Daily Archives: March 13, 2026

Authors: Prof. Tejashree Pangare, Harshad More, Aryan Patil, Raj Patil

Abstract: In an age where digital information has reached an all-time high, it is essential to be able to access the most relevant and correct data out of the exorbitant storage of information from the internet and its various search engines, even more so when it comes to domain-specific knowledge like Space Science. The primary goal of this project is to propose an AI based Help Bot that can be used to do an intelligent search on data from MOSDAC (Meteorological and Oceanographic Satellite Data Archival Centre), a project of ISRO, in a manner that feels conversational and natural to the user through the use of Knowledge Graphs, NLP, and Semantic Search algorithms. In addition to providing a comprehensive and customized, easy, efficient, and smooth information-seeking experience to its users that can be a researcher/scientist, a student, or simply the general public, the implementation of a bot system as such with the state-of-art NLP techniques that understands relationships between entities and dynamic learning algorithms that adapt to newly updated information content in its database, will take us one step closer to achieving a no-miss search experience that also takes into account their intent, thereby improving the accessibility of information and decreasing information search time while truly revolutionizing the way we access Space-derived information regarding Meteorological and Oceanographic data.

DOI: https://doi.org/10.5281/zenodo.18999423

 

Published by: vikaspatanker

Authors: Mrs..K.M.Swarna Devi, Divith S, Jayaprakash C, Madhavan S

Abstract: Timely detection of eye-related diseases is critical for preserving vision and preventing permanent visual loss. With the growing availability of ophthalmic imaging, artificial intelligence has emerged as an effective tool for enabling fast and automated disease screening. This study proposes a real-time artificial intelligence–driven framework for eye disease detection based on deep learning techniques. The system employs convolutional neural networks (CNNs) to process retinal fundus images and optical coherence tomography (OCT) scans for identifying prevalent eye conditions such as diabetic retinopathy, glaucoma, and age-related macular degeneration. To support real-time operation, the model architecture is optimized for low computational complexity and rapid inference without compromising diagnostic accuracy. The proposed system assists ophthalmologists by providing instant diagnostic feedback, reducing manual examination time, and supporting early clinical decision-making. Experimental evaluation demonstrates that the model achieves high detection accuracy along with minimal processing delay, making it suitable for real-time deployment in clinical settings, telemedicine platforms, and large-scale eye screening programs. The results highlight the potential of AI-based solutions to enhance accessibility, efficiency, and reliability in modern ophthalmic diagnosis.

DOI: https://doi.org/10.5281/zenodo.18998750

 

Published by: vikaspatanker

Authors: Ms. K.Madhumitha, Abdul Kareem S, Divakaran M, Gowtham G M

Abstract: Real-Time Voice-Enabled IoT Irrigation for Smart Agriculture introduces an advanced automated irrigation system aimed at improving water management and agricultural efficiency. The proposed framework combines IoT-based environmental sensors with real-time data processing and a voice-interaction interface to support intelligent farm operations. Sensors deployed in the field measure soil moisture, ambient temperature, and humidity, transmitting the collected data to a cloud platform for continuous monitoring and analysis. The system automatically activates or deactivates irrigation based on threshold values and real-time conditions, ensuring precise water distribution. Furthermore, a voice-enabled feature allows farmers to access system updates and manage irrigation through simple spoken commands using smartphones or smart devices. This reduces the need for manual supervision and promotes efficient resource utilization. The solution is particularly beneficial for remote agricultural areas where timely intervention is critical. Experimental validation indicates enhanced water conservation, reduced operational effort, and improved crop growth compared to conventional irrigation practices. Overall, the proposed system offers a scalable, economical, and user-friendly approach to achieving sustainable and data-driven smart farming.

DOI: https://doi.org/10.5281/zenodo.18998450

 

Published by: vikaspatanker

Authors: Deepak Tomar

Abstract: The rapid expansion of cloud computing, artificial intelligence applications, and hyperscale digital services has significantly increased the energy demand of modern data centers, raising concerns about sustainability and operational efficiency. Energy storage systems have emerged as a promising solution for stabilizing power supply, integrating renewable energy sources, and improving overall energy utilization in data center infrastructures. However, conventional energy management strategies often lack the intelligence required to dynamically optimize energy storage and distribution under varying workloads and fluctuating energy availability. This study explores the concept of intelligent energy storage management for sustainable data centers by integrating advanced analytics, machine learning techniques, and real-time monitoring systems to optimize energy storage operations. The proposed framework enables predictive energy demand forecasting, intelligent charging and discharging of storage systems, and efficient integration of renewable energy sources such as solar and wind power. Through intelligent decision-making mechanisms, the system aims to reduce energy waste, lower operational costs, and minimize carbon emissions while maintaining high reliability and performance of data center operations. The findings highlight the potential of intelligent energy storage management systems to significantly enhance energy efficiency and support the transition toward greener and more sustainable data center infrastructures.

DOI: https://doi.org/10.5281/zenodo.18997564

 

Published by: vikaspatanker

Authors: Deepak Tomar, Kismat Chhillar

Abstract: Insider threats remain one of the most complex and costly cybersecurity challenges faced by modern organizations, as malicious or negligent actions originate from trusted users who possess legitimate access to critical systems and sensitive information. Traditional rule-based detection mechanisms often fail to identify subtle behavioral deviations that precede insider incidents, resulting in delayed response and elevated organizational risk. This study proposes a behavioral analytics framework powered by machine learning techniques to detect insider threats through dynamic modeling of user activity patterns. By leveraging multi-source organizational logs, including authentication records, file access events, communication metadata, and network activity traces, the framework constructs individualized behavioral baselines and identifies anomalous deviations indicative of potential threat activity. Both supervised and unsupervised learning models are evaluated using a benchmark insider threat dataset, with careful attention to data imbalance mitigation and model interpretability. Experimental results demonstrate that ensemble learning methods and temporal modeling approaches significantly enhance detection accuracy while maintaining acceptable false positive rates. The findings underscore the importance of integrating behavioral machine learning models into Security Operations Centers to enable proactive, scalable, and context-aware insider threat mitigation strategies.

DOI: https://doi.org/10.5281/zenodo.18996897

 

Published by: vikaspatanker