Daily Archives: June 26, 2026

Authors: Jayshree Pansare, Karan Singh, Affan Ali Sayyed, Rushikesh Langhi, Prathamesh Dive

Abstract: The rapid expansion of digital communication platforms has significantly increased the need for secure and efficient messaging systems. Modern users rely heavily on chat-based applications for academic collaboration, professional coordination, and personal communication. However, traditional messaging systems often fail to provide an optimal balance between data security and efficient information management. While some platforms emphasize usability, they frequently compromise on privacy, whereas others focus on encryption but lack intelligent tools to manage large volumes of conversational data. This research presents a Secret Chat Room with AI Summarization System, a web-based platform designed to address both security and usability challenges. The system integrates end-to-end encryption using AES and RSA algorithms to ensure confidentiality and protect messages from unauthorized access. Additionally, it employs WebSocket-based real-time communication to enable low-latency and efficient message exchange between users. A key contribution of this work is the integration of an AI-based summarization module that utilizes transformer-based model Gemini Summarization. This module processes chat histories and generates concise summaries, allowing users to quickly understand lengthy discussions without manually reviewing all messages. This feature significantly reduces information overload and enhances productivity. The system follows a modular architecture consisting of authentication, encryption, messaging, and AI components. Experimental observations indicate that the system achieves efficient performance with minimal latency while maintaining strong security standards. The proposed solution is suitable for applications in education, enterprise communication, and collaborative environments.

DOI: http://doi.org/10.5281/zenodo.20937546

Published by: vikaspatanker

Authors: Vidya Deshmukh, Samradnyi Patil, Akshada Veer, Jayada Talharkar, Tahareen begampalli

Abstract: Modern security environments, particularly on the battlefield, demand autonomous systems capable of real-time threat detection and neutralization without relying on human intervention. This paper presents the Autonomous Threat Detec-tion and Elimination System (ATDES), an integrated hardware-software platform designed to detect enemy armored threats — specifically tanks — using a vision-based AI detection pipeline and respond autonomously through a servo-controlled targeting and firing mechanism. The system leverages a Raspberry Pi 3B+ as the central processing unit, integrating a 2- megapixel camera for visual acquisition, an IR transceiver pair for friend-or-foe (IFF) identification, an RF receiver for enemy signal detection, and a servo-mounted firing mechanism for threat neutralization. A lightweight deep learning model is deployed on-device for real-time tank detection from camera frames, achieving sub-50 ms inference latency at a resolution of 480×640 pixels. IR-based IFF communication ensures that allied units are correctly identified and excluded from targeting, minimizing the risk of fratricide. Blynk IoT cloud integration enables remote monitoring and event logging. The system operates off-grid using a battery and solar power combination, enabling continuous 24×7 autonomous surveillance. Simulation results confirm consistent real-time de-tection with high confidence scores, demonstrating the feasibility of deploying edge AI for autonomous military threat response. The proposed system contributes a cost-effective, scalable, and intelligent prototype for next- generation autonomous defense systems.

Published by: vikaspatanker

Authors: Ayush Wankhede, Ajinkya Patil, Partth Thombre, Mohit Patil, Mahesh Korade

Abstract: Online pet adoption platforms face significant chal- lenges with fraudulent listings and attribute misrepresentation, eroding user trust. This paper presents a complete pet adoption system integrating CNN-based image verification to authenti- cate listing attributes before publication. Transfer learning with EfficientNet-B0 is applied to 110,425 images spanning 712 breed classes across dogs, cats, and birds. A two-stage training strategy first trains the classification head with frozen base layers, achiev- ing 84.7% validation accuracy, then fine-tunes the top 40 layers to reach 89.3% validation accuracy. The verification pipeline combines breed confidence, color confidence, and prediction certainty into a normalized trust score (VScore, range 0–100). Server-side scoring with an 85-point threshold prevents client manipulation while achieving 98.0% fraud detection accuracy. A one-way privacy gateway protects adopter identities, and automated digital adoption certificates with unique certification IDs formalize successful adoptions. Experimental validation on 128 verification requests demonstrates an 84.4% acceptance rate, 1.8 s average processing time, and only 1.6% false positive rate.

