle enters a restricted zone. GPS provides location information, and a microcontroller processes the collected data. The processed speed-limit information is then sent to the Engine Control Unit (ECU), which automatically adjusts the vehicle speed according to the specified limit. The system reduces the need for manual speed control and helps vehicle maintain safe speeds in critical areas. It can also support smart-city traffic management and improve overall transportation safety. However, regular maintenance of QR codes and scanning devices is required, and performance may be affected by poor weather, low light, or damaged QR codes.Overall, the proposed system provides a practical approach to automatic speed regulation by combining QR scanning and GPS technology, contributing to safer and smarter transportation.
AI-Based Autonomous Warehouse Robot Car: Design, Integration, and Experimental Evaluation of an Edge-AI Sorting Platform
Authors: Almutasim Billa Alanazi, Abdullah Bamufleh, Mohammed Arnaout
Abstract: Warehouse automation is increasingly important for improving operational efficiency, reducing sorting errors, and supporting high-throughput logistics. This paper presents the design, implementation, and experimental evaluation of an AI-based autonomous warehouse robot car for real-time color-based sorting of uniformly sized boxes in an indoor warehouse environment. The robot integrates a dual-camera perception architecture, an ONNX-optimized YOLO object-detection model, ArUco marker-based localization, a Raspberry Pi 5 embedded processor, four Mecanum wheels for omnidirectional mobility, ultrasonic sensing for final approach control, a servo-driven gripper, and a Flask-based real-time monitoring interface. Autonomous task execution is organized using a finite-state machine that sequences Search, Align, Approach, Pick, Deliver, and Exit states, while a deterministic priority rule (Red > Blue > Green) defines the order of box handling. Experimental testing was performed in an approximately 3 m × 3 m simulated warehouse with multiple runs and varying numbers of red, blue, and green boxes. The reported results showed object-detection accuracy of 85%–95%, alignment accuracy of approximately 92%, pick-and-place success of 91%–95% with more than 95% success in most test conditions, and delivery accuracy of approximately 97%. The average detection-to-delivery time was approximately 20 s per box, system latency was below 100 ms, and the priority-sorting logic achieved 100% correctness with no cross-color delivery errors. The findings demonstrate the effectiveness of integrating edge AI, marker-based localization, omnidirectional locomotion, robotic manipulation, and real-time monitoring on a low-cost embedded platform.
Agentic AI Framework For Autonomous Decision – Making In Smart Enterprise Systems
Authors: Dr. Shobana V, Ms.Maheswari P, Ms. Anitha M, Ms.Divyalakshmi.S
Abstract: Enterprise Systems are now becoming increasingly reliant on autonomous decision-making capabilities that go beyond rule-based automation. This paper puts forward an extensive framework for agentic AI systems which include large language models, multi-agent systems, and confidence-based governance which allows for reliable decision-making to be achieved autonomously. The framework takes into consideration the problem of reconciling the need for autonomy and accountability in enterprises and consists of a five-layer system architecture including perception layer, reasoning layer, orchestration layer, execution layer and governance layer. A confidence-based human in the loop system decides on how much governance each decision needs depending on the level of confidence in a particular decision. The experimental evaluation on enterprise workflow systems shows considerable improvement in reducing workflow downtime, improvement in task allocation efficiency and achievement of task autonomy within strict enterprise constraints. Comparisons among the five different agentic AI frameworks show that architecture modularity and governance integration play a critical role in successful implementation of agentic AI systems in real-life settings.
DOI: http://doi.org/10.5281/zenodo.23095376
Lightweight Machine Learning Framework For Secure IoT Device Authentication Using Edge Intelligence
Authors: Dr.M.Vijaya Maheswari, Dr. Mohammad Shahbaz Khan
Abstract: The growth of Internet of Things (IoT) has created some security issues, especially in the authentication process that is often associated with very limited computation capability. This paper suggests a lightweight machine learning approach for IoT device authentication based on edge intelligence. The suggested approach utilizes small scale convolutional neural networks that run on edge gateways to generate unique features based on the information obtained from channel state information and radio frequency fingerprinting. No cloud connection is needed due to this continuous authentication approach. Quantization and pruning are used to make sure the model is lightweight and energy efficient while having good authentication capabilities. Experimental validation on a testbed made of Raspberry Pi shows that good performance of authentication can be achieved using a lightweight model and fast inference per cycle of authentication. Comparison to previous cryptographic and machine learning methods proves that better performance can be achieved in terms of both energy usage and protection from attack like replay and impersonation.
DOI: http://doi.org/10.5281/zenodo.23095607
Ecocriticism And The Changing Human–Nature Relationship In Contemporary Canadian Literature
Authors: Dr. J. Felcia Pushpalatha
Abstract: Contemporary Canadian literature increasingly represents the relationship between human beings and the natural world as complex, interdependent, and historically shaped by colonialism, industrialization, technology, and environmental change. Ecocriticism provides an important framework for examining these changing relationships because it shifts literary analysis from an exclusively human-centred perspective towards an understanding of the interconnectedness of human, non-human, and material worlds. This paper examines the changing human–nature relationship in contemporary Canadian literature through an ecocritical reading of selected works, with particular attention to Margaret Atwood’s Oryx and Crake (2003), Cherie Dimaline’s The Marrow Thieves (2017), Waubgeshig Rice’s Moon of the Crusted Snow (2018), and Lee Maracle’s Celia’s Song (2014). These texts represent different literary responses to ecological crisis, including climate change, technological domination, resource extraction, colonial displacement, loss of biodiversity, and Indigenous relationships with land. The study argues that contemporary Canadian literature has moved beyond the conventional representation of nature as wilderness or passive background. Instead, nature is increasingly presented as an active presence within human history, culture, memory, survival, and identity. The selected Indigenous texts further complicate conventional Anthropocene narratives by connecting environmental destruction with colonial histories and emphasizing relationality, community, traditional knowledge, and land-based responsibility. The paper concludes that contemporary Canadian literature contributes significantly to ecological thought by challenging anthropocentrism and encouraging a more reciprocal understanding of human–nature relationships.
DOI: http://doi.org/10.5281/zenodo.23096860
Economic Zoology: A Quantitative Review of Animal-Based Enterprises, Livelihoods, Resource Recovery and Sustainable Bioeconomy
Authors: Assistant Professor Dr. M. Stephen
Abstract: Economic zoology deals with the utilization of animals and animal-derived products for food, fibre, medicine, agriculture, industry and livelihood development. Traditionally, economically important zoological enterprises have included sericulture, apiculture, lac culture, fisheries, aquaculture, vermiculture and animal husbandry. In recent years, the field has expanded toward insect-based industries, biological control, pollination services, insect-derived biomaterials and circular bioeconomy applications. This review examines the economic, nutritional, agricultural and environmental significance of major animal-based enterprises using quantitative evidence. Global fisheries and aquaculture production reached approximately 235 million tonnes in 2024, including 195 million tonnes of aquatic animals, while aquaculture exceeded 100 million tonnes for the first time. The sector supports more than 600 million livelihoods worldwide.In India, insect-based enterprises such as apiculture, sericulture and lac culture remain important rural livelihood activities, while emerging applications include insect protein, pollination, waste decomposition and pharmaceutical products. Economic zoology therefore represents an important interface between biodiversity, food production, rural employment, industrial raw materials and sustainable resource utilization. The review highlights opportunities for integrating traditional zoological enterprises with emerging technologies to strengthen the circular bioeconomy.
Natural and Multifunctional Preservation Strategies for Cosmetic Formulations: A Review of Antimicrobial, Biofilm-Control and Self-Preserving Approaches
Authors: Paurnima Pahurkar Pingle, Sakina Punjab
Abstract: Microbial contamination is a critical quality and safety concern in cosmetic formulations, particularly in products containing water and other nutrients that can support microbial proliferation. Conventional preservation systems have historically relied on antimicrobial agents such as parabens, phenoxyethanol, isothiazolinones and formaldehyde-releasing preservatives. At the same time, increasing interest in naturally derived, multifunctional and label-compatible ingredients has stimulated research into preservation systems that combine antimicrobial activity with other formulation functions. This review examines the scientific basis of multifunctional preservation systems incorporating glyceryl caprylate, glyceryl undecylenate, caprylyl glycol and propanediol, together with botanical antimicrobial agents including clove oil, tea tree oil, rosemary-derived materials, citronella oil and nutmeg oil. Emphasis is placed on proposed antimicrobial mechanisms, including perturbation of microbial membranes, modification of the physicochemical environment and enhancement of antimicrobial efficacy through ingredient combinations. The review further discusses the concept of self-preserving cosmetic systems and the relevance of microbial biofilms to preservation performance. Approaches for in-vitro evaluation, including minimum inhibitory concentration (MIC) testing, biofilm biomass assays, mature-biofilm disruption studies, scanning electron microscopy (SEM), formulation stability assessment and preservative efficacy testing, are considered as complementary tools for development and characterization. Overall, the literature and formulation concepts reviewed support a multi-hurdle approach in which antimicrobial activity is not attributed to a single ingredient but is evaluated as a function of ingredient interactions, formulation environment and microbial target.
Machine Learning Framework for Student Academic Failure Prediction: A Systematic Literature Review and Research Framework
Authors: Babandi Usman, Salim Ahmad, Zahraddeen Sufyan
Abstract: The early identification of students who are likely to experience academic failure has become an important application of machine learning (ML) and learning analytics in higher education. Although numerous studies have developed predictive models for student performance, the literature remains fragmented across performance prediction, at-risk identification, dropout prediction, and student success modelling, with substantial variation in outcome definitions, predictor availability, validation strategies, and evaluation metrics. This systematic literature review synthesizes research on ML-based prediction of student academic outcomes, with particular emphasis on failure-oriented and early-warning applications. A PRISMA-aligned evidence-synthesis strategy was used to examine systematic reviews, empirical studies, established educational datasets, and recent developments in explainable and responsible artificial intelligence. The reviewed evidence indicates that classification-based approaches, particularly decision trees, random forests, support vector machines, logistic regression, artificial neural networks, and ensemble methods, dominate the field. Academic history, assessment performance, learning-management-system activity, attendance, engagement, and demographic or contextual variables are frequently used predictors. However, high predictive accuracy does not necessarily indicate practical usefulness because information leakage, class imbalance, temporal instability, inadequate calibration, limited external validation, and weak interpretability remain important methodological concerns. Recent studies further demonstrate the importance of explainable AI, fairness assessment, privacy-preserving data generation, and temporal modelling. Based on these findings, this study proposes a Failure-Oriented Explainable Temporal Machine Learning Framework (FET-MLF), integrating leakage-aware temporal feature engineering, predictive modelling, probability calibration, explainability, fairness auditing, and intervention feedback. The framework shifts evaluation from accuracy-centric modelling toward trustworthy, actionable, and institutionally deployable early-warning systems for preventing academic failure.
Mechanical Properties Of Concrete By Incorporating Cementitious Materials (Rice Husk Ash, Fly Ash And Marble Powder)
Authors: Ahmed Raza, Hasnain Ahmed, Ali Hamza, Zaheer Ahmed, Sadam Hussain
Abstract: This study experimentally investigates the influence of Rice Husk Ash (RHA), Fly Ash (FA), Marble Powder (MP) and Expanplast SP100 on hardened properties of concrete. A total of 14 concrete mixtures were prepared, including a control mixture and modified mixtures containing individual and combined replacements of RHA, FA, and MP, while 1% Expanplast SP100 was incorporated in the modified mixtures. Mechanical properties were assessed through 28-day compressive, splitting tensile and flexural strength tests. The results demonstrated that moderate replacement levels generally improved the mechanical properties of concrete. The 10% RHA mixture achieved a compressive strength of 33.92 MPa, compared with 30.00 MPa for the control mixture, representing an improvement of 13.07%. The corresponding splitting tensile strength increased from 2.85 MPa for the control to 3.24 MPa, an improvement of 13.68%. For flexural strength, the 10% RHA mixture achieved 4.76 MPa compared with 4.20 MPa for the control, representing an increase of 13.33%. The highest overall performance was obtained from the ternary mixture containing 5% RHA, 5% FA, and 5% MP, which achieved compressive, splitting tensile and flexural strengths of 35.10, 3.40, and 4.95 MPa, respectively. These values correspond to improvements of approximately 17.0%, 19.30%, and 17.86%, respectively, compared with the control concrete. The improvement is attributed to the combined effects of pozzolanic reactions, filler action, improved particle packing, pore refinement and enhanced cement-particle dispersion provided by Expanplast SP100. The findings demonstrate that the combined utilization of agricultural and industrial waste materials can provide a technically effective and more sustainable approach to concrete production while reducing dependence on conventional cementitious materials.
