Category Archives: Uncategorized

Authors: Siddhi Ekawade, Apurva Jate, Arya Kamble, Sharvari Kate, Tejal Panmand

Abstract: We have seen an rapid increase in the Artificial Intelligence (AI) and Deepfake, due to that it has become very easy for online scams and frauds. It is good but when its in correct hands but they are misused in voice cloning, fake video impression, fraud message it has became a serious cybersecurity concern. This paper reviews different deepfake and AI-driven scams and examines the detection and protection methods. While the precautions are taken to keep the user safe from all scams they still face challenges. The paper focuses on the need of improved solutions along with the awareness measures. There are existing methods show results but struggles with advance deepfakes. Through all the complaints and reviews its seen there should be double verification or multi-layer security to secure the user. The study is mainly focused on analyzing how deepfakes are created using Generative Adversarial Networks (GANs) and previously published papers and reports. The current finding has shown that there is face limitation in identifying the advanced deepfakes. And to tackle this issue this paper highlights the need for combined approach to improve AI-based detection, system review, user awareness to reduce scams an increase the security of the device and realism.

Published by: vikaspatanker

Authors: Vedant Pawar, Hasnain Pathan, Sujal Thombare, Yash Kondake, Aarti Gohade

Abstract: Loan utilization tracking is an important element of responsible lending to ensure that borrowers utilize loan funds strictly for their intended purposes. Traditional monitoring techniques such as physical inspections, manual reporting, and post-disbursement audits are often slow, expensive, and prone to inaccuracies. With the rapid growth of mobile technology, digital finance platforms have transformed loan monitoring through mobile applications, GPS tagging, cloud-based dashboards, digital receipts, photo/video verification, USSD/SMS reporting, and AI-based analysis. This survey paper provides an in-depth review of existing mobile-based loan utilization tracking approaches, studies their benefits in reducing fraud, examines their effectiveness in improving transparency and repayment rates, and analyzes challenges such as privacy issues, digital literacy, and connectivity limitations. The paper also proposes a robust mobile-based framework for lenders to track loan utilization efficiently and discusses future innovations including blockchain, biometric authentication, and advanced analytics to strengthen monitoring systems.

Published by: vikaspatanker

Authors: Hemant Khatpe,, Ayush Pilane, Sahil Choudhari, Chaitanya Jam, Tejal Panmand

Abstract: The rapid dependence on smartphones for storing personal and professional information has introduced new risks during emergencies such as device loss, battery failure, or network unavailability. Xaanaax is a digital emergency backup and information access system designed to ensure uninterrupted availability of essential files, contacts, and notes anytime and anywhere. The platform offers secure, centralized cloud-based storage with 24/7 accessibility across devices. This survey paper reviews existing digital storage and backup solutions, identifies their limitations in emergency scenarios, and analyzes how Xaanaax addresses these gaps through its architecture, security mechanisms, and user-centric design.

Published by: vikaspatanker

Authors: Durvesh Mohan Kavire, Swaraj Pradeep Khade, Saurav Deepak Patankar, Abhijeet Shivraj Waghmar, Ms. Tejal Panmand

Abstract: A Multilingual Chat Application is a communication platform designed to allow users from different linguistic backgrounds to interact seamlessly by integrating real-time messaging with automatic language translation capabilities. The application enables users to send and receive messages in their native language while the system transparently translates the content into the recipient’s preferred language, thereby eliminating language barriers and promoting effective communication. It leverages advanced natural language processing and machine translation technologies to ensure accurate, fast, and context-aware translations, while maintaining message privacy and data security. The system typically supports multiple languages, user authentication, message synchronization, and an intuitive user interface to enhance usability across diverse user groups. Such an application is particularly valuable in global business collaboration, online education, customer support services, healthcare communication, and social networking, where participants may not share a common language. By fostering inclusivity and improving accessibility, the multilingual chat application plays a significant role in connecting people worldwide, enhancing cross-cultural interaction, and supporting efficient information exchange in today’s globally interconnected digital environment.

Published by: vikaspatanker

Authors: Dr Meghna Sharma, Dr Namita Yash

Abstract: This research paper makes an effort to identify the different reasons and methods adopting the downsizing activities by the management in their organizations. The objective is fulfilled with the help of analysing the responses received against the questionnaire. The descriptive analysis has been done to work out the reasons to downsizing in a systematic manner. It’s the employee’s perspective by which the research can be concluded by providing the rational reasons to the layoffs. The reasons can be stated as cost factor, company policy format, performance criteria, company reorganization.

