Conference Proceeding - RECENT INNOVATIONS IN ENGINEERING SCIENCE AND TECHNOLOGY (ICRIET)
Volume 13 | Issue 7

 Abstract

Car parking is today a source of concern in city centers, and with escalating needs alongside a split, inadequate car park provision, one major negative outcome is the fact that road traffic becomes rather hard to handle, supported by the real congestion issues. They can result in inefficiencies as, for instance, massive levels of time wastage, plus greater consumption of fuel. Smart parking systems (SPS) can be described in terms of what they involve in offering innovative solutions through the application of contemporary technologies like sensors, Internet of Things (IoT) devices, and real-time data analytics to maximize the use of parking spaces. They are capable of providing pertinent, real-time information about parking availability to guide drivers into available spaces and enable payment processes. Furthermore, some intelligent parking systems incorporate dynamic pricing schemes, which calculate varying parking prices based on demand, hence reducing imbalances in occupancy levels among parking. This paper gives a general overview of intelligent parking systems, particularly their technological aspects and functionalities, and their impact on urban mobility. In addition, it emphasizes the advantages of the SPS such as traffic congestion reduction, environmental effects, and improvement in user experience. The uptake of smart parking systems is transforming urban parking management at a fast pace with efficient and sustainable strategies for congested cities.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Smart Parking, IoT, AI, Real-Time Parking, Urban Mobility, Traffic Reduction, Autonomous Vehicles, Smart Cities, WSN

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Traditional cryptographic techniques, especially RSA and ECC, which are frequently employed in secure email systems, are seriously threatened by the quick development of quantum computing. We propose a Post-Quantum Secure Email System to solve this problem by including lattice-based cryptography, an encryption method that is impervious to quantum attacks. This system offers secure authentication, end-to-end encryption, and an effective structure for key management. By implementing NTRU and Kyber, secure email storage and retrieval is ensured, by reducing quantum vulnerabilities. A MongoDB-based storage solution combined with a Flask API enables real-time encrypted email processing. Performance evaluations show the system's storage overhead, encryption speed, and computing efficiency compared to traditional cryptographic methods. Its resistance to man-in-the-middle, brute-force, and quantum-enabled decryption assaults is validated by security analysis. In the quantum era, the proposed approach guarantees email secrecy and integrity over the long period.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Performance tests measured encryption and decryption times, database query speeds, and system responsiveness under concurrent user loads, ensuring efficiency despite the higher processing cost of post-quantum cryptography. The system maintained an average email retrieval time of 50-70 ms, balancing security and usability.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Video summarization is a fundamental task of computer vision and multimedia processing to condense lengthy videos into short representations without losing the valuable content and context. The study is founded on the use of object detection techniques in video summarization and relies on the capability of deep learning to automatically recognize and extract discriminative objects and events from video streams. Relying on the benefit of the latest object detection models and new summarization techniques, the study tries to enhance the efficiency and effectiveness of video summarization to allow users to quickly perceive the content and meaning of videos without the requirement of lengthy playback. The approach not only enhances video browsing and comprehension of content but also has its future areas of application in surveillance, video indexing, and content recommendation systems. Video summarization is a crucial element to successfully extract key moments from lengthy video recordings, reducing storage and processing costs, and enhancing the user experience. The work proposes a state-of-the-art video summarization technique founded on real-time object detection based on YOLOv3 and Deep SORT algorithms. Based on the fusion of the new approaches, the proposed method effectively extracts and tracks discriminative objects with improved efficiency, leading to informative and meaningful video summaries. Experimental results exhibit enhanced efficiency and accuracy compared to state-of-the-art summarization techniques, indicating the potentiality of the proposed methodology in its real-world applications like surveillance, sport analysis, and content generation.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Video Summarization, Object Detection, Object Tracking, YOLOv3, Deep SORT, Real-Time Processing

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The deployment of 5G networks introduces a new era of communication technologies, provides higher data speeds, reduction in latency and enhanced connectivity. However, these advancements also increase significant concerns regarding security, especially with the increased attack surface and complexity inherent in 5G's architecture. Intrusion Detection and Prevention Systems (IDPS) are essential for safeguarding 5G networks against malicious threats and ensuring the integrity and availability of services. This paper surveys the state-of-the-art techniques for detecting and preventing in 5G environments, including both traditional and modern approaches. We discuss the role of machine learning and artificial intelligence in enhancing the detection capabilities of IDPS, as well as the challenges posed by the dynamic, distributed nature of 5G networks. Additionally, we explore the integration of IDPS with emerging 5G technologies such as network slicing, edge computing, and the Internet of Things (IoT), highlighting the potential for more adaptive and scalable security solutions. The paper also reviews key issues like real-time processing, scalability, and the need for privacy-preserving methods in intrusion detection. Finally, we identify research gaps and propose directions for future work to enhance the resilience of 5G networks.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

5G Network, Intrusion Detection and Prevention Systems (IDPS), Network Security, Cybersecurity in 5G, Machine Learning (ML)

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

AgroPredict operates as a platform focused on vital agricultural issues such as crop selection and nutrient adjustment and disease prevention that specifically benefits Indian agricultural industries. AgroPredict delivers recommendations for appropriate crop choice through its IoT sensor applications by analyzing authentic soil data with environmental conditions of specific regions. Plant diseases become detectable through image recognition systems which allow early disease identification then the same technology analyzes symptoms while inspecting soil nutrient level to determine proper fertilizers based on plant requirements. AgroPredict achieves accurate predictions through the combination of three ML algorithms that include Random Forest with Support Vector Machines (SVM) and Convolutional Neural Networks (CNN). Users can easily access the system using web and mobile interfaces because the system features intuitive functional design. Through data-driven information delivery AgroPredict helps farmers decrease financial losses and increase yields and adopts sustainable farming practices.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

artificial intelligence, crop prediction, fertilizer recommendation, IoT, machine learning, smart farming

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The proposed loan application processing .system for rural areas is specifically designed to address the unique challenges faced by agricultural communities when seeking financial assistance. This system aims to overcome the obstacles inherent to rural settings, ensuring a seamless and effective process for securing crucial financial support. Tailored to the specific needs of rural users, the system commences with farmers initiating the application process through a user-friendly interface designed explicitly for their use. A paramount focus is placed on robust data storage and management, ensuring the secure preservation of loan application forms. Employing advanced missing data imputation techniques enhances the integrity of the datasets. The website design emphasizes user interfaces that are both intuitive and accessible, accommodating varying levels of technological literacy prevalent in rural settings. The assessment of loan eligibility is facilitated by the integration of a machine learning model, carefully considering factors pertinent to agricultural finance. deployed locally and integrated via APIs, ensuring adaptability to both local systems and external services. The workflow concludes with a transparent and streamlined loan approval or rejection process, accompanied by insightful financial recommendations for approved applicants. This holistic approach, merging technology, effective data management, and machine learning customized for rural contexts, aspires to diminish the financial inclusion gap in rural areas. Ultimately, the system endeavors to empower farmers, enabling them to secure essential financial resources for sustainable agricultural practices.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Machine Learning, Loan, Data, Validation.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This study highlights the importance of cybersecurity and ethical hacking in India by examining the various policies, tools, and methods. The focus revolves around the efficacy of legal structures and policies regarding the cybercrimes that are surfacing in India. Cybersecurity is essential in today's digital world, as it shields private information, sensitive data, and vital infrastructure from online attacks. The swift development of digital technologies in India has resulted in a rise in cybersecurity risks, jeopardizing the availability, confidentiality, and integrity of critical data. Ethical hacking has emerged as a crucial tool in identifying and mitigating these threats. Strong cybersecurity measures aid in preventing monetary, reputational, and even loss of life due to the increase in cyberattacks such as ransomware, malware, and phishing. This report highlights the security concerns and difficulties by examining the present status of cybersecurity regulations and ethical hacking techniques in India. Strong cybersecurity regulations, practical ethical hacking techniques, and awareness campaigns are necessary to counteract cyber threats, according to a thorough review of the literature and professional viewpoints.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