DOI: http://doi.org/10.5281/zenodo.23100719
Optical, Structural, Electrical, and Morphological Properties of Nanostructured ZnO Thin Films Prepared by the Sol–Gel Dip-Coating Technique
Authors: Anil Manjula Ramdas Bari
Abstract: Nanostructured zinc oxide (ZnO) thin films were deposited on ultrasonically cleaned glass substrates using a sol–gel dip-coating technique. Zinc acetate dihydrate, ethanol, and ammonium hydroxide were used as the starting materials. Films with different numbers of coating cycles were prepared and subsequently annealed at 500 °C for 1 h. The structural, morphological, elemental, electrical, and optical properties of the films were investigated using X-ray diffraction (XRD), scanning electron microscopy (SEM), energy-dispersive X-ray spectroscopy (EDX), electrical conductivity measurements, and UV–visible spectroscopy. The crystallite/grain size increased with increasing number of coating cycles. The films exhibited nonstoichiometric ZnO composition and transparent surface morphology. The optical band-gap energy obtained from the absorption spectra was found to depend on the number of coating cycles. The electrical activation energy also increased with increasing grain size. These results demonstrate that the structural and optical properties of sol–gel-derived ZnO thin films can be modified by controlling the number of coating cycles.
GenTutor-HE: A Hybrid Human-in-the-Loop Framework for Personalized, Ethically Governed Generative AI Integration in Higher Education
Authors: Assistant Professor Dr. V. Maria Antoniate Martin, Praveen J
Abstract: Generative Artificial Intelligence (GenAI) has rapidly transitioned from an experimental technology into a mainstream instructional resource across higher-education classrooms. Building on five recent systematic and conceptual reviews of AI in education, this paper synthesizes the reported responses, attitudes, behaviors, applications, benefits, and challenges associated with GenAI adoption in university-level teaching and learning. The review of existing literature reveals that while GenAI substantially improves personalization, engagement, and assessment efficiency, its integration is constrained by ethical concerns, over-reliance risks, unequal infrastructure access, and a persistent lack of human-centered oversight. To address these gaps, this paper proposes GenTutor-HE, a hybrid, human-in-the-loop framework that combines a generative AI tutoring engine, an adaptive knowledge-tracing module, and an explicit ethical-governance layer that filters generated content for bias, privacy risk, and factual reliability before it reaches learners. The proposed methodology follows a design-science research approach comprising requirement analysis, architectural design, prototype development, and mixed-method evaluation. A comparative analysis of existing intelligent tutoring systems, adaptive learning platforms, and generative content-based systems is presented, along with the advantages and limitations of the proposed model. The paper concludes that a balanced, ethically governed hybrid approach offers a more sustainable pathway for GenAI adoption than fully automated systems, and it outlines future enhancements including multilingual support, emotion-aware feedback, and federated-learning-based privacy preservation.
Quantum Randomness at Conception, Chaotic Amplification, and the Infinite Within
Authors: Harsh
Abstract: Every human life begins from a near-infinite space of genetic possibilities, narrowed to one individual by randomness at the molecular scale. This paper argues that such randomness, mostly classical and in a narrow but real case quantum (proton tunneling in DNA), sets the initial conditions of each organism, and that the nonlinear dynamics of development amplify these tiny differences into a materially different life. Evidence from identical twins, clones and isogenic model organisms supports the amplification step. We separate what is defensible physics from what is not, propose an experimental test for a specifically quantum contribution, and close with the author’s philosophical thesis: that in searching for the infinite in outer space and in the quantum realm, we overlooked the only living infinities, human beings themselves.
AI-Based Dynamic Public Transport Crowding Prediction And Intelligent Route Recommendation System
Authors: Dr. C. Venish Raja, Mohana M
Abstract: Public transport in fast-growing cities is under constant pressure, with buses and metro coaches becoming overcrowded at peak hours while remaining underused at other times. Passengers often have no reliable way to know how crowded a vehicle will be before it arrives, while fixed timetables and static journey planners cannot respond effectively to changing demand. Artificial intelligence provides an alternative by treating crowding as a time-series forecasting problem and using predicted occupancy as an input to route recommendation. Research reviewed in this paper reports that recurrent deep-learning models such as Long Short-Term Memory (LSTM) and bidirectional LSTM (BiLSTM), together with tree-based ensembles such as Random Forest and XGBoost, can outperform traditional statistical and schedule-based approaches for occupancy forecasting. Integrated systems that connect prediction with fleet decisions and passenger recommendations are also reported to provide greater operational benefits than isolated prediction models. To confirm that the problem is relevant from the passenger perspective, a primary survey of 123 public transport users was conducted. The survey indicates that overcrowding is regularly experienced by a substantial share of respondents, that many respondents have changed their route because of crowding, and that willingness to trust an AI-based crowding system is divided. These findings support the need for transparent and explainable recommendations rather than an unexplained black-box score. This paper presents the main approaches to AI-based crowding prediction, reports the primary survey findings, explains important feature families and model types, describes a route recommendation method that considers predicted crowding together with travel time, waiting time and transfers, discusses real-time deployment and robustness, and proposes an evaluation plan. The proposed direction combines a data layer, BiLSTM and gradient-boosting prediction models, an ensemble mechanism, and a multi-objective recommendation layer. The overall objective is to support less crowded and efficient journeys while helping transport operators balance capacity. The paper also identifies challenges involving unusual events, new routes, sensor failures, privacy, fairness, passenger behaviour and technology acceptance, and recommends future work on adaptive, lightweight and privacy-aware systems.
Experimental Investigation on the Mechanical and Durability Properties of M40 and M70 Cement-Based Composite Using Waste Glass as Partial Replacement of Coarse Aggregate
Authors: Divakar Yadav, Associate Professor Dr.Ala Satyanarayana
Abstract: The increasing consumption of natural aggregates in concrete production and the accumulation of non-biodegradable discarded glass have created a need for sustainable alternative construction materials. This experimental investigation evaluates the feasibility of utilizing discarded glass as a partial replacement for conventional coarse-grained aggregate in standard-grade M40 and high-strength M70 concrete. Waste glass was incorporated at substitution proportions of 25%‚ 50%‚ 75%‚ and 100%‚ and selected mixes were also modified with microsilica to improve their performance. The study examined fresh properties such as ease of placement and wet density‚ as well as hardened properties including compression strength‚ splitting tensile strength‚ bending strength‚ impact resistance‚ elastic modulus‚ and dry density. Additionally‚ the quality and durability-related characteristics of this concrete were assessed through rebound hammer‚ ultrasonic pulse velocity and sorptivity tests.
Automated and AI Assisted Web Security Auditing a Focused Narrative Review
Authors: Akshat Kumar, Gaurav Sharma, Deep Singh, Roovi
Abstract: Automated web security auditing combines several tasks that produce different kinds of evidence: discovering exposed functionality, exercising application behaviour, interpreting observations, and deciding which weaknesses deserve attention. This focused narrative review examines five selected studies on cross-origin resource sharing, stateful REST testing, GraphQL testing, language-model-assisted penetration testing, and vulnerability prioritization. Four protocol specifications provide supporting definitions. The literature is compared by input requirements, treatment of application state, evidence produced, and limits of interpretation. The synthesis indicates that wider endpoint coverage, successful testing actions, and confirmed security weaknesses are distinct outcomes. Schema-aware testing helps explore structured APIs, while language models can support interpretation and planning; neither removes the need to validate findings against observable behavior. The review identifies evaluation comparability, authentication context, evidence of traceability, and constrained AI assistance as priorities for an AI Security Auditor project. It presents design implications and future evaluation questions without reporting a new implementation, benchmark, or experiment. The source selection is purposive rather than exhaustive, and conclusions are limited to the reviewed material.
AI- Driven Adaptive Resource Allocation For Next- Generation IoT Communication Networks
Authors: Nitish Kumar, Dr. Mohammad Shahbaz Khan
Abstract: The emergence of Internet of Things (IoT) devices in next-generation communication networks has brought about new and complex challenges related to resource management, which include massive connectivity, heterogeneous traffic loads, and strict energy considerations. Static and heuristic resource allocation algorithms have been shown to be incapable of addressing the highly dynamic, non-stationary characteristics of the 6G enabled IoT network ecosystem. This paper examines a number of AI-based adaptive resource allocation techniques utilizing deep reinforcement learning (DRL), federated learning (FL) as well as digital twin (DT) in order to allocate spectrum, power, and computational resources dynamically and efficiently. A hybrid design based on multi-agent DRL as well as hierarchical optimization for the joint optimization of transmit power, sub-channel allocation, and task offloading is presented. It is shown experimentally that the proposed scheme achieves significant reductions in terms of energy consumption and latency, while providing a high QoS satisfaction level compared to traditional heuristic algorithms.
DOI: http://doi.org/10.5281/zenodo.23160226
Stress-Strain Behaviour And Ductility Of Tie-Confined M40 And M70 Concrete Incorporating Waste Glass And PET Bottle Waste
Authors: D.Ramakrishna, Dr.Ala Satyanarayana
Abstract: Concrete is one of the most widely used construction materials and is generally produced using large quantities of natural sand and crushed stone aggregates. Nearly 65–70% of the volume of conventional concrete consists of these natural materials. As a result, the concrete industry places considerable demand on natural resources. With the growing emphasis on sustainable construction, there is a need to reduce the dependence on conventional aggregates by incorporating suitable waste materials into concrete. Materials such as waste glass and discarded polyethylene terephthalate (PET) bottles can potentially serve as alternative aggregates when used at appropriate replacement levels without significantly affecting the required mechanical performance of concrete. The use of glass products has increased considerably in residential and commercial buildings. Glass is commonly used in windows, partitions, interior decoration, furniture, façades and architectural finishes because of its aesthetic and functional advantages. However, its brittle nature results in a substantial quantity of waste during manufacturing, transportation, cutting, handling and installation. A large portion of this waste is often discarded rather than reused, creating disposal and environmental concerns. Similarly, PET, or polyethylene terephthalate, has become one of the most commonly used plastics, particularly for beverage and water bottles. PET bottles are preferred because of their relatively high strength, low density, durability, good appearance and low permeability to gases such as carbon dioxide. The rapid increase in the consumption of PET products has consequently resulted in a significant increase in plastic waste. In many regions, the amount of PET waste generated is considerably higher than the quantity that is recycled. Disposal of such waste in landfills creates long-term environmental problems because plastics degrade very slowly. Therefore, utilizing waste glass and PET materials as partial substitutes for conventional aggregates in concrete can offer a practical method of waste management while also reducing the consumption of natural resources. The present experimental investigation was carried out in different stages to evaluate the possibility of using waste glass and PET bottle waste as aggregate replacements in M40 and M70 grade concrete. In the initial stage, waste glass was introduced separately as a replacement for fine aggregate, coarse aggregate and as a combined replacement for both fine and coarse aggregates. Different replacement percentages were investigated in order to determine the proportion that produced the most suitable compressive strength. A similar procedure was followed for PET bottle waste. Based on the results obtained from these preliminary trials, optimum replacement levels were selected for both M40 and M70 concrete. The confinement index was therefore closely associated with the improvement in the axial response of the concrete specimens. In addition, the stress ratio, strain ratio and ductility factor showed an increasing trend with higher levels of confinement. Overall, the study demonstrates that selected quantities of waste glass and PET bottle waste can be incorporated into M40 and M70 concrete as partial replacements for natural aggregates. When combined with suitable lateral confinement, the resulting concrete can exhibit improved strength and deformation characteristics. This approach provides a useful means of recycling non-biodegradable waste, reducing the consumption of natural aggregates and promoting more sustainable concrete construction.
DOI: http://doi.org/10.5281/zenodo.23180861
The Invisible Threat: Microplastic Biomonitoring, Systemic Translocation, and Human Health Risks
Authors: John Chimerenka Ifenkwe
Abstract: Microplastics are an emerging global contaminant with profound implications for human and environmental health. Defined as plastic particles 1 µm to 5 mm in size, and nanoplastics less than 1µm (<1000 nm), these particles originate from primary manufacturing and secondary degradation of larger plastics. Their ubiquity in air, water, soil, food, and biological tissues has raised significant toxicological concern because microplastics are not inert. They act as physical particulates, chemical vectors for plasticizers, flame retardants, heavy metals, and as carriers for microbial pathogens and persistent organic pollutants. Upon inhalation, ingestion, or dermal contact, microplastics can induce oxidative stress, chronic inflammation, endocrine disruption, immune dysregulation, gut microbiome alteration, and potential genotoxicity. Evidence from in vitro, animal, and emerging human biomonitoring studies suggests associations with respiratory dysfunction, reproductive toxicity, developmental effects, and cardiovascular risk. This review consolidates existing evidence on microplastic sources, human exposure routes, mechanisms of toxicity, and associated health outcomes, while underscoring the critical need for standardized detection protocols, systematic human biomonitoring, development of safer polymers, and effective environmental remediation.
MediXpert: A Vision Language System for Clinical Decision Support
Authors: Mrs. D.R. Nanda Devi, Gnaneshwari Sarala, K. Yasaswitha, A. Likhitha
Abstract: In the context of the current healthcare environment, the accurate diagnosis of a patient’s health condition is only possible by the convergence of the self-reported symptoms by the patient and the analysis of medical images such as a chest X-ray, CT scans, or MRI scans. However, the current state-of-the-art models only provide the analysis of self-reported symptoms by the patient using a natural language-based system, which responds to the queries made by the patient (for instance, using transformer-based models such as BERT or GPT), or the analysis of medical images using a vision-based system (for instance, using ResNet or EfficientNet models). Therefore, the current fragmented approach not only affects the potential of accurate diagnosis by using AI-based models but also affects the comprehensive analysis carried out by the diagnostic professionals in the context of self-reported symptoms as well as medical images. A comprehensive analysis system using medical images as well as natural languages is not available in the current context; hence, the accessibility of healthcare has become a significant concern.
