Volume 14 | Issue 4 |

 Abstract

Abstract The study was carried out to understand the distribution patterns and habitat preferences of Spotted Deer (Axis axis) within the Kuvempu University campus, situated in the Bhadra Tiger reserve, Shivamogga District, Karnataka, India. Employing a multi-method approach, field surveys were conducted for a span of 6 months from January 2025 to June 2025, with methodologies like, line transect sampling, camera trap deployment, and indirect sign identification, across diverse habitat types. Out of 1247 total photographic captures from the camera trap dominated ones were the spotted deer with 26%. The density was high along the semi disturbed habitat with 14.15 compared to forest and the disturbed habitat. Group size varied from 1 to 22 individuals. Camera trap analysis documented peak activity periods during early morning from 05:00-08:00 hours and late afternoon from 04:00-06:00 hours, consistent with crepuscular behaviour patterns. Threat assessment revealed the presence of potential predators like tiger, leopard, dholes and the major effect was observed from the stray dogs. The findings reveal a pronounced association between habitat heterogeneity and Spotted Deer presence and abundance. As the spotted deer has the high ecological significance as a keystone herbivore species and primary prey base, conservation of healthy Spotted Deer populations contributes substantially to broader landscape-scale ecosystem integrity. These results underscore the critical importance of maintaining structurally complex, heterogeneous habitats and minimising disturbances for the effective conservation and management of this keystone ungulate species in shared landscapes.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Keywords: Axis axis, chital, camera trapping, line transect, habitat preference.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Illicit arms trafficking shows up as one of the biggest danger to national and international security in the today's era. Estimated cautiously between $1.7 billion and $3.5 billion within one year, the illegal trade in small arms and light weapons (SALW) propels Rebellions, violence, Criminal syndicates, and Hostilities all across the world. India, with its strategic location in the heart of South Asia and closely bordered by unstable states and Penetrable borders, faces severe problems from this Happening . This research paper tries to do a all round synopsis of India's statutory, departmental, Functional and Active responses to illegal arms trafficking along with comparing it to contemporary global legislation, including the UN Arms Trade Treaty (ATT), the UN Programme of Action on Small Arms and Light Weapons (UNPoA), the Firearms Protocol, and the UNTOC. It tries to check on statutes principally the Arms Act, 1959 (as amended in 2019), the UAPA, and the Bharatiya Nyaya Samhita, 2023, relevant from India's domestic view point and investigates on the loopholes that exists in statutory recommendation and real life circumstances. The paper suggests on how despite having a strong legislative framework, robust institutional support, brave border security, and digital surveillance system, India still need a new, comprehensive and more grounded plan to secure it's National integrity and security along with global interactions.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Illicit Arms Trafficking; Small Arms and Light Weapons (SALW); National Security; Transnational Organized Crime; Border Security

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Index Terms--CO2 Emission Prediction, Machine Learning, Regression Models, XGBoost, LightGBM, CatBoost, MLP Regressor, Streamlit Dashboard, Climate Change, Data Analytics.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The increasing complexity of modern software development and deployment has led to an urgent need for intelligent solutions to predict and prevent system failures. DevOps practices, which emphasize continuous integration, continuous delivery, and automated testing, have significantly improved software reliability. However, unpredictable failures continue to impact system stability, deployment efficiency, and user experience. Artificial intelligence (AI) has emerged as a powerful tool to enhance failure prediction and prevention by leveraging machine learning models, anomaly detection techniques, and predictive analytics. This article explores how AI-powered solutions are revolutionizing DevOps by predicting failures before they occur, identifying root causes, and automating preventive measures. It examines the challenges faced in integrating AI into DevOps workflows, such as data quality, model interpretability, and system scalability. The study also discusses real-world applications, methodologies for implementing AI-driven failure prediction, and future research directions to further optimize AI's role in DevOps.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

AI in DevOps, failure prediction, failure prevention, machine learning, anomaly detection, predictive analytics, DevOps automation.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The Transformer architecture, originally developed for Natural Language Processing, has emerged as a prominent new approach in computer vision. This review concentrates on Vision Transformers (ViTs) and their application in medical imaging. We discuss the fundamental Transformer components, including self-attention, multi-head attention, and encoder-decoder structures, then describe how images are divided into patch-based sequences for transformer processing. The review highlights key elements of ViT architecture, such as patch embedding, positional encoding, encoder design, and feed-forward layers, and compares ViTs to Convolutional Neural Networks (CNNs) regarding feature extraction, local versus global context understanding, efficiency, and benchmark results. We also examine hybrid models that combine CNNs' emphasis on local features with the broad modeling capacity of self-attention. Additionally, we review notable ViT models like the Swin Transformer, DeiT, and those tailored for medical tasks like TransUNet and UNETR. Our findings indicate that, although CNNs remain effective with limited data, ViTs and hybrid models generally outperform pure CNNs in large-scale, resource-intensive medical imaging tasks. These outcomes highlight the increasing importance of Vision Transformers in medical applications such as tumor detection, organ segmentation, and disease classification.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Vision Transformers, Medical Imaging, Self-Attention, Convolutional Neural Networks, Swin Transformer, DeiT, Patch Embedding, Deep Learning.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