DOI: http://doi.org/10.5281/zenodo.20930615

Published by: vikaspatanker

Authors: Mayuri Dongre, Arshiya Sahare, Sarang Dumbhare

Abstract: In the age of digitalisation, a lot of transaction information is produced online, and it is significant to understand customer behaviour and market trends. This paper aims at examining the behavioural patterns in transaction data based on a data-driven approach. The information is gathered using web scraping on Flipkart, primarily in the electronic products categories of mobiles, headphones, smart watches, speakers, accessories with the help of Selenium WebDriver and Python. The obtained data is saved in the CSV format and processed with Python libraries, such as Pandas and NumPy, that allow cleaning data, eliminating duplicates, missing values, and categorizing products. Additional analysis is conducted to establish customer preferences, expenditure trends and product demand trends. The end results are presented in visual representations in the form of dashboards and reports to aid in improved business decision-making. This research assists in interpreting the behaviour of transactions and is useful in the data-driven strategies.

DOI: https://doi.org/10.5281/zenodo.20929867

Published by: vikaspatanker

Authors: Deepa Barethiya, Deepak Vinod Chouksey, Ankur Sanjeev Khurpadi

Abstract: Cardiovascular diseases remain the leading cause of mortality worldwide, accounting for approximately 17.9 million deaths annually according to the World Health Organization. Early detection and accurate risk assessment of heart disease are critical for effective clinical intervention and improved patient outcomes. Traditional diagnostic methods often depend heavily on subjective clinical judgment, which can be inconsistent and time-consuming. This research proposes a Machine Learning-based predictive system that leverages clinical data to assess the risk of heart disease with high accuracy. The proposed system employs multiple classification algorithms including Logistic Regression, Random Forest, Support Vector Machine (SVM), and XGBoost, and evaluates their performance on the UCI Cleveland Heart Disease dataset. Feature selection techniques such as correlation analysis and Recursive Feature Elimination (RFE) are used to identify the most significant clinical predictors. The proposed ensemble model achieves an accuracy of 91.8%, sensitivity of 93.2%, and specificity of 90.4%, outperforming individual classifiers. The results demonstrate that machine learning can serve as a reliable and scalable decision-support tool for cardiologists and general physicians in early heart disease diagnosis.

DOI: https://doi.org/10.5281/zenodo.20929505

Published by: vikaspatanker

Authors: Deepa Barethiya, Ankita Vairagade, Harshal Kathalkar

Abstract: The role that Artificial Intelligence plays in changing the way people work around the world is really big. Artificial Intelligence makes things more efficient. Creates new jobs but it also makes people worry about losing their jobs and having to be more flexible at work. This paper looks at how Artificial Intelligence's affecting people’s jobs and it uses surveys and reviews of what other people have written to do this. The results show that there is a difference between how worried people are about losing their jobs and how much they are doing to learn new things because things, like money and time get in the way. Artificial Intelligence is making it really important for people to learn skills it is not just something people can do if they want to it is something people have to do.

DOI: https://doi.org/10.5281/zenodo.20928264

Published by: vikaspatanker

Authors: Deepa Barethiya, Himani Shirpurkar, Drushti Dharmik

Abstract: Artificial Intelligence (AI) is increasingly integrated into clinical decision-making, particularly in disease-based treatment selection. AI systems promise efficiency, predictive accuracy, and personalized care by analyzing large datasets and recommending tailored therapies. However, these benefits are accompanied by ethical challenges that must be addressed before widespread adoption. Issues of transparency, bias, accountability, privacy, and patient autonomy are consistently reported in recent literature [1][5]. This paper synthesizes findings from 20 peer-reviewed studies published between 2023 and 2026, offering a systematic review of ethical concerns and governance strategies. By combining thematic analysis with case studies in oncology, cardiology, infectious disease, and neurology, we propose a framework for ethically responsible AI deployment in healthcare.