Numerical Characterization of Aperiodic Dynamics, Recurrence, and Initial-Condition Sensitivity in the Classical Lorenz System
Authors: Audu Abdulmalik, Onubedo, Nura A.Y Musa, Youngkumaphew Iliyasu
Abstract: This study presents a numerical investigation of nonlinear motion, recurrence, and predictability in the classical Lorenz system using a compact three-dimensional convection model. The analysis combines equilibrium considerations, local stability concepts, temporal integration, phase-space visualization, recurrence mapping, and sensitivity to nearby initial conditions. Numerical simulations show that the state variables remain bounded while developing irregular oscillations after an initial transient, with repeated reversals reflecting transitions between distinct regions of state space. The reconstructed three-dimensional trajectory exhibits the characteristic two-lobed attractor, while its planar projections reveal repeated circulation around the two non-zero equilibrium regions without convergence to a stationary state or stable periodic orbit. Successive local maxima form an organized nonlinear return relation, demonstrating that the apparently irregular oscillations retain an underlying deterministic structure. Sensitivity analysis further shows that initially close trajectories separate substantially with time, limiting reliable long-term prediction despite deterministic governing equations. The combined temporal, geometric, recurrence, and sensitivity analyses therefore provide a consistent computational characterization of bounded aperiodic motion in the Lorenz system. The results reinforce the importance of low-dimensional nonlinear models for understanding deterministic unpredictability and recurrent chaotic dynamics in mathematical physics.
Agentic AI: Transforming Autonomous Decision-Making In Intelligent Systems
Authors: K.Petchiammal, S.Subha, M.Pon Abisha, M.Maheswari, K.Rama Subramanian, P.Syed Mohamed Mydeen
Abstract: Agentic Artificial Intelligence (Agentic AI) is an emerging approach that enables intelligent systems to perform tasks with greater autonomy by combining reasoning, planning, memory, tool usage, and decision-making capabilities. Unlike conventional AI systems that mainly respond to predefined instructions, Agentic AI systems can interpret objectives, divide complex tasks into smaller steps, execute appropriate actions, evaluate outcomes, and modify their strategies when required. This paper presents a closed-loop framework for autonomous decision-making based on goal analysis, task planning, reasoning, tool selection, execution, feedback, verification, and adaptive re-planning. The proposed approach aims to improve the ability of intelligent systems to handle multi-step and changing tasks rather than producing a single response. The paper also discusses existing research, the research gap, experimental evaluation criteria, challenges, limitations, and future opportunities. The framework provides a structured foundation for developing more adaptive, reliable, and efficient autonomous intelligent systems.
Software-Defined Mobility: Decarbonising The Automobile By 2050
Authors: Professor (Dr.) Lalith Kumar Dharvath
Abstract: The global automotive sector faces an unprecedented existential crisis in the 21st century, placing the traditional industry under siege from climate mandates and changing consumer behaviors. To endure, automobility must undergo a profound century of transformation, moving beyond basic electrification toward Software-Defined Mobility (SDM). As of 2026, the global vehicle footprint has reached an immense scale, with approximately 1.6 billion active cars worldwide and India producing over 34.7 million vehicles annually. Managing the carbon footprint of this massive global fleet requires a radical technological shift. Software-Defined Vehicles (SDVs) provide a vital opportunity for adaptation by shifting value from physical hardware to intelligent code. SDM drives deep lifecycle decarbonisation through five critical levers. First, continuous over-the-air (OTA) updates extend vehicle lifespans, breaking the carbon-heavy manufacturing replacement cycle. Second, cloud analytics deploy advanced algorithms for intelligent battery optimization, maximizing efficiency. Third, digital twins virtualize testing, saving massive factory energy. Fourth, structural consolidation enables hardware dematerialization, lowering vehicle weight and energy consumption. Finally, connected platforms power smart infrastructure and shared mobility ecosystems, directly reducing urban congestion. Concurrently, the nature of competition is shifting. Traditional manufacturing metrics are being replaced by computational performance and software integration. As the industry races toward net-zero emissions by 2050, computational intelligence—rather than mechanical assembly—will determine which automakers survive this disruptive landscape.
DOI: http://doi.org/10.5281/zenodo.23186797
Assessment of Behaviour problems among Non-ADHD and ADHD Children in Relation to Child Variables
Authors: Research Scholar T. Deepa Reddy, Professor K.Anuradha
Abstract: Attention Deficit Hyperactive Disorder (ADHD) is a common neurodevelopmental disorder characterized persistent patterns of inattention, hyperactivity and impulsivity. Children with ADHD may also experience significant behavioral difficulties that can affect their academic performance, social relationships and family functioning. Comparing behavioural problems among children Non-ADHD and ADHD children can help identify areas early intervention. In the present study behavioual problems were assessed by using a standardized behaviour scale. The findings indicated a significant difference in overall behavioural scores between the ADHD and Non-ADHD groups.
DOI: https://doi.org/10.5281/zenodo.23208744
Design, Analytical Dimensioning And Field Evaluation Of A Low-Cost IP-Camera And DVR Security Monitoring System For A University Computer Engineering Laboratory
Authors: Owigho Benedicta, Edward O. Odesa, Omolegho A. Ibok, Jones U. Ekwemuka
Abstract: University computer-engineering laboratories contain high-value technology, but in resource-constrained settings, they are typically protected solely by locks, attendance registers, and manual supervision, none of which provide a continuous record of events. The design, analytical dimensioning, and field testing of an Electronic Laboratory Security Monitoring System (ELSMS) built from commercially available parts consisting of three 2-megapixel IP cameras, a Gigabit Ethernet switch, an 8-channel DVR with a 1 TB surveillance drive, Cat6 cabling, a regulated supply, and a 650 VA UPS are presented in this paper. The system is dimensioned using closed-form models for camera coverage, pixel-density-based identification range, storage retention, network load, and dependability. With the specified 1920-pixel sensors, identification-grade pixel density (250 px/m) is only achievable within 2.2-3.8 m of a camera (120°–90° field of vision); three cameras load less than 2% of a gigabit link; and 1 TB offers roughly 7.7 days of continuous recording at 4 Mbit/s per camera (15.4 days at 2 Mbit/s). 83 of the 86 structured tests were successful (96.5%; 95% Wilson CI 90.2-98.8%). Nine out of ten controlled occurrences were found, the recording and network tests each reported one failure in thirty tries, and the camera-coverage, video-quality, and playback tests all passed. The lag between display and recording was roughly two seconds. The findings demonstrate that evidence-grade picture detail, recording retention, and power autonomy rather than network capacity are the binding design limitations and validate the architecture’s appropriateness for low-cost deployment. The models are reusable by other institutions.
DOI: https://doi.org/10.5281/zenodo.23209130
Equilibrium Configurations and Stability Analysis of the Photogravitational Restricted Four-Body Problem with a Circumstellar Belt
Authors: Audu Abdulmalik, Onubedo, Idris Babaji Muhammad, Mohammed Garba
Abstract: The present study investigates the equilibrium structure and local stability characteristics of a planar photo gravitational restricted four-body problem under the combined influence of radiation pressure and an axisymmetric circumstellar belt. The dynamical model considers three primaries arranged in an equilateral triangular configuration, while the motion of an infinitesimal particle is governed by radiation-modified gravitational forces and the gravitational potential of the belt. The equilibrium equations are solved numerically using a systematic search procedure with multiple initial conditions, followed by residual verification and eigenvalue-based stability analysis. The influence of the dominant-primary radiation factor and belt mass parameter on equilibrium locations is examined. The results show that radiation pressure produces systematic displacement of equilibrium points by modifying the effective gravitational balance, although the stability classification remains unchanged for the investigated parameter sets. In the stellar application, the numerical search identifies additional axial equilibrium solutions beyond the previously reported configurations. These solutions are verified using high-precision calculations to confirm their satisfaction of the equilibrium conditions. Linear stability analysis reveals that the near-origin equilibrium possesses purely imaginary characteristic roots and is therefore linearly stable within the adopted infinitesimal-particle approximation, whereas the remaining detected equilibria exhibit instability due to characteristic roots with positive real components.The results demonstrate that circumstellar belt perturbations and radiation effects significantly influence equilibrium multiplicity and local dynamics in modified restricted four-body systems. The study highlights the importance of accurate numerical exploration and stability verification for identifying closely located equilibrium solutions. Further investigations involving nonlinear stability analysis and global dynamical evolution are required to extend the present results beyond the linear approximation.
Spatial Memory Behaviour of the Tower of Hanoi Algorithm on the Java Virtual Machine: An Analytical and Empirical Study
Authors: Omar Khan Durrani, Almas Sadaf Farooqi, Harshitha Sindhe
Abstract: Asymptotic analysis of recursive algorithms usually emphasises time complexity, leaving space underexamined on managed runtimes, where realised memory cost depends on object layout and collector behaviour as well as algorithmic structure. This study uses the Tower of Hanoi (TOH), whose optimal solution comprises 2^N-1 moves, as a controlled workload for characterising spatial performance in Java. Three paradigms are formalised: (i) pure recursion, with auxiliary space bounded by call-stack depth, O(N); (ii) an iterative variant with an explicit stack, relocating the same O(N) state to the heap; and (iii) a move-materialising variant retaining the full sequence, requiring Θ(2^N) space. Analytical footprint models are derived from frame size, object-header size, reference width, and alignment padding. The models are validated using the Java Runtime API, sampling live-heap occupancy after forced collection across increasing N until resource exhaustion. Repeated trials yield mean estimates, and model fit is assessed by regression on log-transformed data. Sensitivity to heap bounds, thread stack size, and collector choice is analysed to separate algorithmic effects from runtime effects. Asymptotic space classes are preserved on the JVM, but materialisation exhausts the heap at far smaller N than recursion, while deep recursion fails through StackOverflowError. Lazy move generation is recommended.
Investigations on the Replacement of Fine and Coarse Aggregates with Ceramic Tile Waste in M40 and M70 Concretes
Authors: G.poojitha, A.Ram kumar
Abstract: Industrial ceramic tile waste represents a persistent construction-related solid waste stream with potential for use as aggregate in concrete. This study evaluates crushed ceramic tile waste as partial replacement of natural fine aggregate and combined fine/coarse aggregates in standard-grade M40 and high-strength M70 concretes. Concrete mixtures were proportioned using the ACI method and evaluated through workability, density, compressive strength, split tensile strength, flexural strength, stress-strain response, modulus of elasticity, rebound hammer, ultrasonic pulse velocity (UPV), sorptivity, and chemical durability. Replacement trials covered 0-100% ceramic waste; selected mixtures were subsequently examined in detail. For M40 concrete, the 50% fine-aggregate replacement mixture achieved a 28-day compressive strength of 55.36 MPa, while the selected combined-replacement mixture achieved 52.10 MPa. For M70 concrete, the control, 50% fine-aggregate replacement, and combined-replacement mixtures recorded 28-day compressive strengths of 71.03, 67.36, and 63.12 MPa, respectively. UPV values for the investigated mixes were generally above 4.2 km/s, indicating good concrete quality. Sorptivity decreased with time, whereas exposure to 5% H2SO4 and 5% HCl produced pronounced mass and compressive-strength losses. The results demonstrate that ceramic tile waste can be incorporated as aggregate while maintaining useful mechanical performance at selected replacement levels, although higher replacement levels and aggressive acidic environments adversely affect performance.
DOI: http://doi.org/10.5281/zenodo.23206314
AI-Based Smart Health Care Monitoring Using IoT
Authors: M.Ajithaveni, M.Mahalakshmi, A.kavibala, G.Maharishi, M.Kavitha, M.siva
Abstract: Healthcare monitoring is an important area where technology can significantly improve the quality, accessibility, and efficiency of medical services. Traditional healthcare systems mainly depend on periodic hospital visits and manual monitoring of patients, which may not provide continuous information about a patient’s health condition. The combination of Artificial Intelligence (AI) and the Internet of Things (IoT) provides an effective solution for continuous and intelligent healthcare monitoring. In an AI-based smart healthcare monitoring system, IoT sensors are used to collect real-time physiological information such as heart rate, body temperature, blood oxygen saturation, blood pressure, and other health-related parameters. The collected data is transmitted through an IoT communication network to a cloud or processing platform. AI and machine learning techniques can then analyze the collected data, identify abnormal patterns, and support early detection of potential health problems. The proposed system provides continuous monitoring, automatic data analysis, alerts, and remote access to health information. This reduces the need for frequent manual observation and helps healthcare professionals make informed decisions based on real-time patient information. The system can be particularly useful for elderly people, patients requiring continuous observation, individuals with chronic health conditions, and remote healthcare environments. The integration of AI, IoT sensors, cloud computing, and mobile applications creates a smart healthcare environment that improves monitoring efficiency and supports timely medical intervention.