DR is one of the most prevalent diabetic retinopathy. preventable causes of vision loss, which cause an acute. requirement of effective and early detection means. The present review examines the joint implementation of EfficientNet, a current deep-learning. Gradient-weighted Class Activation Mapping, architecture, and Gradient-weighted Class Activation Mapping. (Grad-CAM) to enhance the classification and interpretability of DR. in medical imaging. The compound scaling strategy employed by EfficientNet makes it very accurate and calculationally efficient, which is why it is suitable with largescale retinal screening. Grad-CAM also enhances the diagnostic pipeline since it produces intuitive. visual descriptions to emphasize areas of clinical importance, Hence making it more transparent and practitioner-confident. Recent studies which used this combination to determine the severity of DR. grading is evaluated in terms of performance variables including accuracy, sensitivity and specificity. Key challenges--including Unequal size of data set, inconsistency in image quality, and scarcity. generalization in varied groups of patients- are also talked about. The review concludes that the combination of EfficientNet and GradCAM presents an effective direction of the creation of scalable, explainable, and clinically reliable automated DR screening.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Federated Learning, Mental Health Monitoring, PHQ-9 Depression Assessment, Emotion Recognition, Privacy Preservation, Multimodal Behavioral Data, Secure AI Chatbot, Flower Framework

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Maha Kumbh 2027 at Nashik is expected to welcome hundred million of devotees from entire world. This presents significant challenges in managing transportation, parking, and emergency response. Traditional systems often fail to handle large crowds effectively, resulting in congestion, delays, and security problems. This paper introduces KumbhSahyogi, a GPS-based, AI-powered smart mobility and service management system designed to maximize parking allocation, optimize routes, and coordinate transportation for large events like the Kumbh Mela cordinate transportation for large events like the Kumbh Mela. The system includes various features such as real-time parking slot availability, AI-powered route optimization through Dijkstra's algorithm and congestion heatmaps, multimodal transport integration, and a smart chatbot for user support.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Kumbh Mela 2027, Smart Transportation, Artificial Intelligence (AI), GPS Navigation, Smart Parking, Dijkstra's Algorithm, Crowd Management, AWS.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The Travel Details Smart Logging Application is an intelligent mobile-based solution developed to modernize and streamline the process of capturing, managing, and storing travel-related information in a structured digital format. Traditional methods of recording travel details often depend on manual entry or paper-based documentation, which can lead to inaccuracies, data loss, and inefficiencies in managing travel records. This ap-plication addresses these challenges by providing an automated, user-friendly platform that allows users to record origin and destination locations using real-time GPS services, automatically capture date and time information, and manage additional travel parameters such as mode of transportation and number of travelers through an intuitive interface. The system enhances accuracy by minimizing manual intervention and ensures reliable data logging through automated location detection and time stamping. Designed with a focus on usability and efficiency, the application supports seamless data entry and retrieval, making it suitable for individuals, organizations, and institutions that require structured travel documentation for purposes such as expense tracking, travel monitoring, logistics management, and reporting. The lightweight architecture and scalable design allow easy deployment across devices while maintaining performance and accessibility. By integrating mobile technology with location-based services, the application demonstrates how digital solu-tions can significantly improve everyday travel documentation processes. The proposed solution promotes efficiency, reduces administrative overhead, and provides users with a convenient and dependable method to maintain organized travel records, thereby contributing to smarter and technology-driven travel management practices.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Travel Monitoring, GPS Tracking, Mobile Ap-plication, Data Analysis, Smart Tourism

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Lalpur Chhedipara is a small village located in Dhuliyan town of Murshidabad district, West Bengal. This paper attempts to explore the cultural significance of Loko Utsab in this village. The study focuses on the active participation of villagers in preserving this tradition through performances. Gambhira is one type of Loko Utsab. It also highlights Gambhira's role as a medium of social awareness and cultural identity.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Lalpur Chhedipara, Village, Tradition, Gambhira, Folk

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Abstract: The Spread of Christianity among the Bodos brought about different socio-cultural changes among them. The spread of Christianity among the Bodos of Borshijhora village began in about 1980 under Rev Fr. Joseph Zubizaretta and Fr. Edward D' Souza. The spread of the religion in the village, in due course of time, shaped the socio-cultural life of the villagers. This study aims to highlight the spread of Christianity in Borshijhora village, causes and impact of conversion to the faith.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Christianity, Borshijhora, causes, impact, Social life.

  License

Creative Commons Attribution 4.0 and The Open Definition