DOI: https://doi.org/10.5281/zenodo.17986288

Published by: vikaspatanker

Authors: Aryan Pate, Mohit Patil, Yadnesh Dalvi, Kanchan Birari, Ashutosh Kale

Abstract: Education changes dramatically today, with personal and adaptive learning experience needs at an all-time high. Amidst this dynamism, acampus.ai stands in the helm of revolutionary change with a product using artificial intelligence to fundamentally change the way people learn. The cutting-edge platform brings into every classroom personalized AI tutors with a customized educational experience tailored for each learner's specific needs. ACampus.ai is nothing less than an AI tutor system. So, it's much more than just a classroom approach. Instead of customized courses and real-time assistance on doubts, the teaching methodology adjusts dynamically to the learner's pace and style. It analyzes user behavior and learning patterns by finding out what needs to be done in relevance and alignment to the goal of the learner. It's either deep and comprehensive understanding about the subject or short answers to doubts-there is empowerment through a holistic and individualized experience provided by acampus.ai.

DOI: https://doi.org/10.5281/zenodo.17985744

Published by: vikaspatanker

Authors: Sali Radha, Sali Radha, Bacchav Jayesh, Patil Harshal, Boraste Siddesh

Abstract: In today’s educational landscape, institutions increasingly recognize the value of student feedback for enhancing learning experiences. However, traditional methods like manual reviews and basic statistics often fail to capture the complex and complicated patterns within this feedback. Our project proposes a novel approach using Long Short-Term Memory (LSTM) algorithms to analyse student feedback and predict sentiment more effectively. LSTM’s strength in handling sequential data enables us to uncover deeper insights into student experiences and trends. This innovative method aims to transform feedback analysis into a comprehensive, data-driven evaluation tool, ultimately improving educational practices. Additionally, we implement a Generative Pre-trained Transformer (GPT) model to provide dynamic, tailored suggestions for student growth. By combining advanced machine learning techniques, our system not only analyses feedback but also offers actionable recommendations, fostering a more supportive and effective learning environment. This holistic approach aims to enhance both student outcomes and institutional practices.

DOI: https://doi.org/10.5281/zenodo.17985712

Published by: vikaspatanker

Authors: Aaditya Said, Vivek Khalate, Shubham Sanap, Aayush Nahire, Deepali Suryawanshi

Abstract: Modern agriculture faces challenges such as soil degradation and suboptimal crop selection, leading to reduced productivity. The “Smart Farming-IOT based” project focuses on three key components: crop recommendation, fertilizer recommendation, and climate-based predictions. Using historical data, climate conditions, and soil characteristics, a machine learning model predicts the most suitable crops for a given area, ensuring optimal land use and increased yield. The project empowers farmers with real time insights to enhance productivity and support sustainable agriculture.

DOI: https://doi.org/10.5281/zenodo.17985636

Published by: vikaspatanker

Authors: Chaitanya Madhav Mate, Sanil Nivrutti Shinde, Aakansha Ganesh Tambe, Niranjan Deepak Lahane, Kirti Patil

Abstract: This paper describes a "Dynamic Job Market Analysis Platform" that allows capturing real-time analytics and prediction of the employment fluctuations with the goal of benefiting universities students and employer. By utilizing machine learning models to forecast trends accurately, the platform fills this gap between student skill sets and the leading demands of the industry. It provides students with actionable insights to help them ensure their career paths are in line with market requirements, and it helps employers better understand trends in the workforce. The results highlight an accuracy of over 97% in prediction of the employment patterns, showcasing the ability of the platform to fuel data-based decision making. This project helps to improve employability and provides a better alignment between academia and market requirements by addressing issues, such as data imbalance and dynamic changes in the market. Future directions include expanding the scope of real-time data integration and refining prediction models for broader applicability.

DOI: https://doi.org/10.5281/zenodo.17985571

Published by: vikaspatanker

Authors: Bhavesh Avinash Gadekar, Neha Sanjay Gaikwad, Swayam Vijay Bhosale, Prasad Ambadas Bidgar, Ganesh K Gaikwad

Abstract: Dementia diagnosis is a critical challenge in neurology, often relying on time-intensive and subjective manual analysis of MRI scans. This research proposes a novel hybrid AI-based system for early and accurate dementia detection, combining traditional neuroimaging techniques with advanced deep learning models. The system employs a hybrid 2D-3D pipeline that integrates slice-based 2D convolutional models with volumetric 3D CNN architectures, ensuring a balance between computational efficiency and spatial pattern recognition. The 2D models focus on extracting detailed features from individual MRI slices, while the 3D models capture spatial relationships across the entire brain volume. Additionally, clinical metrics such as cognitive scores are integrated with the MRI data to enhance diagnostic accuracy. Attention mechanisms and Grad-CAM visualizations improve model interpretability by highlighting critical brain regions, addressing the need for transparent AI-driven clinical tools. This hybrid approach significantly improves diagnostic accuracy, generalizability, and explainability compared to conventional methods. The system classifies scans into Strongly Demented, Mildly emented, or Non-Demented categories, providing actionable insights for clinicians. By bridging AI with neuroimaging and multimodal data integration, the proposed system aims to revolutionize dementia detection, enabling earlier intervention and improved patient outcomes.

DOI: https://doi.org/10.5281/zenodo.17985501

Published by: vikaspatanker