DOI: https://doi.org/10.5281/zenodo.23237531
AI-Based Student Performance Prediction Using Machine Learning
Authors: Mrs. Mercy Smilin, N. Vigneshwari, A. Naseeha, M. Madhushree, V. Pon Azhagi, P. Raghunanthan, I. Manikandan
Abstract: Student performance prediction is an important application of Artificial Intelligence in the education sector. Academic performance can be influenced by attendance, internal assessment marks, assignment performance, study habits, previous examination results, and other learning-related factors. Traditional evaluation mainly depends on examinations and manual analysis, which may not i dentify students who require academic support at an early stage. This research proposes an AI-based student performance prediction system using machine learning techniques. The system collects relevant student information, preprocesses the data, selects useful features, trains machine learning models, and predicts expected academic performance. Algorithms such as Decision Tree, Random Forest, Logistic Regression, and Support Vector Machine can be applied and compared. The models can be evaluated using suitable performance measures such as accuracy, precision, recall, F1-score, and confusion matrix. The proposed system can help teachers identify students who may need additional guidance and can support data-driven academic planning. The system is intended as a decision-support tool and not as a replacement for teachers. With reliable data and proper evaluation, machine learning can provide useful information for early academic intervention and improved learning support.
DOI: http://doi.org/10.5281/zenodo.23238306
Remaining Tool Life Monitoring System Based On Error And Machine Learning
Authors: Kush Joshi, Aryan Yardi, Anshika Goyal, S Ahmed Hamdan, Saravanakumar R
Abstract: – Unplanned tool failure in CNC machining leads to poor surface quality, scrapped parts and machine downtime. This paper addresses the design and implementation of a low-cost, real-time system for monitoring tool wear and estimating the remaining useful life (RUL) of cutting tools. The hardware architecture comprises an ADXL345 3-axis accelerometer mounted on the spindle housing, an electret microphone for acoustic emission sensing and an ACS712 Hall-effect sensor for spindle motor current. The signals are conditioned by a custom low-pass filter board and digitised by an Arduino Mega or an NI myDAQ. The software is developed in LabVIEW and Python and applies DC offset removal, Daubechies db4 wavelet denoising, windowed FFT and statistical feature extraction (RMS, kurtosis, crest factor, tooth-passing amplitude and sideband energy). Tool condition is classified as Healthy, Early-Wear or Critical using calibrated thresholds, and RUL is estimated by extrapolating the polynomial trend of the RMS feature, with machine learning models identified as the next stage of development. A real-time dashboard with alert thresholds presents the results to the operator. The system is being validated on a benchtop CNC router against four known tool conditions, with targets of at least 90 % classification accuracy, a false alarm rate below 5 % and an RUL error below 10 %. By enabling the early detection of tool wear, this framework is intended to support predictive maintenance, reduce unplanned downtime and extend tool life in automated manufacturing.
A Comparative Analysis of Machine-Learning Models for Plant Acoustic Stress Classification
Authors: S. Pradeepa Ph.d Scholar, Associate Professor Dr. Hari Prasad D
Abstract: In this paper, the authors develop a comparative framework for acoustic stress analysis of plants, which is based on a primary sensing modality, acoustic observations, and a secondary one, environmental measurements. The provided evidence includes a dataset containing 3,991 environmental records labeled with 10 classes and a performance comparison table for six different machine learning models: Support Vector Machine (SVM), Random Forest (RF), CatBoost, XGBoost, Improved XGBoost, and Stacking Classifier. The reported accuracies are 79.97%, 82.73%, 82.10%, 82.35%, 82.23%, and 82.60%, respectively. Due to the fact that the attached tabular data does not include the actual acoustic descriptors but rather environmental variables, these numerical results are explicitly reported as an environmental/contextual computational baseline, and are not considered to be proven acoustic-only performance. The manuscript thus offers a leakage-aware, step-by-step approach for gathering acoustic signals, reducing environmental noise, extracting acoustic features, separating acoustic and contextual information, training the six classifiers, and testing plant level generalization. The framework is designed to enable a subsequent controlled acoustic experiment with plant-level splits to independently label and assess the Healthy, Unstressed and Stressed states. The final manuscript keeps the provided results and also keeps a good separation between evidence measured and the proposed acoustic experiment.
DOI: https://doi.org/10.5281/zenodo.23237375
PriceSense: A Quick-Commerce Price Comparison Platform
Authors: Omkar Anand Sharma, Divij Nagpal, Akhand Pratap Singh Sengar, Devansh Shringi
Abstract: The rapid growth of quick-commerce platforms has provided consumers with faster access to groceries, household products, and everyday essentials. However, users often need to search across multiple platforms to determine where a product is available at the lowest price. This manual comparison process be-comes inefficient when product names, package sizes, availability, and prices differ across platforms.PriceSense is a web-based quick-commerce price comparison platform developed to consolidate product pricing information from Blinkit, Zepto, Swiggy Instamart, and BigBasket through a unified interface. The system uses a Flask backend and Playwright-based browser automation to interact with dynam-ically rendered web pages. Product information is processed us-ing product-name normalization, package-size extraction, token-based similarity matching, and sequence-based similarity calcu-lation.The updated system additionally incorporates user authenti-cation and location context. Users can register and log in using an email address and password, while location information can be provided through a delivery address or pincode and browser geolocation. Geographic coordinates can be reverse-geocoded to obtain a delivery postcode, after which the location context is passed to the scraping engine. Browser automation sessions are configured with the corresponding geographic coordinates.To improve data reliability, the system distinguishes directly extracted live prices from estimated and unavailable values. Estimated values are calculated only when at least two legitimate live price observations are available and are not used to determine the winning platform. The resulting comparison is presented through an interactive dashboard with platform-wise prices, summary indicators, and graphical visualization.
DOI: https://doi.org/10.5281/zenodo.23261119
Artificial Intelligence Emerging Trends In Libraries
Authors: Dr. Bezwada Ravi Kumar
Abstract: Artificial intelligence (AI) is one of the emerging trends and applications of computing in libraries. It involves programming computers to do things, which if done by humans, would be said to require intelligence. The ultimate promise of artificial intelligence in libraries is to develop computer systems or machines that think, behave, and in fact rival human intelligence, and this clearly has major implications on librarian ship. The application of artificial intelligence in the library has become pervasive. They include expert systems for reference services, book reading and shelf-reading robots, virtual reality for immersive learning among others. Although the incorporation of artificial intelligence in libraries can be perceived to alienate librarians from their users, it will probably help libraries do more rather than taking over the jobs of librarians. It will enhance their services delivery. Artificial intelligence will greatly improve library operations and services and will upgrade and heighten the relevance of libraries in an ever-changing digital society.
DOI: https://doi.org/10.5281/zenodo.23275770
IoT-Based Automatic Cotton Leaf Disease Detection, Infection Percentage Estimation, and Treatment Recommendation Using Raspberry PI
Authors: Deepak Ramkrishna Khadse, Dr. Pankaj H. Zope
Abstract: Cotton is one of the most important commercial crops worldwide, and its productivity is significantly affected by various leaf diseases such as bacterial blight, alter aria leaf spot, Cercospora leaf spot, powdery mildew, and leaf curl disease. Early and accurate identification of these diseases is essential to minimize crop losses and improve yield. Traditional disease diagnosis relies on manual field inspection by experts, which is time-consuming, labour-intensive, and often impractical for large agricultural fields. The proposed system integrates a Raspberry Pi5, a high-resolution camera module, environmental sensors, and wireless communication technologies to create an intelligent crop monitoring platform. Images of cotton leaves are captured using the Raspberry Pi camera and processed through computer vision techniques for leaf segmentation and feature extraction. A lightweight deep learning model based on MobileNetV2 is employed to automatically identify different cotton leaf diseases. In addition to disease classification, image segmentation techniques are used to calculate the percentage of infected leaf area, enabling the system to determine the severity level of infection. The system also collects environmental parameters such as temperature, humidity, soil moisture, and light intensity using IoT sensors. These parameters, along with disease diagnosis results, are transmitted to a cloud-based dashboard for real-time monitoring and analysis. Automated alerts can be sent to farmers through SMS or mobile notifications when severe infections are detected. Experimental evaluation is expected to achieve disease classification accuracy above 93%, infection area estimation error below 5%, and real-time inference within one second per image on the Raspberry Pi platform. The proposed solution offers a cost-effective, portable, and scalable approach for precision agriculture by combining artificial intelligence, image processing, and IoT technologies. The system enables early disease detection, accurate infection assessment, timely treatment recommendations, reduced pesticide usage, and improved cotton crop productivity, making it a valuable tool for smart farming applications.
DOI: https://doi.org/10.5281/zenodo.23280693
Digital Twin-Based Performance Optimization of Thermal Systems in Smart Manufacturing
Authors: Assistant Professor Pushpendra Kumar Sharma, Assistant Professor Surinder kumar
Abstract: One of the key issues associated with the intelligent manufacturing is the thermal management. The temperature homogeneity and efficiency define the quality of goods to be produced as well as the production costs. The authors suggest a novel intelligent hybrid digital twin architecture which allows for real-time optimization of the industrial thermal processes and integrates the physics-informed reduced-order model of thermal processes and the deep reinforcement learning algorithm for the closed-loop control of the system. There are two main problems that can be solved using this architecture; the first problem is the computational time required for the high-fidelity simulations while the second problem is the insufficiency of the necessary data required for the adaptive decision-making process in the physical system. Spatiotemporally reduced-order modeling is applied for the fast estimation of the temperature field on the basis of the data collected by the sensors while the deep Q-Network is controlling the temperature distribution according to several objectives.
Explainable Large Language Model for Intelligent Decision Support in Complex Domain
Authors: Associate Professor Dr.P.Jayalakshmi
Abstract: Despite showing great skills in working with unstructured data and generating appropriate answers, large language models (LLMs) cannot be applied in critical settings due to their lack of transparency and susceptibility to hallucination. The paper describes an approach to creating explainable LLM-based decision support systems, which includes breaking down complex questions into small subtasks that are processed via auditable pipelines. The approach employs retrieval-augmented generation along with domain-specific knowledge bases and computational instruments for the purpose of producing explanations and demonstrating intermediate reasoning steps while emphasizing data limitations. The approach is tested on clinical decision support problems for a multiple myeloma population. Comparing the results with those obtained when using monolithic LLMs as well as explainable machine learning approaches proves that structured decomposition leads to better decision traceability while achieving similar performance.
YOLOv26 For Automated Inventory Management: A Study On Dense Object Detection In Retail Environments
Authors: Hagenimana Samuel, Dr. Tang Hai
Abstract: : For retail to manage its inventory well, it needs to be able to accurately detect objects. However, shelves that are full of products make this difficult because they can block each other, overlap, and change size. This study is the first to look at YOLOv26, a new YOLO architecture that doesn’t need NMS for inference, uses Progressive Loss Balancing, and the MuSGD optimizer. It does this on the SKU-110K dataset, which has over 11,000 images and 1.7 million annotated objects. Inference runs at 158 frames per second, which means it can be used in real time on low-end hardware. Analysis shows that 77% of objects are correctly detected. Most of the errors come from false positives in the background and items that are heavily blocked. YOLOv26n is 22% better than YOLOv8 in mAP50 and gets close to bigger models with very little extra processing power. Integrating with ByteTrack for counting inventory shows that it can be used in real life, setting a strong standard for efficient, edge-capable retail automation.
The Physiology Foundation Of Human Sleep Mechanism Stages And Life Span Dynamic
Authors: Mukkamalla Sai Darshan Reddy
Abstract: Sleep is a fundamental physiological process essential for cognitive function, metabolic homeostasis, emotional regulation, and overall human health. This review examines the physiological foundations of human sleep, focusing on the neurobiological mechanisms regulating the sleep–wake cycle, sleep architecture, systemic physiological changes, lifespan dynamics, and environmental influences. Sleep regulation is primarily governed by the interaction between homeostatic sleep pressure (Process S) and the circadian rhythm (Process C), supported by coordinated neurochemical and hormonal pathways. The review further characterizes the physiological features and functions of non-rapid eye movement (NREM) and rapid eye movement (REM) sleep, including their effects on neural activity, autonomic function, cardiovascular regulation, metabolism, and memory. Polysomnography remains the gold standard for evaluating sleep architecture and physiological disturbances. Sleep patterns also undergo substantial changes across the lifespan, with aging associated with reduced slow-wave sleep and increased sleep fragmentation. Environmental factors, particularly thermal conditions and bedding systems, can further influence sleep continuity and restorative sleep stages. Conversely, sleep deprivation and fragmentation are associated with cognitive impairment, metabolic dysregulation, cardiovascular strain, and adverse systemic health outcomes. Overall, sleep represents a highly organized, dynamic physiological state that is essential for maintaining homeostasis and health. Future research should emphasize personalized sleep interventions and the integration of advanced wearable technologies for longitudinal assessment of sleep physiology.