cybersecurity, cybercrime, ethical hacking, cybercrime in India, government policies on cybersecurity in India, data protection laws in India, cybercrimes in India

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Agriculture is the primary source of food for the global population. It provides essential nutrients and sustenance for billions of people worldwide. Without agriculture, there would be no reliable food supply, leading to hunger and malnutrition. Agriculture plays a significant role in the economy, source of Raw Materials, Biodiversity and Ecosystem Services etc. This paper explores the Real-World uses of Machine Learning (ML) in sustainable agriculture to confront the difficulties of a growing global population and climate change. With traditional methods suffering from reduced effectiveness due to changing climate patterns and rising food demands, ML offers a data-centric approach to revolutionize crop management. The study investigates ML algorithms such as neural networks, support vector machines, decision trees, and ensemble models to analyze key factors like soil quality, climate conditions, and past crop performance. The aim is to optimize crop selection for specific regions, resulting in higher yields and reduced environmental impact.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Smart Farming, Crop Yield Prediction, Soil Fertility Assessment, Machine Learning, AI Applications, IoT Applications

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This paper aims to develop an advanced system that detects and maps cyberbullying incidents on social media using AI technologies. The system will leverage Transformer techniques to identify relevant patterns and contextual signals from social media posts, accurately spotting instances of cyberbullying. Teachers can use the system to identify and support students affected by online harassment. Policymakers can benefit from precise data on cyberbullying trends and locations, enabling informed decision-making and the development of robust policies to combat online abuse. Social media platforms can utilize this technology to monitor and mitigate cyberbullying incidents, ensuring a safer digital environment for users.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Cyber bullying, BERT, FSSDL-CBDC, Deep learning, Machine learning, Transfer learning, CNN

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Colonoscopy is essential for detecting colorectal cancer (CRC) and pre-cancerous polyps, allowing for timely intervention and better patient care. Nonetheless, the manual analysis of colonoscopy images can be slow and prone to human mistakes, which increases the likelihood of overlooking polyps or making incorrect diagnoses. This study examines the use of deep learning techniques to automate the detection and classification of polyps in colonoscopy images. By employing convolutional neural networks (CNNs) and sophisticated image processing methods, the research seeks to improve the accuracy, efficiency, and dependability of colonoscopy analysis, aiding healthcare providers in diagnosing conditions related to the colon. The focus of this work is on preparing colonoscopy images, isolating significant regions, and extracting important features to train a deep learning model for classification purposes. The suggested system framework combines the segmentation and classification models to differentiate between normal and abnormal colon tissues. The method has been evaluated using a thorough dataset of colonoscopy images, showing significant enhancements in detection accuracy compared to traditional methods.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

A DIAGNOSIS OF COLON CANCER USING DEEP LEARNING ALGORITHM

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Workplace safety, particularly in construction and industrial places, is a critical concern due to the high risks faced by the workers in the hazardous zones. To address these challenges, our study introduces a computer vision based system for virtual border identification and safety monitoring. Using real-time object detection algorithm YOLOv5, the system tracks human movements and monitors proximity to danger zones by overlaying virtual boundaries on live video streams. It instantly detects safety violations, triggering visual and audible alarms while notifying supervisors, and significantly reducing the chance of any accidents.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Virtual Safety Zones, Computer Vision, YOLOv5, Real-Time Monitoring.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Agriculture is the backbone of food security and economic stability, but crop diseases significantly impact yield and quality. Traditional disease detection methods often require expert intervention, making them time-consuming and inaccessible to many farmers. To address this challenge, this research proposes an Augmented Reality (AR)-Driven Disease Detection System for smarter farming. The system integrates computer vision, deep learning, and AR technology to provide real-time disease identification and treatment recommendations. The proposed framework utilizes image processing and convolutional neural networks (CNNs) to detect plant diseases from images captured by AR-enabled devices such as smartphones. The processed results are overlaid on the plant using AR visualization, allowing farmers to recognize affected areas and access treatment solutions instantly. Additionally, the system provides real-time recommendations, preventive measures, and expert consultation options, ensuring a proactive approach to disease management. By leveraging machine learning, real-time data processing, and AR-based visualization, this solution enhances precision farming, reduces dependency on chemical treatments, and improves crop health monitoring efficiency. Farmers receive immediate, visual feedback on their crops, highlighting potential disease symptoms through AR interfaces. Furthermore, the system facilitates teleconsulting features, enabling farmers to seek expert advice remotely. This approach not only reduces crop losses but also promotes sustainable agricultural practices by minimizing excessive chemical use and optimizing disease management strategies. Implementing this AI-driven smart farming technology aims to empower farmers, enhance decision-making, and contribute to a more efficient and resilient agricultural sector.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Augmented Reality (AR), Disease Detection, Smart Farming, Precision Agriculture, Deep Learning, Computer Vision, Convolutional Neural Networks (CNNs), Real-time Monitoring, Image Processing, Sustainable Agriculture, AI- driven Farming, Smart Crop Management