DOI: https://doi.org/10.5281/zenodo.20927818

Published by: vikaspatanker

Authors: Amit C. Kamble, Himmat T. Jadhav

Abstract: Energy security and tariff volatility are growing concerns for Indian higher-education institutions (HEIs) due to rising digital infrastructure, cooling loads, and escalating electricity prices. This paper presents a multi-year, bill-validated assessment of a 180.18 kWp grid-connected rooftop PV system in- stalled at Shivaji University, Kolhapur (SUK). Beyond reporting measured performance (average generation ≈283,824 kWh/yr; CUF ≈18%), the study introduces a Performance Stability Index (PSI) and a Tariff Resilience Index (TRI) to quantify interannual energy stability and financial robustness under adverse tariff scenarios. A 25-year discounted-cash-flow model, incorporating real tariff evolution, yields an IRR of 18.4%, NPV of about INR 517 lakh, and payback of ∼5.3 years. Annual CO2 avoidance is estimated at ∼233 tCO2/yr using the CEA grid factor. A benchmarking framework situates the system against Indian HEI PV case studies, and a replication pathway is outlined for campus-scale deployment. The results demonstrate that rooftop PV can significantly enhance HEI energy security while support- ing national solar and NEP-2020 sustainability goals.

DOI: http://doi.org/10.5281/zenodo.20927527

Published by: vikaspatanker

Authors: Deepa Barethiya, Dhanashri Pannase, Gangasagar Kashyap

Abstract: Ensemble machine learning has pushed crop recommendation accuracy past 99% on standard soil-weather benchmarks — yet this milestone conceals a troubling gap. Systems built around Random Forest, XGBoost, and gradient boosting produce ranked crop labels while leaving the economic viability of each suggestion entirely unexamined. A farmer told "grow rice with 99% confidence" still does not know whether that choice will leave a positive margin after seed, fertiliser, and irrigation costs. This paper proposes a decision-oriented framework that moves beyond the accuracy plateau by coupling a soft-voting ensemble with per-crop yield regressors and a configurable economic layer that estimates expected profit. Where conventional pipelines terminate at a suitability label, the proposed architecture extends the output to a Risk-Adjusted Expected Profit, mathematically formulated as E[Π_c ]_(risk-adjusted)=P_ensemble (c│X) Π_c, where P_ensemble (c│X)is the Ensemble Suitability Probability and Π_c=((Y_c ) ̂(X)×P_(market,c)×1000)-Total Cost_cis the Nominal Net Profit. This coupling mathematically discounts the apparent value of high-risk crops by their probability of soil-weather failure — a correction absent from every reviewed system. To illustrate the theoretical decision dynamics of this framework, we construct a conceptual walkthrough across 200 hypothetical soil-weather scenarios derived from standard agricultural benchmarks. This analysis suggests that the agronomically top-ranked crop and the economically top-ranked crop diverge in roughly 46% of cases — a finding that, if borne out in empirical deployment, would have direct implications for farm-level income planning. A conceptual Streamlit dashboard design is also proposed, embedding real-time what-if sliders and SHAP-based feature attributions to make the system transparent to extension workers and farming cooperatives. The central argument of this paper is simple: a classifier that ignores profit is only half a tool. This framework proposes the other half.

DOI: https://doi.org/10.5281/zenodo.20927114

Published by: vikaspatanker

Authors: Assistant Professor Deepa Barethiya, Sakshi Jibhkate, Samiksha Daronde

Abstract: This paper compares Java-based frameworks and ASP.NET Core for web applications used by companies. It looks at how they work and how well they handle a large number of users. The study checks things like how long it takes for the application to respond, how much work it can handle, how much of the computer’s brain it uses and how much memory it needs when a lot of people are using it at the same time. They ran tests to see what would happen if many people used the application. The results show that ASP.NET Core is really good at responding and using resources wisely. Java-based frameworks are good at handling a lot of users and working with computers at the same time. This study tells us what is good and what is not so good about Java-based frameworks and ASP.NET Core. It helps people choose the tools to build big web applications for companies.

DOI: https://doi.org/10.5281/zenodo.20926193

Published by: vikaspatanker