DOI: https://doi.org/10.5281/zenodo.23312780
CAMP CHAMP – College Management System
Authors: Duraimurugan M, Omprakash S, Sakthivel R, Priyadharshini J
Abstract: Camp management involves organizing activities, communicating with participants, maintaining schedules, and coordinating resources. When these activities are managed manually or through disconnected communication channels, information may become difficult to maintain and access. CAMP CHAMP is proposed as a mobile application intended to support the organization and management of camp-related activities through a centralized digital platform. The proposed concept focuses on presenting relevant camp information, organizing schedules, facilitating communication, and improving access to important updates. This paper discusses the motivation, objectives, conceptual architecture, proposed methodology, functional requirements, data management considerations, and evaluation approach for CAMP CHAMP. The design emphasizes usability, maintainability, data protection, and the possibility of future enhancements. The actual implementation, technology stack, and performance must be evaluated using the completed application and its observed results.
DOI: https://doi.org/10.5281/zenodo.23334208
Epistemic Trust, Methodological Opacity, And Hallucination Risks In AI-Augmented Educational Scholarship: A Critical Literature Review
Authors: Debasish Sarkar
Abstract: The rapid integration of Generative Artificial Intelligence (GenAI) and Large Language Models (LLMs) into academic inquiry has triggered an unprecedented paradigm shift in educational research. While AI-augmented literature synthesis, automated data coding, and language refinement offer remarkable productivity gains, they simultaneously introduce profound epistemic and methodological vulnerabilities. This doctoral dissertation provides an exhaustive, critical literature review exploring the complex interplay among Epistemic Trust, Methodological Opacity, and Hallucination Risks in modern educational scholarship. Synthesizing extensive peer-reviewed literature published between 2020 and 2024, this study adopts a virtue epistemology framework to analyze how algorithmic cognitive offloading impacts scholarly rigor. The findings reveal a growing systemic tension between productivity acceleration and epistemic vigilance: heavy reliance on generative tools correlates with a decline in primary-source verification, increased vulnerability to algorithmic hallucinations, and an over-reliance on opaque “black-box” data interpretations. Furthermore, this study evaluates current institutional policy responses across international higher education frameworks, highlighting a critical deficit in systematic verification protocols and doctoral AI literacy training. To resolve these systemic vulnerabilities, this review formulates the Epistemic Verification and Transparency Protocol (EVTP), providing actionable guidelines for researchers, journal editors, and university ethics committees to preserve scholarly integrity in an increasingly automated academic landscape
DOI: https://doi.org/10.5281/zenodo.23324179
Multi-sensor Fault-Aware Road-Safety Monitoring For Powered Vehicle: A Review And Proposed Framework
Authors: Saravanakumar R, Aishali Datta Chowdhury, Aeindrila Roy, Ananya Rajesh
Abstract: Motorcycle and scooter riders are among the most vulnerable road users because they have little physical protection, limited rear-side awareness, and depend on timely emergency response after a crash. This paper reviews sensor-based Internet of Things (IoT) road-safety systems for powered two-wheelers, covering inertial crash and abnormal-motion detection, alcohol-vapour screening with low-cost gas sensors, blind-spot sensing, communication and alerting, and the detection of sensor and communication faults. It organises the field in a five-category taxonomy, compares the strengths and limits of the main sensing technologies, and examines recent patent filings to show how these functions are being combined in practice. The analysis points to recurring gaps: single-function designs, fixed thresholds, little reported validation, and few checks on whether the sensors and links themselves are working. To address these gaps, the paper proposes a conceptual fault-aware multisensor framework that validates data, confirms events using several conditions, and classifies sensor, communication and data-consistency faults before raising an alert. The framework has not yet been validated experimentally; a test protocol and open research problems are outlined.
DOI: https://doi.org/10.5281/zenodo.23338186
Research Paper On Optimal Allocation Of Delivery Partners To Minimize Food Delivery Time
Authors: Garima Goyal, Neti Thakkar, Shivani Goyal, Swastika Sakshi, Tavishi Jain
Abstract: Online food-delivery platforms must decide which vehicles serve which kinds of orders. This paper formulates that decision as a transportation problem and solves it with real operating data. Using 43,641 complete delivery records from the Metropolitan and Urban segments of a Zomato dataset (44 days, 11 February to 6 April 2022), delivery capacity by vehicle type (motorcycle, scooter, electric scooter) is treated as the supply, eight city-by-traffic conditions as the demand, and the historical mean delivery time of each vehicle in each condition as the unit cost. A dummy destination balances the table. The problem is solved with the North-West Corner, Least Cost and Vogel approximation methods, the optimality of the best starting solution is tested with the MODI method, and the result is confirmed in Excel Solver. On a typical day the optimal plan lowers total expected delivery time from 25,680 to 24,357 minutes, a reduction of 5.15% (27.12 to 25.72 minutes per order), by keeping motorcycles on low-traffic orders and giving scooters and electric scooters the congested ones. Over all 44 days the reduction is 5.06%. The saving remains positive under a 25% demand increase combined with a 10% workforce cut (2.87%) and falls to about 1.7% when partner capacity is set close to its observed average. Shadow prices show that an extra order of scooter or electric-scooter capacity saves about 2.3 and 2.8 minutes, while extra motorcycle capacity has no value on a typical day.
DOI: https://doi.org/10.5281/zenodo.23340443
Design of A Security-Aware Smart Home Automation Architecture With Mobile Application Control: A Software-in-the-Loop Evaluation
Authors: Kariola Basira Ogaga, Michael Oghale Ighofiomoni, Omokaro Idama, Friday Elohor Odoh
Abstract: Smart-home systems increasingly combine networked sensors, embedded controllers, software services, and mobile interfaces to make domestic devices observable and controllable. This paper presents the design of a smart home automation architecture with a mobile application and manual controls, emphasising low-cost architecture, responsive control, security-aware messaging, and graceful local operation. The proposed system adopts an ESP32-class controller at the device layer, Wi-Fi connectivity, sensor inputs for temperature/humidity, illumination, motion and gas condition, relay outputs for representative household loads, and a mobile interface for authentication, status monitoring, direct switching, automation rules, alerts, and event history. A layered Internet-of-Things and cyber-physical feedback perspective guided the architecture, while recent user-acceptance literature informed interaction choices such as state visibility, simple device grouping, and explicit manual override. Because physical deployment conditions vary with router quality, mains wiring, sensors, and actuators, the quantitative evaluation reported here is deliberately limited to a reproducible, host-local software-in-the-loop environment rather than field measurements: the command-processing API path and the deterministic automation-rule engine were implemented and exercised, while the ESP32 wiring, sensor mapping, and relay behaviour are specified as a build-ready design that has not yet been physically assembled or timed. Across 880 authenticated API requests covering one-, two-, four-, and eight-device command batches, mean response time ranged from 1.747 to 2.059 ms and the 95th-percentile response time ranged from 2.220 to 2.762 ms, with all requests completing successfully. Six negative API/security checks returned their intended error codes, and thirty deterministic automation-rule tests covering ordinary automation, safety conditions, and manual override all passed. The results demonstrate that the command-validation, authentication, and rule-evaluation logic underlying a mobile-controlled home platform is sound and computationally light, and they establish a reproducible software baseline against which a future hardware-in-the-loop study, connecting the specified ESP32 controller, sensors, and relays, can measure genuine end-to-end response time, energy behaviour, and field reliability.
DOI: https://doi.org/10.5281/zenodo.23363529
Nunnarri: An AI-Enabled Universal IoT Management Platform Integrating MQTT, Local Language Models, and Model Context Protocol
Authors: Harshar A T, Jebakumar R, Pranav K, Balaji V, Santhipriya .s
Abstract: The growing use of heterogeneous Internet of Things (IoT) devices creates operational challenges in device onboarding, telemetry collection, control, data persistence, and integration with intelligent services. This paper presents Nunnarri, a modular IoT management platform designed to bring these functions into a single deployable system. The public implementation combines a React and Vite dashboard, Node.js and Express API services, an MQTT/WebSocket gateway, two PostgreSQL database services, an Ollama-connected AI backend, and a Model Context Protocol (MCP) server. The architecture separates identity and access data from IoT telemetry, provides a virtual-pin abstraction for device interaction, and exposes selected device-management operations to compatible AI clients through MCP tools, resources, and prompts. The system is packaged as a multi-container Docker Compose application and includes firmware integration material for ESP8266-class devices. The contribution of this work is an implementation-oriented architecture that connects conventional IoT messaging and web management with locally hosted language-model assistance and a standardized tool interface. The repository documents the intended system components; however, quantitative claims about latency, throughput, anomaly-detection quality, reliability, and resource use require controlled evaluation. A repeatable evaluation plan is therefore defined for message delivery, end-to-end command delay, service recovery, AI-assisted task completion, and security controls
DOI: https://doi.org/10.5281/zenodo.23383258
Design and Analysis of HVAC System for a Residential Building Using BIM Framework – A Review
Authors: Mr. Anshul S. Pande, Dr. Tejpal C. Parshiwanikar
Abstract: Heating, Ventilation and Air Conditioning (HVAC) systems comprise spatially interconnected ducts, air terminals, equipment, fittings and accessories that must be coordinated with architectural, structural and other building services. Conventional two-dimensional Computer-Aided Design (CAD) drawings remain useful for drafting, but they can limit spatial understanding and make coordination difficult when multiple services occupy constrained ceiling and shaft zones. Building Information Modeling (BIM) supports the representation of building systems as three-dimensional, information-rich objects and enables model-based coordination and documentation. This review synthesizes literature on BIM-enabled HVAC workflows, CAD-to-BIM conversion, multidisciplinary MEP coordination and clash detection. Published studies indicate that CAD-based duct-model generation can be formalized, BIM coordination benefits from structured information and management processes, and automated clash results still require engineering interpretation. A practical workflow is proposed for the ENEC Staff Accommodation Unit-02 (SAU-02) case study using AutoCAD, Revit MEP and Navisworks. The workflow connects drawing interpretation, HVAC model development, federated coordination, clash classification and resolution, model verification, and documentation. The paper’s contribution is an application-oriented synthesis and evaluation framework rather than a new HVAC thermal-design method. Quantitative performance outcomes are not claimed here; they are to be established through project-based model checks and coordination records.
A Hybrid EfficientNet-B0 and Machine Learning Framework for Multi-Class Retinal Disease Classification on RFMiD
Authors: Md. Alamgir Hossain, Paritosh Chandra Ray, Md. Mehedi Hasan, Abdul Al Mahin, Ferdowsi Khatun
Abstract: Automated analysis of retinal fundus images can support scalable screening, but multi-disease classification remains difficult because retinal abnormalities may exhibit overlapping visual characteristics and clinically important classes are often underrepresented. This paper investigates a four-class retinal image classification framework using a strictly filtered subset of the Retinal Fundus Multi-Disease Image Dataset (RFMiD). The considered classes are No_Disease_Risk, diabetic retinopathy, age-related macular degeneration, and macular hole. Images with multiple disease labels, unselected diseases, or inconsistent annotations were excluded, resulting in 855 training, 294 validation, and 305 test images. Training data were balanced to 401 entries per class through sampling with replacement and stochastic augmentation. A pretrained EfficientNet-B0 with frozen parameters was used to extract a 1,280-dimensional representation from each image. Support vector machine, Random Forest, K-nearest neighbors, XGBoost, and CatBoost classifiers were subsequently evaluated. On the official test subset, CatBoost attained the highest accuracy of 0.9016, weighted F1-score of 0.9022, and macro-ROC-AUC of 0.9842. SVM achieved the highest macro precision and macro F1-score, at 0.9075 and 0.8628, respectively, while XGBoost produced the highest macro recall of 0.8774. Grad-CAM visualizations were generated for selected test images to examine the retinal regions associated with the model predictions. The results show that fixed deep representations can support effective retinal image classification with conventional learners, although the small number of unique ARMD samples limits the statistical reliability and generalizability of the findings.
Soil Intelligence 2.0: A Grounded, Validated, and Multilingual Multimodal Ai Framework for Smartphone-Based Precision Agriculture
Authors: Mrs. Divya C, Nandish K, Vishnu Prasad K, Dushyanth Gowda N, Nithin K.B
Abstract: In the current existing scenario, one of the main limitations to sustainable agricultural practices and crop production is the gradual degradation of soil quality. The expensive and time-consuming nature of traditional laboratory-based soil testing also poses a significant challenge for small- and marginal-scale farmers, consequently leading to inefficient soil utilization. In this study, Soil Intelligence 2.0- an artificial intelligence-based precision agricultural platform that leverages smartphone images for preliminary and real-time estimation of soil properties and cropping suggestion- is proposed to overcome this problem. The system integrates Vision Transformer (ViT) architecture with multimodal LLMs to evaluate the various soil visual properties such as texture, texture and pattern of the soil on an imageBased on these features, the system estimates important soil parameters such as nitrogen, phosphorus, potassium, and pH, and uses the estimated conditions to determine suitable crops. The platform further provides personalized agricultural recommendations through traditional, organic, and hybrid farming approaches. An AI-based agricultural assistant enables farmers to interact with the system and receive context-specific guidance related to crop selection, soil improvement, and cultivation practices. The proposed approach is designed to deliver preliminary soil insights within seconds using commonly available smartphones, offering a low-cost and accessible decision-support solution for precision farming. The system can assist farmers in making more informed field-level decisions while supporting efficient and sustainable agricultural practices.