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Attendance management is a fundamental task in educational institutions and workplaces, but traditional methods, such as roll-calling or card- swapping, are prone to inefficiencies, inaccuracies, and time wastage. Keeping track of attendance the traditional way can often lead to mistakes and even manipulation. To solve this, our paper introduces a smarter approach--a Face Recognition-Based Attendance System. By using advanced computer vision and machine learning, this system makes attendance tracking more accurate, efficient, and hassle-free. The system utilizes OpenCV and dlib for detecting real-time face and recognition, enabling a seamless and contactless way to mark attendance. Our system uses the ResNet50 model to capture unique facial features, creating 128-dimensional embeddings to accurately match individuals with stored records. Attendance is securely logged in an SQLite database, and a simple Flask-based web interface makes it easy to access records anytime. By automating the process, this system removes the hassle of manual entry, reduces errors, and provides a reliable solution for schools, colleges, and workplaces. Additionally, the integration of Tkinter for the face registration interface ensures ease of use, while SQLite offers a reliable storage system. By implementing this solution, administrative workload is significantly reduced, accuracy is enhanced, and attendance management becomes more streamlined and efficient.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Face Recognition, Smart Attendance System, Computer Vision, OpenCV, dlib, Face Detection , Flask Web Application, Tkinter GUI, Automation, Attendance Management System, Scalable Solution.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Human-machine collaboration presents persistent challenges in aligning human intent with machine comprehension and execution. Large Language Models (LLMs) offer promising solutions by leveraging advanced natural language processing capabilities to bridge this gap. This paper surveys a novel framework that integrates LLMs into a vehicle "Co-Pilot," enabling autonomous systems to interpret and execute driving tasks based on human commands and contextual information. The proposed framework incorporates a robust interaction workflow and a memory mechanism to systematically organize and retrieve task-relevant data. By dynamically selecting appropriate controllers and planning optimal trajectories, the Co-Pilot adapts its operations to fulfill user-defined goals while maintaining safety and efficiency. Simulation experiments demonstrate the framework's ability to understand natural language instructions, plan actions, and execute driving tasks effectively, highlighting both its practical viability and limitations. Furthermore, the study emphasizes the importance of real-time adaptability in addressing complex driving scenarios and explores the concept of human-machine hybrid intelligence. This work illustrates the potential of LLMs to revolutionize autonomous driving by enabling more intuitive and effective human-machine collaboration.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Human-machine collaboration, Large Language Models (LLMs), Vehicle Co-Pilot, Human-machine interaction, Natural language understanding, Trajectory planning, Autonomous driving, Hybrid intelligence.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This paper presents a conceptual framework for a Women's Safety Protocol -- a proposed Python-based web application aimed at enhancing personal security through emergency alerts and real-time location tracking. By leveraging APIs such as Twilio and Geopy, the system is designed to discreetly notify emergency contacts when a user is in distress. Though still in the ideation stage, this proposal outlines the planned architecture, potential use of wearable integration, and future AI enhancements for emotional state detection. Future enhancement could involve AI-driven emotion detection using wearable devices. The app utilizes Geopy to track the user's location with up to 200-meter accuracy, ensuring swift response capabilities. In the event of an emergency, it notifies multiple trusted contacts simultaneously and requests the user to verify their safety using a secure passcode. If the passcode is not entered, the system escalates the alert automatically. Through the combination of real-time tracking, automated alerts, and a discreet, intuitive interface, the Women's Safety Protocol offers a reliable and effective solution for women to access help promptly--enhancing personal safety and enabling rapid assistance from friends, family, or authorities.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

In emergencies, real-time alerts and quick-response protocols, powered by the IoT, ensure women's safety by tracking location and enhancing security.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

To predict the price of second-hand cars, we have implemented three approaches namely, XGBoost, Random Forest, and Decision Tree Regressors. The Kaggle sourced dataset is cleaned, feature selected and treated for outliers in order to improve the accuracy. To evaluate the performance of the model and for knowing which factors in the price affect, we use the R2 score and Metrics evaluation of the Scikit learn module available in Python. Therefore, XGBoost is likely to outperform the rest because it is more efficient. Thus, we build a simple to use web application where Users may input Automobile details and get the same time price estimates. Our project serves to help buyers and sellers use the data to accurately predict a second-hand car price.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Random Forest, Decision Tree, XGBoost Regressor, Machine Learning, Car Price Prediction.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Accurate age detection has emerged as a useful tool in a constantly growing digital environment, spanning sectors including identity verification, wellness, healthcare, and personalized services. Conventional age estimation techniques use static inputs, but new opportunities for data-driven, real-time estimation have been made possible by developments in machine learning and predictive analytics. By utilizing deep learning models with real-time processing, we aim to create a robust system that adapts to diverse datasets and varying environmental conditions. The proposed framework employs optimization techniques to enhance efficiency of the model, ensuring seamless integration into cloud-based platforms for scalable deployment. The proposed study also indulges in deployment of the age detection model as a Software-as-a-Service for various security-based applications, such as security in net-banking. Beyond technical implementation, this study addresses the ethical considerations surrounding age detection, emphasizing fairness, transparency, and data privacy. The results of this research contribute to the growing field of machine learning-driven security applications, providing more insights, and secure identity verification. Through this work, ChronoDetect aims to demonstrate how AI-driven age estimation can be both efficient and ethically responsible, paving the way for broader applications in digital security domains.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Machine Learning, Predictive Analytics, Age Detection, Cloud Computing, Software-as-a-Service (SaaS), Net-Banking, Optimization Techniques, Real-time Processing, OpenCV, TensorFlow.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The main aim of this project is to develop and deploy an Expense Tracker application that enhances the management of finances through the utilization of advanced technologies. The application incorporates Optical Character Recognition to scan and extract information from invoices and receipts, along with a Large Language Model that classifies expenses into pre-defined categories. Using Flask for the frontend and Python for the backend, the project promises a smooth and friendly user interface. The OCR component employs sophisticated text recognition techniques to pull out transaction information from receipts, while the LLM categorizes transactions into typical expense categories such as food, utilities, and entertainment. Additionally, the app shows these categorized expenditures through interactive graphs and detailed transaction lists on the dashboard, offering users insightful information about their expenditures. This project showcases the successful integration of OCR and LLM technologies in actual applications, underlining the potential for automation of everyday financial work. The system seeks to enhance financial literacy and allow users to make knowledgeable choices, hence encouraging efficiency as well as access in managing expenses.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Automated Expense Tracking, Optical Character Recognition (OCR), Machine Learning Algorithms, Financial Management, Expense Categorisation, Tesseract OCR, Spending Insights, Neural Networks (CNN, LSTM), Real- Time Expense Monitoring, Data Extraction

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This solution introduces a web-based report generation tool that streamlines academic writing by integrating advanced formatting and data management features. The user- friendly interface supports autoformatting, including standardized font sizes, justified content, and 1.5-line spacing, while offering seamless data input and session management with auto-save functionality. AWS integration ensures secure storage and processing, and the tool includes robust data management capabilities such as historical access to previously generated documents. This comprehensive solution aims to enhance student productivity and reduce stress associated with academic report creation by providing an efficient, reliable, and userfriendly platform that adheres to academic standards.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Report Generation Tool, Web-Based Application, AWS Integration, Academic Support, Data Management, User-Friendly Interface, Document, Google AI Studio APIs

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The growing popularity of cryptocurrency has increased the need for accessible and efficient Bitcoin mining platforms. Traditional mining methods often involve high power consumption, costly hardware, and complex setups, making them unsuitable for individual users. This paper introduces cryptoMiner Pro, a lightweight Bitcoin mining solution that eliminates the need for specialized equipment. The platform emphasizes ease of use and energy efficiency, offering a clean and intuitive interface tailored for both novice and experienced users. Robust security measures are incorporated to safeguard user data and ensure secure transactions. Additionally, an integrated administrative module supports effective oversight of users and transactions. By simplifying the mining process, cryptoMiner Pro aims to democratize Bitcoin mining and make it more inclusive for a broader audience.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Accessibility, Bitcoin Mining, Lightweight Mining Platform, Secure Transactions, Transaction Management, User Interface

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This study has been undertaken to investigate the determinants of revenue generation and financial performance in the e-GramPanchayath system, using two analytical frameworks: the traditional Financial Ratio Analysis and an Econometric Model based on Arbitrage Pricing Theory (APT). To test the financial model, basic revenue indicators such as tax collections and service charges are used, while macroeconomic variables are applied in the APT framework. The macroeconomic variables include inflation, rural employment rate, government grants, and agricultural output. For this purpose, monthly time series data has been compiled from January 2015 to December 2020 from various Gram Panchayath records and government databases. The analytical framework includes both correlation analysis and regression modeling to identify the significant factors influencing revenue trends and financial sustainability in local governance systems.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Digital Governance, E-Government, Rural Development, Transparency, Online Services

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The project ClickTalk Interface is a web-based application that takes human-computer interaction to the next level by combining voice commands and hand gestures. Built with Next.js and Tailwind CSS, it features tools like virtual mouse control, virtual volume control, speech-based commands, text-to-speech conversion. Using cutting-edge libraries like OpenCV, Media pipe, and speech recognition, the website makes tasks easier, boosts productivity, and improves accessibility. Its modular design helps users control their system in intuitive ways, all hands free.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

CLICKTALK INTERFACE

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

As software development has advanced at a high speed with artificial intelligence (AI), AI-enabled code assistants are now indispensable assets to enhance the productivity of developers and the quality of code. But how can AI truly transform the way we write, debug, and optimize code? This paper presents an AI-facilitated VS Code extension, "AI Powered Pair Programming Assistant," which can function as a virtual coding companion through intelligent code recommendation, inline descriptions, test case generation, and feedback. By utilizing Gemini AI, our system combines cutting-edge code analysis and test case automation to accelerate the development process, minimize manual debugging efforts, and improve collaborative coding effectiveness. This integration tackles remote teams' challenges of time zone disparities and miscommunication through instant AI-backed support and suggestions. Through experimentation and evaluation, we explore the potential of AI-based assistants to maximize method generation, enhance test coverage, and facilitate smoother software development practices.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

SMART CODING PARTNER An AI-Powered Assistant for Better Code and Productivity

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This project develops an AI system to detect and classify diabetic eye diseases early using retinal images. It uses image enhancement and CNNs to extract key features. A hybrid model combining deep learning and traditional ML classifies disease types and stages. The system is trained on public datasets and considers patient info like age. It achieves high accuracy in identifying conditions like diabetic retinopathy and glaucoma. This tool helps doctors with faster, more accurate diagnosis and supports telemedicine use.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Advanced Classification Technique for Diabetic Eye Disorders

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Blood donation systems in many regions continue to face significant challenges related to timely donor-recipient matching, efficient communication, and accurate prediction of blood demand. This paper presents PulseMatch, a smart, web-based blood donation platform designed to address these issues using cloud technologies and machine learning. Developed with a React.js frontend and Firebase backend, PulseMatch facilitates seamless interaction between hospitals and potential blood donors. The system enables real-time donor registration, request submission, and intelligent donor matching based on blood group, location, and availability. In addition, machine learning modules are proposed for fraud detection and blood shortage forecasting to improve reliability and response times. PulseMatch integrates a scalable architecture that supports live data synchronization, automated alerts, and future extensions including mobile compatibility and explainable AI. This paper details the system design, implementation workflow, and integration strategy for intelligent automation in blood donation, demonstrating the potential to modernize and optimize healthcare logistics through data-driven approaches.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Blood Donation System, Smart Healthcare, Machine Learning, Firebase, React.js, Donor Matching, Fraud Detection, Shortage Prediction, Cloud-based Platform, Healthcare Automation.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This paper presents the development of an AI-powered healthcare Chabot utilizing Retrieval-Augmented Generation (RAG) to provide accurate, reliable, and multilingual medical assistance. By integrating advanced natural language processing (NLP), image recognition, and speech processing, the Chabot offers personalized and context-aware health support, retrieving real-time information from open-access medical databases to enhance accuracy and reliability. Unlike traditional rule-based chatbots, which rely on predefined responses, our sys-tem dynamically generates answers using large language models (LLMs), ensuring adaptability to evolving medical knowledge. A key feature is its multimodal interaction, supporting multilingual voice conversations and computer vision for analyzing skin conditions, making healthcare assistance more inclusive. This enables users to engage via text, speech, or images, improving accessibility for non-native speakers and individuals with disabilities. Experimental results highlight improvements in response accuracy, efficiency, and user engagement, demonstrating the system's potential to bridge healthcare accessibility gaps.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Healthcare, Chatbot, Retrieval-Augmented Generation, Natural Language Processing, Computer Vision, Multilingual Support, Artificial Intelligence

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Floods pose a significant threat to human life, infrastructure, and the economy. This paper presents Flood Sense, an advanced flood prediction system powered by machine learning. The system integrates historical meteorological data, real-time hydrological parameters, and satellite imagery to provide early warnings and risk assessments. Various machine learning techniques, including Decision Trees, Random Forest, and Artificial Neural Networks (ANN), are employed to enhance predictive accuracy. A web-based dashboard, built using Flask, enables real-time monitoring and alert dissemination. The goal of this system is to aid government agencies, disaster management teams, and local communities in making informed decisions to mitigate flood damage.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Flood Prediction, Artificial Intelligence, Real- Time Monitoring, Disaster Management, Hydrological Analysis, Remote Sensing.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This paper presents MindMate, an AI-powered mental health chatbot designed to offer accessible, anonymous support through interactive, voice-enabled conversations. Leveraging fine-tuned language models, LangChain, FAISS, and Streamlit, the chatbot delivers personalized guidance and retrieves relevant information in real time. It uses retrieval-augmented generation (RAG) to enhance accuracy and integrates validated psychological tools like PHQ-9 and GAD-7 for sentiment-aware responses. The chatbot is trained on a diverse dataset of mental health dialogues, FAQs, and synthetic conversations. Built with Hugging Face transformers and a FAISS-powered retrieval system, it dynamically adapts to user inputs. MindMate is accessible through a streamlined web interface, enabling users to seek help anytime, free from judgment.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Mental health support, FAISS, LangChain, LLM, RAG, Sentiment analysis

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Early Identification of crop infections is a critical component of sustainable agriculture, as it is essential to maximize crop yields and minimize losses. Corn plants are susceptible to various infections that can have a substantial impact on crop production, including Northern Corn Leaf Blight, Common Rust, and Gray Leaf Spot. This research introduces a computational framework that incorporates artificial intelligence (AI) and employs deep learning techniques to identify diseases in maize plants while assessing their severity. The system will use high-resolution images that will be used to train a convolutional neural network (CNN) for robustly diagnosing disease. The system design incorporates a MongoDB database that will allow for efficient storage, retrieval, and management of disease-related data. The system will be able to provide growers with flexibility through real-time tracking and instant feedback to help growers make informed decisions to help control and prevent plant disease. The models are implemented utilizing TensorFlow and PyTorch, and are designed to be scalable and accurate. A well-organized interface will allow farmers and agronomists ease of access to the prediction process. The system design shows how automated disease detection can be combined with real-time information for smart farming. Future improvements will depend on improving the detection accuracy and later applying these models to more crops. This study reinforces sustainable agricultural practices and integrates AI-based precision farming systems.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Maize crop disease identification, AI-driven plant health monitoring, CNN-based disease classification, advanced deep learning in agriculture, intelligent farming systems, automated crop disease recognition, real-time agricultural diagnostics.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