DOI:
Heart Rate Monitoring And Emergency Alert System Using IoT
Authors: Mrs. Chethan Kumar T, Devaraj V M, Akash L, Thimmegowda G, Dhananjay M M
Abstract: The increasing need for continuous health monitoring has encouraged the development of Internet of Things (IoT)-based healthcare systems. This paper presents an IoT-based heart rate monitoring and alert system prototype using an ESP32 microcontroller and a MAX30102 optical sensor. The proposed system estimates heart rate and peripheral oxygen saturation (SpO₂) from sensor signals and displays the readings locally through an LCD or serial monitor. The ESP32 supports wireless data transmission over Wi-Fi to a Python Flask server, where readings can be stored in an SQLite database and accessed through a web dashboard. A configurable threshold-based mechanism is proposed to flag readings that fall outside predefined limits for further observation. The system aims to provide an affordable and modular approach to collecting, displaying, and reviewing physiological measurements. Its practical performance depends on sensor placement, motion, signal quality, network connectivity, and validation of the measurements. The prototype is intended for educational and preliminary monitoring purposes and is not designed to diagnose heart attacks or replace professional medical assessment.
DOI:
Effects of STEAM Education Practices on Problem Solving Ability of Elementary School Students
Authors: preeti Nara, Dr.Raj Kumar
Abstract: The study aims to examine the effects of STEAM Education Practices on elementary students’ problem-solving ability, and it conducted hands-on activities incorporating STEAM components for 7th-grade students. The study used a pre-test, post-test, control-group design. The research participants were students in a public school located in the Farmai, Sirsa district. The study included 80 elementary students. The researcher developed 30 science modules that cover eight units based on the NCERT syllabus. The Experimental treatment was applied for 30 days in two groups: experimental group A (STEAM hands-on activity practice and Control group B (through conventional method). The researcher obtained the data using the standardized Problem Solving Ability Test by L. N. Dubey. The data were analyzed by t-test. The results showed that hands-on, activity-based STEAM practice effectively improved problem-solving ability among elementary students
DOI: https://doi.org/10.5281/zenodo.23466990
Representing Folk Tradition And Cultural Practice In Tamil Cinema: A Qualitative Content Analysis Of Mandaadi (2026)
Authors: Priya Palanimurugan, Mr. R. Praveen Kumar, S Kalaiselvan
Abstract: This study examined the representation of folk tradition and cultural practice in the Tamil film Mandaadi (2026), directed by Mathimaran Pugazhendhi. It focused on the depiction of coastal life and competitive boat racing and explored how these elements were connected with community identity, maritime knowledge, collective participation and personal rivalry. The study adopted a qualitative content analysis approach and used publicly available film descriptions, reviews, trailer information and scholarly literature on intangible cultural heritage and fishing communities in Tamil Nadu. UNESCO’s framework for intangible cultural heritage provided the conceptual foundation for interpreting cultural practices as living traditions shaped by community recognition and transmission.The analysis identified boat racing as the central cultural theme in the film’s reported narrative and examined its potential significance as a symbol of community belonging, practical skill and collective pride. It also considered how cinematic storytelling may have connected sporting competition with interpersonal relationships and social identity. The study distinguished the film’s cultural representation from verified ethnographic evidence and avoided treating fictional events as direct records of real-world traditions. The findings suggested that Mandaadi offered a useful context for exploring the relationship between Tamil cinema and coastal cultural representation. However, the historical origins, local terminology and authenticity of the practices depicted required further verification through direct scene analysis, community interviews and independent historical sources. The study concluded that cinema could contribute to the visibility of cultural practices while also reshaping their meanings through dramatic storytelling. It recommended further research involving systematic film analysis and community-based evidence to establish a more comprehensive understanding of the relationship between the film and living cultural heritage.
DOI:
Structural Properties Of Exact Modules Under Change Of Rings
Authors: Dr. Sumit Kumar Dekate
Abstract: This paper is extension of my work in my previous paper titled “EXACT MODULES UNDER CHANGE OF RINGS’’ and “GENERALIZATION OF EXACT MODULES UNDER CHANGE OF RINGS” which studies the preservation of exactness of modules under change of rings. The principal results concern surjective ring homomorphisms and the transfer of exactness between module structures induced by such homomorphisms. The paper also records a converse-type result, introduces weakly exact modules, proves a criterion for balanced bimodules, and gives a finite-length criterion involving direct powers M^((n)) .To make the results more transparent for readers, explicit examples are included after the principal results. The present version replaces the earlier illustrations by new examples involving quotient rings, polynomial rings, finite fields, triangular matrix rings, and finite-dimensional modules.
BlindAssist: A Near-Real-Time Object Detection And Voice-Based Assistive System For Visually Impaired Users
Authors: Mr. Vaibhav Shrivastava, Nishchay Sakure, Chanchal Pagare, Md. Eliyas Sheikh, Sapna Debnath, Saurabh Sharnagat, Harsha Keche
Abstract: BlindAssist is a Python prototype that uses a laptop webcam and a YOLOv8 object detector to tell a visually impaired user, by voice, which objects are in front of the camera and roughly where they are. Frames are captured with OpenCV and passed to a COCO-pretrained YOLOv8 model (Ultralytics), and the results are spoken offline with pyttsx3. Each announcement gives the object’s class name and whether it lies to the left, ahead, or to the right, based on the horizontal position of its bounding box. For seven object classes an uncalibrated estimate of distance is computed from the box width and shown on screen; it is not spoken. Objects are tracked by class name, so that a class is announced when it first appears, when its direction changes, and once when it has been missing for three consecutive detection passes. Face modules using YuNet and SFace (OpenCV) and the DeepFace library recognize known people from reference photographs, and a DeepFace emotion model announces the dominant facial expression of each detected face; a custom pen-and-pencil model is under development. The paper describes the design, gives the algorithms and configuration, compares the system with published assistive systems, and describes the tests we plan to run. No labeled test set or user study has been completed, so no accuracy, speed, or user-study figures are reported.
DOI:
Artificial Intelligence And Creative Writing: Transforming Authorship In English Language And Literature
Authors: Dr V. V. Malinee, M.Miraclin Chella Vailankanni
Abstract: Incorporation of artificial intelligence in the process of creative writing entails a paradigmatic change in literary text production, assessment, and perception. In this research paper, the focus is made on an innovative role of large language models in the sphere of English language and literature authorship. Based on systematic analysis of recent empirical studies and theoretical approaches, three interrelated aspects will be discussed: interaction between human authors and AI systems, reception of artificial literature by the readership, and conceptualization of authorship in the digital age. For the purpose of this paper, the mixed-method approach involving both quantitative analysis of empirical studies and qualitative analysis of literary theoretical approaches will be used. As the results of the analysis indicate, the texts generated by AI are rated aesthetically the same way as texts written by human authors when there is no indication of the author’s identity; however, readers maintain biases against texts that are perceived to be machine-generated.
DOI: https://doi.org/10.5281/zenodo.23530821
AI-Driven Customer Sentiment Analysis For Brand Reputation Management
Authors: Pooja Dwivedi, Ashish Shukla
Abstract: With the rise in user-generated content generation on digital media, brand perception monitoring has become unfeasible using manual methods. In this paper, we present an AI-powered sentiment analysis approach that aims at reputation management of brands across social media platforms and e-commerce review channels. For this purpose, the proposed system adopts a hybrid deep learning architecture based on contextual embeddings generated by transformers and bidirectional recurrent networks for capturing sophisticated sentiments in informal text. To handle various forms of noise along with domain-specific linguistic patterns, multi-stage text preprocessing pipeline is employed. The proposed methodology uses aspect-based sentiment extraction to provide a fine-grained reputation analysis at the aspect level, not document-level sentiment polarity. The experiments conducted on a specially curated dataset of social media posts and product reviews demonstrate the ability of our framework to process real-time data streams in a consistent manner. Comparative analysis of the performance against existing state-of-the-art approaches highlights the benefits of hybrid models in dealing with sarcasm, implicit sentiment, and multilingual sentiment expressions.
DOI: https://doi.org/10.5281/zenodo.23530964
Block Chain Based Degree Verification System
Authors: Aditya Dhakarge, Ramsha Siddiqui, Priyatam More, Ms. Chetana Bhagat, Dr. Madhuri Kawarkhe
Abstract: Verification of academic degrees still relies on institutional records, manual checks, and inspection of documents, resulting in delays and leaving room for openings when a certificate is digitally altered. This paper introduces SOET VeriTrust, a blockchain-based system for verifying academic credentials, which combines deterministic cryptographic validation and automated visual forensics. The implemented system uses Ethereum Sepolia smart contracts, Keccak-256 Merkle-tree commitments, a packed revocation bitmap, an off-chain SQLite database, and a Python FastAPI forensic microservice.Student records are encoded as canonical cryptographic leaves, while only the Merkle root is anchored on-chain, thus preventing direct storage of student personally identifiable information. During verification, the application performs the database lookup, issuer authorization, Merkle proof validation,revocation check, and optional Error Level Analysis (ELA) with a PyTorch convolutional neural network for visual tampering. An eight-state diagnostic engine separates the successful verification from the claim mismatch, invalid proof, revocation, missing records, issuer failure, and visual forgery. Experimental evaluation with five test documents and repeated verification requests revealed fast deterministic verification, with full forensic analysis adding extra processing time. The reported batch anchoring and packed revocation results also reflect the practical reduction in persistent storage on the blockchain as compared to registering credentials individually. The findings suggest that the combination of cryptographic integrity and visual forensic analysis provide a more informative verification process than simply matching document hashes.
DOI:
Design And Implementation Of A Dual-Axis Solar Tracking And Real-Time Monitoring System Using Arduino, MATLAB, And NI LabVIEW-
Authors: Anshuman Singh, Shubham Manish Chhaparia, Anurag, Saravanakumar R
Abstract: Low-cost photovoltaic (PV) tracking prototypes are frequently reported as hardware demonstrations, with little attention to how their internal signals can be observed, compared, and reproduced. This paper presents the design and implementation of an Arduino Uno-based dual-axis solar tracking prototype that senses the direction of stronger illumination with four light-dependent resistors (LDRs) behind a cross-shaped divider, positions the panel with two hobby servos on a 3D-printed pan–tilt frame, and additionally monitors panel voltage and soil moisture, with a relay-switched pump load for threshold-based irrigation. The controller streams one nine-field comma-separated frame per cycle over USB serial. Two monitoring clients are described. A MATLAB dashboard reads the serial stream directly and plots the four LDR channels, panel voltage, soil moisture against dry and wet thresholds, and pump state in real time. An NI LabVIEW virtual instrument implements a virtual-sun model of the quadrant error calculation, deadband stepping, and a simulated solar-power estimate, and polls a local Flask HTTP endpoint through its HTTP client functions. We derive the quadrant error model, show that the LabVIEW model’s errors reduce to −12 and +12 times the angular offsets, and report an eight-point, 140 s observation window of the live dashboard in which the top-right LDR operated close to ADC saturation. No tracking gain, energy yield, or communication latency is claimed. Closed-loop convergence, fixed-versus-tracking comparison, and endpoint latency are instead specified as a reproducible test protocol and identified as future work. The paper therefore documents what was built, what was measured, and what remains to be validated, so that the prototype can be reproduced and extended.