With the era of personalized entertainment, it is essential to get movie recommendations spot on with the user sentiments. Our paper introduces SentimentSync an artificial intelligence-based kannada movie recommendation system that relies on sentimental analysis of YouTube trailer comments and web scraping of dynamic movie listings of the Times of India in contrast to other traditional recommendation systems SentimentSync combines locally hosted sentimental analysis using NLTK's Vader with sophisticated web-scraping techniques through selenium the system generates aggregate sentiment scores for a user-queried film and a group of upcoming releases then rank-filter and returns recommendations based on similarity. An interactive flask web interface shows recommendations with an auto-generated explanation utilizing large language models (llms) through the langchain platform. Experimental outcome shows that our hybrid solution increases recommendation relevance as well as the ease of using an interactive web interface over having to use a paid APIs.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Sentiment Analysis, Movie Recommendations, YouTube API, Selenium, Flask, Kannada Movies, Web Scraping, LangChain, NLTK, LLM

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Cyber deception threats are becoming increasingly sophisticated, often evading traditional security measures such as firewalls and intrusion detection systems. Attackers can exploit unpatched systems or use advanced techniques to infiltrate networks. This paper introduces an AI-powered IP-based cyber deception system designed to confuse and deceive attackers using intelligent honeypots and anomaly detection. Our approach enhances threat intelligence and adapts dynamically to evolving threats.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Component, formatting, style, styling, insert.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The rapid advancement of deepfake generation techniques has created significant challenges in preserving the authenticity of digital media. This comprehensive survey examines the state-of-the- art in deepfake video detection, with a particular focus on hybrid Long Short-Term Memory (LSTM) models that combine spatial and temporal analysis capabilities. We analyze over 50 recent studies (2019-2024) to evaluate the effectiveness of various architectural approaches, including Convolutional Neural Network- Long Short-Term Memory (CNN-LSTM), Three Dimensional Convolutional Neural Network- Long Short-Term Memory (3DCNN-LSTM), and attention-enhanced variants. The paper provides a detailed comparison of model performance across benchmark datasets such as FaceForensics++ and Celeb-DF, while discussing key evaluation metrics like AUC-ROC and F1-score that are critical for assessing detection reliability. We systematically identify current limitations in generalization capability, computational efficiency, and adversarial robustness that hinder real-world deployment. The survey concludes by outlining promising research directions, including multimodal fusion techniques, lightweight model architectures for edge deployment, and explainable AI approaches to enhance forensic credibility.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Hybrid Long Short-Term Memory (LSTM), Convolutional Neural Network- Long Short-Term Memory (CNN-LSTM), Three Dimensional Convolutional Neural Network- Long Short-Term Memory (3DCNN-LSTM), multimodal fusion techniques.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Deepfake technology, driven by artificial intelligence, has developed rapidly over the past few years, raising issues of misinformation, privacy violations, and online security threats. This project is centered around creating a robust Deepfake Detection System based on machine learning methods to distinguish real media from the manipulated one. The system has a user authentication module for secure access via a login system. In addition, it incorporates an advanced deepfake detection algorithm that can scan images and videos to verify whether they are authentic. The detection model generates a fake accuracy percentage, reflecting how much media are likely manipulated. This measure adds transparency and gives users quantifiable feedback into possible deepfake risks. The system utilizes convolutional neural networks (CNNs) and deep learning to make high-precision identification of synthetic content. The technology can be applied to real-world scenarios such as media authentication, law enforcement, and social media surveillance, helping in the mitigation against misinformation. To make it scalable and efficient, the platform will be developed with an easy-to-use interface where individuals and organizations can upload and examine media easily.Through the creation of a correct and accessible detection system, we are moving closer to maintaining trust in digital content and preventing the risks involved in synthetic media manipulation.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Deepfake Detection, Convolutional Neural Networks (CNNs), deep learning techniques, AI-driven cybersecurity

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This paper presents the design and implementation of a comprehensive web-based College ERP system aimed at enhancing the efficiency of academic and administrative operations in educational institutions. The system automates critical functions such as student enrolment, faculty management, attendance tracking, and examination processing, replacing traditional manual methods that often lead to inefficiencies and errors. Developed using modern web technologies, the solution ensures scalability, robust data security, and user-friendly access across various roles within the institution. It incorporates features such as role-based access control, a modular architecture, and real-time analytics to support data-driven decision-making and institutional transparency. By streamlining operations, reducing administrative workload, and improving communication among stakeholders, the system fosters a more organized and technology-driven educational environment. Furthermore, it is designed with future extensibility in mind, supporting cloud deployment and integration with advanced tools such as AI analytics and Learning Management Systems (LMS). This ERP system not only provides a practical approach to managing college operations efficiently but also serves as a foundational step toward ongoing innovation in educational technology.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

College ERP, Student Information System (SIS), Role-Based Access Control (RBAC), Attendance Management, Examination Management, Web-Based ERP System, Database Management, Cloud-Based Deployment & Data Security.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

In the rapidly evolving landscape of e-commerce and retail, the integration of predictive modelling, real-time data analytics, and artificial intelligence (AI) has significantly transformed pricing strategies, customer engagement, and operational efficiencies. This research investigates the implementation of predictive analytics techniques for new product pricing, the role of real-time data processing in enhancing business agility, and the transformative impact of AI in delivering personalized consumer experiences. Predictive modelling techniques leverage historical data, market trends, and consumer behavior to optimize pricing decisions, while real-time analytics architectures utilizing technologies like Apache Kafka and Apache Flink facilitate immediate insights into inventory management, customer preferences, and dynamic pricing.This paper presents a comprehensive analysis of how these technologies collectively empower businesses to achieve operational excellence, enhance customer satisfaction, and sustain competitiveness in the digital marketplace. Ethical considerations regarding data privacy and algorithmic fairness are also highlighted, ensuring responsible deployment of AI- drivennsolutions. The study ultimately emphasizes the critical role of data-driven, real-time, and AI- augmented approaches in shaping the future of e-commerce and retail industries.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Predictive Modelling, Real-Time Data Analytics, Artificial Intelligence, Dynamic Pricing, Customer Personalization, E-Commerce.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The calculation of the Air Quality Index (AQI) in India differs greatly from global norms due to regional characteristics such as geographical diversity, seasonal fluctuations, and pollution sources. While most countries use consistent techniques that emphasize pollutants such as PM 2.5, NO2, and O3, India's methodology favors PM 10 and PM 2.5 due to high dust levels, industrial emissions, and biomass combustion. The AQI calculation for India includes adaptive seasonal modifiers to account for crop burning, festivities like Diwali, and climatic conditions such as monsoons and winter inversion. Additionally, regional weightage variables are added depending on local pollution sources, which improves accuracy. Unlike worldwide models, which rely mainly on static pollution criteria, India's model makes dynamic modifications to account for real-time environmental and demographic conditions. This approach provides a more relevant and accurate representation of air quality, catering to India's unique climatic, industrial, and cultural conditions. In addition, we present a detailed investigation of chemical processes and how their various quantities influence the toxicity of the compounds produced. We investigate the significance of five key gases. We assess the adverse effects of the produced items utilizing data from internet sources and a variety of calculation and visualization methodologies. The evaluation is based on established threshold values for all gases involved.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Air pollution, Pollution Control Board, Pollutant data analysis, Predictive modelling, Random Forest ML algorithm, User Friendly website, Data visualization.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The AI-Powered Afforestation Planner project aims to address the growing issue of air pollution through strategic afforestation. By leveraging advanced remote sensing and machine learning techniques, the project identifies barren land areas suitable for tree planting to improve air quality. The study focuses on the Kanakapura Taluk in Ramanagara District, where land classification is performed using Google Earth Engine (GEE) with manually provided training samples. These samples were used to classify the region into urban areas, water bodies, vegetation, and barren lands using the Random Forest algorithm. The project fetches real-time Air Quality Index (AQI) data to assess pollution levels and recommends the optimal number and species of trees for planting. The final output is a web application that provides users with land classification results, barren land area calculations, and tree species recommendations tailored to improving air quality based on AQI levels. The web-based approach ensures accessibility for end users, offering an interactive tool for better environmental decision-making.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Afforestation, Land Classification, Google Earth Engine (GEE), Random Forest, Air Quality Index (AQI)

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

As virtual statistics turns into an increasing number of popular ensuring strong safety for sensitive files is critical this studies introduces a complicated protection framework that mixes biometric watermarking with device mastering to establish a tamper-resistant and adaptive protection system by way of encoding intricate iris and fingerprint patterns the usage of a custom designed rubiks cube encryption algorithm the method creates a comfy embedded watermark that is tremendously proof against manipulation in parallel convolutional neural networks CNNs examine and authenticate biometric statistics permitting real-time detection of spoofing tries and unauthorized changes the adaptive gaining knowledge of functionality of CNNs lets in the system to refine its detection accuracy through the years strengthening its resilience against rising threats this precise integration of encryption and shrewd pattern recognition gives extensive improvements in file security with ability packages in sectors which include healthcare finance and authorities wherein records integrity and authentication are paramount.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Biometric Watermarking, Document Security, Rubik Encryption, Convolutional Neural Networks (CNN), Machine Learning, Iris and Fingerprint Fusion, Zero-bit Watermarking, Authentication, Spoofing Detection, Fraud Detection.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The AI Based Crop Recommendation System in Karnataka is an online platform designed to assist farmers in making optimal decisions regarding crop cultivation. It incorporates features such as analysis of soil health, weather forecasting, and market price predictions. Farmers can also access information on suitable crop varieties, irrigation management, and pest control methods through the system. By using this AI-driven system, farmers can improve crop yields, reduce input costs, and enhance overall agricultural productivity. This digital solution streamlines agricultural decision-making and supports sustainable farming practices in Karnataka by providing farmers with essential information and recommendations.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

artificial intelligence, crop recommendation, agriculture, Karnataka, precision farming, sustainable agriculture.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The threat of ransomware is considerable in cybersecurity risk and often goes undetected by traditional signature-based detection approaches. In this paper, we present a deep learning-based behavioral analysis framework supporting pro-active detection and disruption of ransomware. Rather than depending on signatures, the framework analyzes system-level activities, such as file encryption, abnormal access, and process relations. The framework utilizes Long Short- Term Memory (LSTM) networks to analyze temporal activities and Recurrent Neural Networks (RNNs) to extract features, enabling real-time identification of ransomware. Our system detects anomalies present in suspicious behavioral patterns, it provides warnings to the administrators, and automatically either quarantines files or isolates from the network. By using deep learning, our framework detects better and has fewer false positives compared to traditional methods. This study demonstrates the potential for deep learning for analyzing behavior for ransomware protection purposes, giving us a strong and adaptive means of defending against evolving cybersecurity threats.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Ransomware Detection, Deep Learning, Behavioral Analysis, Long Short-Term Memory (LSTM), Recurrent Neural Networks (RNNs).

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This paper presents a novel image enhancement technique that integrates Stationary Wavelet Transform (SWT) and Discrete Wavelet Transform (DWT) with cubic spline interpolation and Contrast Limited Adaptive Histogram Equalization (CLAHE). The method decomposes low-resolution images into frequency sub-bands, processes these sub-bands to estimate high-frequency details, and reconstructs enhanced high-resolution outputs. By addressing challenges such as edge blurring and loss of fine details, the algorithm offers significant improvements over traditional methods. Experimental evaluations on the Kaggle Super-Resolution Dataset demonstrate enhanced Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM), particularly for medical and satellite imaging applications. The approach's adaptability and efficiency make it a promising solution for diverse imaging needs.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Image enhancement, wavelet transform, cubic spline interpolation, CLAHE, super-resolution, SWT, DWT, thresholding

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The proposed project is a cab fare comparison prototype designed to help users estimate and compare cab service prices effectively. Instead of relying on direct API integrations from platforms like Ola, Uber, and Rapido, this prototype uses custom-developed APIs based on publicly available pricing information and predefined parameters. Users can register, authenticate, and manage their profiles, while the platform utilizes location-based inputs to assist with fare estimation. The system also offers trip history, user reviews, and a fare analytics page to help both passengers and drivers.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Cab Fare, Custom APIs, Prototype, Location-based Estimation, Analytics, Django, React.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

A restaurant's ability to take orders for food is essential. This is something that waiters do for customers when they dine at restaurants. Typical restaurant ordering procedures may lead to a number of problems. Server and client misunderstandings during order taking are the root cause of all problems. A short wait for the server to come and take the order is also required of the customer. The current setup is somewhat antiquated, using paper and printed menus to keep track of customer orders. Consequently, a real-time ordering system developed to manage the ordering process for restaurants is the Food Ordering System using QR Code technology. Therefore, the QR Code meal ordering system is a remedy for that problem. Smartphones serve as the foundation of the system since they are now indispensable in modern culture. The restaurant will include a QR code on the menu that customers must scan. Using this method, the buyer may also be sure they got what they requested. Additionally, the restaurant staff has access to the order list and may review the menu.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