DOI:
LabVIEW Based HVAC Building Management Energy Optimizer For Occupancy And Air-Quality Responsive Ventilation Control
Authors: Lokesh S, Sakthi Deepashika N, Dr.Saravanakumar R
Abstract: Heating, ventilation, and air-conditioning (HVAC) systems contribute significantly to total building energy consumption because conventional ventilation systems often operate continuously at fixed power levels regardless of real-time spatial demand. This paper presents the design, hardware implementation, and graphical validation of a low-cost, real-time HVAC Building Management Energy Optimizer that dynamically regulates ventilation based on room occupancy and environmental sensing. The hardware architecture utilizes an ESP32 microcontroller as the central control unit, an HC-SR501 passive infrared (PIR) sensor for occupancy detection, an MQ135 analog environmental sensor, an L298N driver module for Pulse-Width Modulation (PWM) fan control, a 6 V DC geared motor representing the ventilation fan, and an SG90 servo motor serving as a motorized HVAC airflow damper. Laboratory Virtual Instrument Engineering Workbench (LabVIEW) communicates with the ESP32 via a USB serial (VISA) interface to provide graphical real-time data acquisition, parameter visualization, decision monitoring, and dynamic power/energy estimation. The system employs a rule-based control strategy where occupancy acts as the primary gate: when unoccupied, fan speed and damper position are set to 0% and 0°, respectively. Under occupied conditions, ventilation is scaled across three demand tiers—30% (30° damper), 60% (60° damper), and 100% (90° damper)—based on analog air-quality threshold boundaries. Experimental validation confirms the full hardware-software execution chain, establishing an accessible, transparent prototype for demand-controlled ventilation (DCV) and building energy optimization
DOI:
LabVIEW-Based Adaptive Environmental Noise Monitoring And Source Classification System
Authors: Aniritha S, Rithanya S, Ajissa Jasmin, Dr. Saravanakumar R
Abstract: Environmental noise monitoring systems commonly focus on sound-level measurement, while source-aware interpretation and adaptive detection are often treated as separate functions. This work presents a low-cost LabVIEW-based framework for real-time environmental noise monitoring and source classification. An electret microphone and LM358-based signal-conditioning stage acquire the acoustic signal, while an Arduino Nano provides analog-to-digital conversion and serial communication with LabVIEW. The software computes the root-mean-square (RMS) magnitude, calibrated sound level, fast Fourier transform (FFT), equivalent continuous sound level (Leq), maximum sound level (Lmax), and duration of threshold violation. Instead of using only a fixed alarm level, the proposed monitoring strategy estimates the local background-noise condition and adapts the detection threshold to reduce unnecessary alerts. Frequency-domain features are then used to distinguish representative environmental sources such as traffic/engine noise, horns, speech, music, and construction noise. A LabVIEW dashboard provides real-time visualization, alerts and data logging. The hardware prototype has been integrated and the Arduino-LabVIEW interface and experimental validation are being completed. The final evaluation will compare the prototype against a reference sound-level meter and quantify measurement error and classification performance
DOI: https://doi.org/10.5281/zenodo.23570976
Smart Water Quality Monitoring System _392
Authors: Musab Shaikh, Anas Ansari, Kamish Khan, Dr. Madhuri Kawarkhe, Mr. Vishal Vijay Saswade
Abstract: Checking whether water is safe to drink normally means collecting a sample and sending it to a laboratory, and the answer can take hours or even days to come back. This paper describes a small Smart Water Quality Monitoring System that gives a quick first indication on the spot instead. An ESP32 microcontroller reads an analog Total Dissolved Solids (TDS) sensor and an analog turbidity sensor, converts their output voltages into ppm and NTU values, and places the sample in a Safe, Borderline or Unsafe band. The verdict is shown on a 16×2 LCD attached to the device and can also be sent to an online dashboard for later viewing. All parts are joined with jumper wires on a breadboard, so the prototype is easy to rebuild or change. The testing plan compares the device against a handheld reference meter using several kinds of water, and the paper explains how that comparison is carried out and reported. The system is meant as a screening aid and is not a replacement for certified laboratory analysis.
DOI:
Development of A LabVIEW-based Electronic Nose Using ESP32 And Machine Learning for Food Spoilage Detection
Authors: Suraj Kalyanaraman, Nandish S, Aswin N, Dr Saravanakumar R
Abstract: The electronic nose (E-Nose) provides a rapid and non-destructive method for evaluating the quality and safety of food products. In this, we try to address the design and implementation of a low-cost E-Nose for detecting the extent of food deterioration. The hardware architecture comprises an array of metal oxide semiconductor (MOS) gas sensors, including MQ-135, MQ-4, and MQ-2 sensors, which are used to detect volatile organic compounds (VOCs), carbon dioxide (CO₂), methane (CH₄), and ammonia. To ensure high precision, the ADS1115 16-bit ADC is employed, along with the DHT22 sensor for environmental correction. The array is integrated into an acrylic sealed compartment equipped with an ESP32-S3 Xiao microcontroller and an active cleaning fan. The software architecture is constructed using NI LabVIEW and incorporates producer-consumer design patterns and event-based state machine designs. During classification, differential voltages and slopes are extracted and transmitted through a Python node to principal component analysis (PCA) and k-nearest neighbour (k-NN) classification algorithms. This system demonstrates high accuracy in classifying different thresholds of deterioration and offers the food industry a real-time, scalable, and highly efficient surveillance solution. By enabling the early detection of spoilage, this framework directly contributes to environmental sustainability by reducing food waste and lowering the carbon footprint of the agricultural supply chain
DOI:
A Real-time Multimodal Human Fatigue Monitoring System Using GSR And Vision-based Facial Analysis
Authors: shivraahul B, vishal Sanjay R, Dr. Saravanakumar R
Abstract: Human fatigue and drowsiness can reduce attention and task performance, motivating the development of real-time monitoring systems that combine behavioral and physiological indicators. This paper presents a low-cost prototype that combines galvanic skin response (GSR) sensing with vision-based facial analysis. An R-1085 GSR sensor is interfaced to an ESP32 for analog signal acquisition and USB serial transmission. A conventional laptop camera provides facial images, while Python, OpenCV and MediaPipe facial landmarks are used to estimate eye aspect ratio (EAR) and identify prolonged eye closure. LabVIEW 2022 is proposed as the supervisory monitoring layer for visualization and alert presentation, with a buzzer available for warning output. Initial GSR experiments using 35 finger-contact samples produced values from 2242 to 2270 (mean 2262.4; standard deviation 4.75),whereas 37 no-contact samples ranged from 2366 to 2490 (mean 2400.38; standard deviation 23.81). The camera prototype produced EAR values of 0.39 for an open-eye frame and 0.19 for a closed-eye frame. These preliminary results demonstrate operation of the individual sensing components. The prototype is intended for research and educational evaluation; the measured thresholds are not clinically validated fatigue criteria.
DOI:
Hardware-in-the-Loop PID-Based Speed Regulation Of A DC Motor Under Variable Mechanical Load
Authors: Hamand Srekand, Suman T, Dr. Saravanakumar R
Abstract: This study presents a Hardware-in-the-Loop (HIL) closed-loop speed regulation of a physical DC motor using a LabVIEW-based PID controller. In DC motors, maintaining the desired speed under changing mechanical loading is an important control problem in electromechanical systems, where variations in load torque result in speed deviations and transient disturbances. To overcome these issues, mathematical modelling and software simulation are useful for controller development; experimental evaluation with a hardware setup is necessary to account for practical effects associated with motor dynamics, sensing, actuation, and mechanical loading. An optical quadrature encoder provides real-time shaft speed feedback, and a variable friction brake is employed to introduce controlled mechanical load disturbances. A discrete-time mathematical motor model is incorporated as a supporting simulation tool for controller development and analysis, while the physical DC motor remains the experimental plant. The system is evaluated under various mechanical-load conditions with speed references. Motor-speed response, tracking error, controller output, and actual performance metrics are analysed to assess reference tracking and disturbance-rejection behaviour. The proposed setup provides an experimentally oriented platform for investigating closed-loop DC motor speed regulation under controlled physical loading and establishes a basis for systematic controller tuning, experimental validation, and further development of HIL-based electromechanical control systems.
DOI:
Battery Thermal Management Systems For Electric Vehicles: A Review Of Liquid Cooling And Mist-Assisted Approaches
Authors: Sankalp Sonkusare, Dr. Ritesh Sambare
Abstract: Lithium-ion batteries generate heat during charge and discharge, and their performance, ageing and safety depend on temperature and temperature uniformity. This review synthesizes foundational and recent literature on battery thermal management systems (BTMS), with emphasis on indirect liquid cooling, serpentine and microchannel passages, and evaporative mist cooling. The literature shows that cooling performance depends on the battery heat load, cell-to-coolant thermal contact, channel geometry, coolant conditions, heat-exchanger capacity and auxiliary power. Serpentine channels can improve heat transfer coverage, but their hydraulic resistance and pumping demand must be considered alongside peak temperature and cell-to-cell temperature spread. Mist cooling can add evaporative heat rejection, while its outcome depends on humidity, airflow, droplet evaporation and water carry-over. Findings from the two papers central to this project—a 2024 mist-cooling study and a 2026 serpentine microchannel experiment—are therefore interpreted within their own test boundaries rather than treated as direct performance forecasts. The review applies this synthesis to a proposed laboratory system using a copper serpentine-tube liquid loop and optional mist introduced at the radiator air inlet. Since assembly and testing are pending, the manuscript reports no experimental result for that system. It recommends matched test cases and combined thermal, hydraulic and electrical measurements to establish whether additional cooling justifies pump, fan and mist-system energy use.
DOI: https://doi.org/10.5281/zenodo.23594498
The Narmada Valley Project
Authors: Karthika. A. R
Abstract: The Narmada Valley Project, one of India's largest river valley initiatives, aims to provide irrigation, electricity, and economic growth through dams like the Sardar Sarovar Dam. However, the relocation of over two lakh people, mainly from rural and tribal communities, has led to significant human costs. Although constitutional guarantees and the 2013 Land Acquisition Act exist, rehabilitation efforts remain inadequate and inconsistent. This study adopts a qualitative and doctrinal approach, drawing upon statutory frameworks, landmark judicial decisions, environmental assessments, and civil society narratives to critically examine the legal and social consequences of the project. Through the lens of key legal cases, including Narmada Bachao Andolan v. Union of India, and stakeholder perspectives, the paper critiques the contradiction between state-led development imperatives and constitutional rights.
DOI: https://doi.org/10.5281/zenodo.23609946
Design And Simulation Of A Seismic Building Vibration And Damage Monitoring System Using LabVIEW
Authors: Bhavya Ajit Agarwal, Toushif Ahmed Khan, Krishnanshu Agrawal, Shreyansh Choudhary, Dr. Saravanakumar R
Abstract: Continuous structural health monitoring is vital for assessing civil infrastructure integrity under seismic excitations and environmental degradation. Traditional manual visual inspections are periodic, qualitative, and incapable of capturing transient dynamic responses. This paper presents a virtual instrumentation framework developed in LabVIEW for real-time seismic vibration simulation, dynamic response monitoring, and structural damage detection. A base-excited single-degree-of-freedom mass-spring-damper model is implemented to simulate structural dynamics under ambient shaking, resonance, sudden stiffness reduction, structural tilt, and pseudo-random earthquake ground motion. The virtual instrument tracks critical dynamic indicators including fundamental natural frequency, roof-to-base transmissibility ratio, instantaneous roof acceleration, and tilt angle, triggering a multi-criteria threshold alarm logic upon abnormal state detection. Simulation findings demonstrate that a 48.5% loss in lateral stiffness (from 1300 N/m to 670 N/m) induces a 0.51 Hz downward shift in natural frequency (from 1.81 Hz to 1.30 Hz), confirming frequency tracking as a sensitive damage indicator. A proposed hardware acquisition architecture incorporating dual MEMS accelerometers (ADXL345 and MPU6050) interfaced via microcontroller serial streaming to LabVIEW VISA is documented. This prototype provides an educational testbed for structural dynamics and non-destructive damage diagnosis.
DOI:
A Computationally Efficient Alternative To MPC For Real-Time Torque Vectoring In 4WID EVs
Authors: Pallab Basak, Wang Bao Hua
Abstract: Four-Wheel Independent Drive (4WID) electric vehicles offer unprecedented authority over individual wheel forces, enabling propulsion-based yaw stability control through torque vectoring. Model Predictive Control (MPC) has emerged as the dominant academic paradigm for multi-objective vehicle dynamic control owing to its constraint-handling capability and predictive horizon structure. However, the online solution of constrained quadratic programs incurs O(n³) computational complexity with respect to the prediction horizon length, rendering real-time deployment on standard automotive embedded controllers impractical for most production-grade applications. This paper proposes an architecture comprising a Sliding Mode Control (SMC) yaw stability controller integrated with a two-dimensional (2D) motor efficiency map (η-map) based torque allocator that resolves the corrective wheel torque distribution problem in O(1) time via direct memory lookup. The SMC formulates a sliding surface combining yaw rate error and sideslip angle error, deploying a boundary-layer saturation function to eliminate chattering while preserving robust stability margins under nonlinear tire dynamics and variable road friction. The efficiency-aware torque allocator pre-computes optimal torque-speed operating regions offline and retrieves them via a 2D lookup table at each control step, achieving multi-objective performance simultaneous stability enhancement and energy efficiency improvementwithout requiring any online optimization. Comparative MATLAB/Simulink evaluation under a standardized step-steer maneuver at 60 km/h on a dry surface (μ = 0.85) demonstrates that the proposed SMC reduces yaw rate root-mean-square error (RMSE) by 51.2% relative to the proportional (P) baseline, by 33.5% relative to PID, and by 37.6% relative to the Linear Quadratic Regulator (LQR), while consuming only 139.8 kJ compared to the highest-consumption controller at 142.7 kJ. Theoretical execution time analysis confirms that the proposed architecture is compatible with 1 ms embedded controller cycle times on standard 32-bit automotive-grade microcontrollers, a target that MPC-based alternatives cannot reliably meet without dedicated co-processor hardware.