QR Code, Food, Restaurant

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Educational institutions face significant challenges in maintaining effective communication between parents and mentors, which directly impacts student performance and engagement. ConnectEd is a web-based application designed to bridge this communication gap by providing a structured and efficient platform for tracking student progress, facilitating real-time interactions, and ensuring transparency between parents and mentors. Technologically, ConnectEd is built using ReactJS, HTML, CSS, and JavaScript for an interactive and responsive frontend, while the backend is powered by Node.js and MongoDB, ensuring scalable and secure data management. The system also incorporates Pinata for decentralized storage where necessary, reinforcing data integrity. Multi-language support is integrated to eliminate communication barriers, making the platform accessible to a diverse user base.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Educational Communication, Parent-Mentor Interaction, Student Performance Monitoring, Web-Based Educational Platform, Academic Progress Tracking, Attendance Management, Timetable Scheduling, Secure User Authentication, Data Privacy, ReactJS Frontend, Node.js Backend, MongoDB Database, Multilingual Support, Admin User Dashboard.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The distribution of humanitarian aid is susceptible to fraud, inefficiencies, and a lack of transparency. This paper presents an AI-integrated blockchain framework to detect fraud and ensure secure aid distribution. The system incorporates machine learning (ML) to detect anomalies in aid transactions and Hyperledger Fabric to maintain immutable, decentralized transaction records. Zero-Knowledge Proofs (ZKPs) facilitate privacy-preserving beneficiary verification, ensuring safe and transparent transactions. The proposed system increases trust, accountability, and efficiency in aid distribution. Experimental findings illustrate enhanced fraud detection accuracy and real-time transaction verification.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Blockchain Technology, AI-Based Fraud Detection, Hyperledger Fabric, Zero Knowledge Proofs, Humanitarian Aid, Cryptographic Security

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The rapid advancements in artificial intelligence (AI) and embedded systems have paved the way for innovative healthcare solutions, especially for chronic disease management. This paper presents the design and implementation of Smart Nurse, an AI-powered robotic healthcare assistant for cancer patients. The system integrates real-time patient monitoring, emergency alert mechanisms, and medication dispensing functionalities using Raspberry Pi. Key features include a fall detection system based on YOLO and sensor data fusion, an emotion detection model combining facial expressions (ResNet-50), voice tone (MFCCs), and sentiment analysis (BERT), and an AI-driven chatbot utilizing GPT-based natural language understanding with OpenAI Whisper for speech recognition. The robot navigates autonomously using SLAM-based AI navigation with ESP32CAM and ultrasonic sensors. It communicates via a hybrid MQTT and API-based system to synchronize with a Django backend and a locally hosted database. Emergency situations trigger a loud SOS alarm and Twilio SMS alerts to caretakers. The hardware framework includes a battery-powered structure with servo-controlled gravity-based medication dispensing. The system is designed to function without cloud dependency, ensuring affordability and privacy. This paper details the system architecture, hardware-software co-design, and the implementation methodologies for AI models, real-time communication, and embedded robotics. The experimental results indicate that the system enhances patient safety, ensures timely medication, and provides emotional support, making it a promising solution for remote healthcare assistance.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

healthcare robotics, artificial intelligence, cancer care, patient monitoring, emotion detection, fall detection, medication management, embedded systems, YOLO, BERT, ResNet50, SLAM.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This project introduces a blockchain-based e-vault system that ensures secure, transparent, and tamper-proof storage and verification of digital certificates. By utilizing the immutable and decentralized nature of blockchain technology, the system effectively eliminates risks associated with certificate fraud and unauthorized alterations. It features two types of users: Admin/Authorized Users, who can upload certificates with recipient details, and Normal Users, who are permitted to verify them. Upon successful upload, certificates are stored on the blockchain and recipients are notified via email with the certificate ID and related information. Users can access all their certificates through email-based login, with an optional Merge Account feature to combine multiple accounts for unified access. Additional functionalities include a Portfolio Page for resume generation, a dynamic pricing model to support institutional sustainability, a user guidance feature for easier navigation, and a live chatbot for real-time assistance. This system not only secures digital credentials but also empowers users and organizations with tools for professional development and efficient certificate management.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Blockchain, Digital Certificates, E-Vault, Tamper-Proof Storage, Authentication, Certificate Verification, Immutable Ledger, Credential Management, Email-Based Access, Resume Generation, Portfolio Page, Real-Time Support.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The Face Recognition Attendance Management System is an innovative solution developed to eliminate the need for manual roll calls. This system provides a quick and accurate replacement of traditional attendance methods by applying computer vision techniques such as Convolutional Neural Networks (CNN) and Haar Cascade. The system captures images, detects faces, and matches them with pre-stored images for automated attendance recording. Future enhancements include cloud integration and mobile app support for real-time monitoring

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Face Recognition, Attendance Management, HOG, CNN, OpenCV, Machine Learning, Deep Learning.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The health of grape plants is crucial for ensuring high-quality vineyard yields and maintaining the economic sustainability of viticulture. Effective disease detection is a pivotal aspect of modern agricultural management, as diseases such as black measles, leaf blight, and black rot can significantly impact crop production. This paper discusses research on some advanced methods in the field of grape plant disease detection by incorporating machine learning algorithms and image processing techniques. In this paper, the use of spectral imaging, neural networks, and field-based monitoring systems for early, precise, and cost-effective diagnosis of diseases is discussed and the user interface is in the regional language Kannada for better usability of farmer. By addressing the limitations of traditional manual inspection methods, this research aims to highlight innovative approaches that enhance efficiency and reduce the environmental impact of disease management practices. The findings underscore the potential of precision agriculture in revolutionizing disease control strategies in viticulture.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Grape Plant Disease Classification, Image Processing, Deep Learning, Feature Extraction, CNN

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Precision agriculture is an emerging area that applies modern information technology and machine learning to create news ways of identifying and diagnosing plant diseases to promote sustainable farming practices. This paper aims to review the application of machine learning and deep learning techniques in plant disease detection and classification in precision agriculture. It also proposes a different approach in classifying relevant literature which is based on the employed methodology - classification or object detection, and reviews the literature on datasets available for plant disease detection and classification. This work comprises a comprehensive analysis within the scope of object detection and classification of plant diseases utilising the PlantDoc dataset. The conclusion reached in this research is that YOLOV5 is the best object detection algorithm and that ResNet50 and MobileNetv2 models are the best image classifier models relative to the time cost of training the models and the accuracy of produced images.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Classification,deeplearning,disease detection,machine learning,object detection,precision agriculture

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Cardiovascular diseases (CVDs), especially myocardial infarctions (heart attacks), represent a leading cause of death globally. Traditional diagnostic approaches such as ECGs, biomarkers, and clinical assessments often fall short due to delayed response and limited sensitivity. The integration of Artificial Intelligence (AI) and Machine Learning (ML) introduces novel possibilities for early prediction, personalized diagnostics, and continuous patient monitoring. This paper presents a comprehensive review of AI's role in cardiovascular risk assessment by highlighting the limitations of conventional techniques, the structure of AI architectures, their clinical advantages, key case studies, and future potential involving hybrid and federated learning systems. Furthermore, it emphasizes data privacy, ethical concerns, and regulatory preparedness to ensure real-world deployment and trust in AI-driven systems

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Artificial Intelligence (AI), Machine Learning (ML), Myocardial Infarction, Cardiovascular Disease (CVD), ECG, Risk Stratification, Predictive Modeling, Explainable AI (XAI)