DOI: https://doi.org/10.5281/zenodo.23612861
CRISPR-Based Rapid Diagnostics for Urinary Tract Infections: Detection of Uropathogens and Antimicrobial Resistance
Authors: Assistant Professor Apurva, Assistant Professor Dr. Bhawna Negi, Assistant Professor Swati Trivedi, Assistant Professor Iti Sharma
Abstract: Urinary tract infections (UTIs) are among the most common bacterial infections and contribute substantially to antimicrobial use in community and hospital settings. Conventional diagnosis relies on clinical assessment, urinalysis and urine culture, with culture-based identification and antimicrobial susceptibility testing often requiring 24–72 hours. This delay can encourage empirical antibiotic prescribing and may be particularly problematic when infections are caused by resistant or less commonly detected organisms. Clustered regularly interspaced short palindromic repeats (CRISPR) and CRISPR-associated (Cas) proteins have emerged as programmable nucleic-acid recognition tools that can be coupled with amplification methods and simple signal readouts. Cas12-based systems detect target DNA and can generate collateral cleavage of reporter molecules, whereas Cas13-based systems recognise RNA targets and can provide an alternative route for rapid detection. In UTI applications, these platforms may support identification of common uropathogens, including Escherichia coli, and detection of selected antimicrobial resistance determinants. Early studies have demonstrated rapid genotypic resistance testing and urine-to-result workflows, while newer approaches are exploring direct pathogen identification and susceptibility assessment from urine. However, CRISPR assays do not automatically replace culture: detection of a resistance gene does not always predict the full resistance phenotype, and absence of a targeted gene does not prove susceptibility. Sample preparation, target coverage, mixed infections, contamination control, clinical validation, cost and regulatory requirements remain important challenges. This review summarises the biological basis of CRISPR diagnostics, current approaches for uropathogen and resistance-marker detection, potential clinical applications, limitations and priorities for translation into routine diagnostic services.
Microgrid Battery Manager: A Labview-Based Virtual-Instrumentation System for Multi-Cell Li-Ion Monitoring, State-Of-Charge Estimation and Protection
Authors: Amishi Pandey, Kushagra Shukla, Ilangganiyan Rajkumar, Sourya Pradhan
Abstract: Li-ion packs in microgrid energy-storage units need continuous supervision, because cell imbalance, overcurrent and overheating shorten pack life and create safety risks. This paper presents a Microgrid Battery Manager built as a LabVIEW virtual instrument on an Arduino platform. A four-cell 18650 series pack (14.8 V nominal, 2.6 Ah) is scanned through a CD4051 analog multiplexer that carries four cell-voltage taps, two NTC thermistors and one shunt-based current channel. The host VI estimates State of Charge (SoC) by Coulomb counting, corrected by open-circuit voltage (OCV) whenever the pack is at rest, and uses nested Case Structures to classify overcurrent, over-temperature and cell-imbalance faults. Solid-state switches perform passive balancing and automatic load disconnect, and a front-panel dashboard shows SoC, cell voltages, temperatures and protection status. The system is validated against four targets: ±10 mV cell-voltage accuracy, SoC error within ±5 %, cell spread below 20 mV after balancing, and an overcurrent trip in under 100 ms.
DOI: https://doi.org/10.5281/zenodo.23614387
Arduino Based, Bluetooth Controlled E-Vehicle
Authors: Vishwa Bharathi S Assistant Professor, Sonu Kumar, Bs Mani Kumari, Uzma Ummehani, Susheel Kumar
Abstract: An Arduino-based Bluetooth-controller electric vehicle (E-Vehicle) intended for wireless operation and efficient transportation of moderate loads. The proposed system uses and Arduino UNO microcontroller as the central control unit, an HC-05 Bluetooth module for wireless communication, motor drivers for controlling the drive motors, and rechargeable batteries as the power source. The vehicle can be operated through a Bluetooth-enabled mobile device, allowing the user to control its basic movements such as forward, backward, left, right. the Arduino receives control commands from the mobile device through the Bluetooth module and generates appropriate signals for the motor drivers to control the vehicle. The system is designed with multiple DC motors to provide sufficient traction and stability while carrying a specified load. The developed prototype demonstrates the integration of embedded systems, wireless communication, motor control, and electric mobility technologies, providing a foundation for further development of smart and autonomous electric vehicles.
DOI: https://doi.org/10.5281/zenodo.23647989
Virtual Instrumentation-Based Simulation and Monitoring of A Control Valve in A Water Flow System Using ESP32 and MATLAB
Authors: Ishayu Basu, Adhrit Ghosh, Ananya Misra, Dr. Saravanakumar R
Abstract: This paper presents a low-cost virtual instrument for observing the operation of a manually actuated control valve in a water flow system. Two digital vibration sensors are clamped to a pipe section near the valve and connected to the GPIO pins of an ESP32 microcontroller. The ESP32 samples both outputs, attaches a millisecond timestamp to every sample and streams the result as comma-separated text over a USB virtual serial port at 115200 baud. A MATLAB application, named AQUAGUARD, reads and parses the stream and presents a software front panel consisting of a status card for each sensor, two live strip charts over a 15-second moving window and a footer line that summarises the combined sensor state. The rationale is that flow through a partly open valve generates turbulence, which excites pipe-wall vibration that a pipe-mounted sensor can detect as a change of state. Because the sensors give only a binary output, the instrument reports a change in vibration state rather than a flow rate, and no leak diagnosis is claimed. With the valve closed, the dashboard consistently received Sensor 1 = LOW and Sensor 2 = HIGH, which confirmed correct acquisition, serial parsing and display. A simulation mode of the dashboard, driven by synthetic data, is also used to demonstrate the panel behaviour for every combination of sensor states and for pulsed inputs. We also identify a limitation of the present firmware, which transmits only when at least one sensor is HIGH and therefore leaves a stale reading on the panel when both sensors fall LOW, and we describe the correction. The paper documents the architecture, hardware set-up, firmware, software design, operating procedure and a roadmap towards analog vibration sensing, data logging and closed-loop PID valve control.
DOI: https://doi.org/10.5281/zenodo.23658843
Structural And Thermal Analysis Of A V6 Engine Cylinder Block: A Review
Authors: Jayshree Nakhate, Mrs. Seema Rewatkar Assistant Professor
Abstract: The cylinder block is the main structural component of an internal combustion engine. In a V6 engine it carries six cylinders arranged in two banks, supports the crankshaft and contains the coolant passages. During operation it is exposed to combustion pressure, thermal loads, vibration and mechanical forces, which produce stress, deformation and temperature variation in different regions of the block. Testing different block designs on physical prototypes is costly and time consuming, and Finite Element Analysis (FEA) is therefore widely used to predict structural and thermal behaviour before manufacturing. This review examines published research on thermal deformation of cylinder blocks, coupled thermal and mechanical analysis, FEA-based material comparison, residual stress in engine blocks and thermal management of multi-cylinder engines. The reviewed literature shows that temperature distribution, thermal loading and material properties strongly influence stress, deformation and dimensional stability. However, most studies treat structural and thermal behaviour separately, concentrate on single-cylinder, inline or diesel configurations, or compare materials only thermally. Integrated structural and thermal comparison of Aluminium Alloy and Grey Cast Iron for a V6 cylinder block under identical conditions is limited. Based on these findings, the proposed project develops a 3D CAD model of a V6 cylinder block and analyses it in ANSYS using static structural and steady-state thermal analysis.
DOI: https://doi.org/10.5281/zenodo.23659515
Evaluation of Welding Techniques for Fabricating Corrosion-Resistant Marine Components
Authors: Andrew John, Siedougha Clement Afoegba, Samuel Bagudu Elkanah
Abstract: The fabrication of corrosion-resistant marine components is essential for improving the durability, safety, and reliability of ships, offshore structures, and marine equipment exposed to aggressive seawater environments. This review evaluates welding techniques used in fabricating marine components, focusing on their influence on corrosion resistance, mechanical properties, weld quality, and fabrication cost. The welding processes examined include gas tungsten arc welding, gas metal arc welding, shielded metal arc welding, flux-cored arc welding, submerged arc welding, and friction stir welding. The review discusses commonly used marine materials, including carbon steel, stainless steel, duplex stainless steel, aluminium alloys, copper-nickel alloys, and nickel-based alloys. It also examines major corrosion mechanisms, such as pitting, crevice, galvanic, intergranular, and stress corrosion cracking. Furthermore, the effects of welding parameters, heat input, filler-metal selection, and microstructural changes on welded-joint performance are highlighted. Corrosion evaluation methods, including immersion testing, salt spray testing, potentiodynamic polarization, and electrochemical impedance spectroscopy, are discussed. The review emphasizes that welding performance depends on material selection, process control, surface preparation, and post-weld treatment. It identifies research gaps concerning standardized comparative testing and long-term seawater exposure. Overall, an integrated assessment of corrosion resistance, mechanical strength, productivity, and lifecycle cost is necessary for selecting suitable welding techniques for reliable marine component fabrication.
A Recurrent Deep Learning Framework for Seismic Wavefield Separation in Ocean-Bottom Node Surveys
Authors: Abubakar Ubaidu Ahmed, Zhong Yu
Abstract: Accurate separation of upgoing and downgoing seismic wavefields is essential for improving subsurface imaging in ocean-bottom node (OBN) surveys, where recorded wavefields may overlap and complicate seismic data interpretation. This study proposes a deep learning framework based on stacked long short-term memory (LSTM) networks augmented with batch normalization (BN) and dropout regularization for the automated, data-driven of reflected P-wave wavefields recorded by OBseparation N sensors. The network is trained on a synthetic dataset generated from a realistic 1-D velocity model, with supervised labels derived from one-way wave-equation modelling. Training converges at epoch 38 with a best validation mean squared error (MSE) of 0.049293, corresponding to an R² of 0.9508. Quantitative evaluation demonstrates consistent superiority over conventional f-k filtering and competing deep learning baselines including multi-layer perceptrons (MLP), one-dimensional convolutional neural networks (CNN-1D), and gated recurrent unit (GRU) networks. An ablation study confirms the additive contribution of each architectural component. Three-dimensional visualization of predicted reflected P-wave amplitudes reveals physically coherent wavefield geometry consistent with theoretical expectations. The proposed methodology offers a computationally efficient and geologically interpretable alternative to established separation workflows, with direct applicability to real-world OBN processing pipelines.
DOI: https://doi.org/10.5281/zenodo.23662723
More Alerts, Less Prevention? A Computational Study of Hazard Detection, Response Capacity and Threshold Choice in Occupational Health and Safety
Authors: Ashish Mulraj Ved, Assistant Professor Ravindra Kumar
Abstract: AI-enabled hazard detection is usually judged by whether it finds risky conditions. Prevention requires another step: somebody must have the time, authority and competence to respond. This article evaluates how that second step alters the interpretation of predictive safety performance. It reports a factorial computational experiment derived from a doctoral study of AI in occupational health and safety (OHS). Across 54 simulated operating conditions, five alert thresholds and 200 replications per condition, a score-based detector classifies 5,000 opportunities per replication. The design varies hazardous-opportunity prevalence (1%, 5% and 15%), score separation (1, 2 and 3), drift (0 and 0.75) and nominal response probability (0.40, 0.65 and 0.90), with response capacity limited to 500 alerts. At the reference threshold of 1.5, one 5%-prevalence, separation-2, no-drift condition yields mean residual hazardous opportunities of 180.8, 138.2 and 95.7 across the three response assumptions. With the same detector and nominal 65% response, imposing drift raises mean residual opportunities from 138.2 to 185.5. A sensitivity calculation also changes the preferred threshold when the relative weight of unresolved hazards and false alerts changes. The simulation shows why sensitivity, precision, workload and verified action belong in the same OHS decision model. All numerical data are computer-generated. The study does not report a deployed machine-learning system, workplace incidents, observed worker behaviour or injury reduction. Its contribution is a transparent stress test of the management choices that connect detection to prevention.
Integrated Solar–Piezoelectric Hybrid Energy Harvesting with IoT-Based Monitoring: A Comparative Analysis of Reported Systems and a Unified Architecture
Authors: Y Santhosh Kumar Associate Professor,, B Nagaveni, Ruksar Fatima, Sachik Standly, Rohit
Abstract: Hybrid energy harvesting that combines photovoltaic (PV) generation with piezoelectric footstep harvesting promises uninterrupted low-power supply in public spaces, but published prototypes use different tile designs, converters, monitoring schemes and reporting conventions, which makes them hard to compare. This paper critically consolidates three recent studies: an Arduino/ESP8266 solar–piezoelectric system with ThingSpeak monitoring, a footstep floor tile with a closed-loop buck converter, and a tracked-PV plus 48-sensor footstep system for DC loads. Their reported figures are normalised to common metrics (power per unit area, daily energy, and the piezoelectric share of total energy), and inconsistencies in the source data are identified. A unified architecture is then proposed in which a PV string and a piezoelectric tile array feed a common 12 V DC bus through independent converters, with a microcontroller and IoT gateway providing real-time monitoring and fault flagging. Using only values reported in the source studies, an analytical scale-up shows that a 30-tile array (60 Wh over 8 h) can supply a 5.4 W camera for about 11 h, but contributes only about 3–5% of the daily energy of a tracked PV panel of the class reported. The results indicate that piezoelectric tiles are best positioned as a supplementary and sensing-grade source, while PV remains the dominant energy contributor. Experimental validation of the proposed architecture is identified as future work.
DOI: https://doi.org/10.5281/zenodo.23665143