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Non playable character (NPC) is the core of an immersive and realistic game. As gaming turns into a huge industry, traditional script following NPCs need to be replaced with more dynamic characterization. This can be done through integrating an AI to the NPCs to make the game more immersive and enhance the realism of the game. Taking advantage of AI, an NPC can be given a framework which uses reinforcement learning and conversational AI within a simulation environment. This allows the NPC to engage in natural conversations with the player, learn from their past interactions and dynamically adapt their behavior with respect to the player. Using reinforcement learning the NPC are able to enhance their decision making based on their previous interactions with the player. Conversational AI makes the dialogue have more depth and context aware of the in-game environment. Testing the NPC in stimulated environment, the results demonstrate a more realistic and self-aware NPC.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Reinforcement learning, Conversational AI, Simulation, Dynamic-Decision making, Animation, Open-world exploration, virtualization

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

In this Generation of emails, messages spam continues to pose several challenges to email ecosystem. Spam detection in emails have been a concern because the user security depends on the classification of emails as spam or ham. The existing methods for spam detection lack in precision and is a time consuming process. This paper provides a spam detecting model that accounts for the dynamic nature of spam mails and learning based clustering techniques for classifying spam and ham messages. The model contains various Machine Learning (ML) algorithms used for detection and classification of spam emails. The model is integrated with Artificial Intelligence (AI) for automatic detection of spam or ham messages, which is most advanced form of detecting spam compared to other methods. The model present a novel approach to detect spam using Random forest (RM) classifier which is further enhanced by the designed methodology. The model claims the effective methodology with robust and interpretable features for detecting the spam messages.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Deep Learning . Email Spam Detection . Machine Learning

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The growth of deepfake technology has raised great concerns about privacy, misinformation and cybersecurity. Advanced AI can make it difficult to say real and false media, as deeper and more visual content can change. In this paper we will explore current methods to recognize and prevent deepfakes and check how well they work and have limitations. It also explains how deeplearning to create faces changes, focusing on Stylegan and how it is used to edit, restore and change faces in different styles.We also look into the famous deepfake tool deepfacelab and sketch it to work with high resolution facial films. Apart from Visual Deepakes, we look at FluentLip, the latest audio conditioned LipenSthesis model that improves synthesis language synchronization and smoothness. Finally, let's look at recent advances in speech production. We present an approach to using emotions to create more natural and controllable facial expressions. Regarding existing procedures, limitations, and trends, this review suggests more efficient identification measures, the ethical design of AI, and better public education to combat the growing threat of deepfakes.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Deep learning, deepfakes, face generation, deepfake detection, face-swapping, StyleGAN, AI ethics, audio-driven synthesis, talking face generation

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The Plastic Waste Management and Reward System is designed to promote responsible plastic disposal by leveraging technology to incentivise users for recycling efforts. The system integrates Flask for backend processing, AngularJS, HTML, CSS, and JavaScript for a dynamic frontend, and ImageDB for cloud-based image storage. MySQL is used for secure transaction and user data management, ensuring reliability and efficiency. The platform employs AI-powered image recognition models to classify plastic waste accu-rately, allowing users to earn rewards based on proper disposal. Rigorous testing methodologies ensure performance, security, and scalability. Future enhancements, including blockchain-based rewards, AI-driven classification improvements, and IoT-enabled smart bins, will further optimize waste tracking and management. This project presents an innovative approach to tackling plastic waste pollution by merging technology with sustainability.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Plastic Waste Management, Reward System, Recycling Incentives, Flask Backend, AngularJS Frontend, ImageDB Cloud Storage, MySQL Database, AI-powered Image Recognition, Waste Classification, Secure Transactions, Performance Testing, Security Testing, Scalability, Blockchain Rewards, IoT Smart Bins, Sustainability, Waste Tracking, Environmental Technology

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Agriculture forms the core of the Indian economy, but farmers still grapple with inefficient crop planning, unpredictable weather, volatile market prices, and limited access to real-time farm insights. 'Raitha Bandhava - Farmer's Companion' is an end-to-end Farming Management System that leverages the latest technologies like artificial intelligence (AI), machine learning (ML), and application programming interfaces (APIs) to empower farmers with data-driven insights. The platform enhances productivity through AI-based crop guidance, weather forecasting, and smart market trends analysis. With API integration, it connects the farmer with the government and private farm database to give insights regarding policies, subsidies, and trends within the market. Further, the platform offers AI-based disease detection, virtual input/output market for produce, and supply chain optimization features. This article discusses the architecture, implementation, and impacts of the system and how it addresses existing technological gaps in agricultural solutions. Pilot implementations initially have reported 20% increased yield and 15% reduction in input costs and affirmed the efficiency of the system. It also discusses digital literacy and connectivity challenges and proposes remedies such as offline capabilities and language capability.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Smart Farming, Crop Planning, Market Price Analysis, Weather Forecasting, Supply Chain Management, Digital Agriculture, Farmer Marketplace, AI-Based Disease Detection, Agricultural Technolo

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Lung cancer remains a major global health concern, with early diagnosis playing a crucial role in enhancing patient survival rates. According to the Global Cancer Observatory (GLOBOCAN 2024), lung cancer remains the most common cause of cancer-related deaths worldwide, accounting for over 2.4 million new cases and 1.8 million deaths annually. The application of Artificial Intelligence (AI) in medical imaging has opened new avenues for improving lung cancer detection. This review examines the role of AI, particularly deep learning algorithms, in analysing medical images such as CT scans, X-rays, and MRIs for lung cancer diagnosis and prognosis. Various AI-based techniques, including Convolutional Neural Networks (CNNs), Support Vector Machines (SVMs), and meta-heuristic approaches like the Crow Search Algorithm (CSA), have shown substantial progress in identifying and categorizing lung nodules as benign or malignant. Pre-processing steps such as image segmentation, edge enhancement, and resampling contribute to improving image clarity, thereby enhancing the accuracy of AI-driven diagnostic models. Despite these advancements, challenges such as data imbalance, model interpretability, and generalization persist. This paper also explores the potential of Computer-Aided Diagnosis (CAD) systems in complementing AI methodologies for more precise and reliable clinical applications. Additionally, the study reviews the limitations of conventional histopathological diagnostic techniques and the potential of molecular biomarkers in refining lung cancer classification. The growing use of AI in healthcare is paving the way for personalized treatment strategies, yet the necessity for diverse and extensive datasets remains critical for improving model reliability. Through this review, we aim to provide a structured overview of AI-driven medical imaging advancements in lung cancer detection, offering insights to guide future research and development.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Deep Learning, Convolutional Neural Networks (CNNs), Transfer Learning, Lung Cancer Detection, Lung Nodule Classification, Computer-Aided Diagnosis (CAD), Machine Learning (ML), Artificial Intelligence (AI), Feature Extraction, Deep Neural Networks (DNNs), ResNet, GoogleNet, MobileNetV2, VGG16, InceptionV3, Support Vector Machines (SVM), Random Forest (RF), Optimization Algorithms, Segmentation Techniques, Generative Adversarial Networks (GANs), Conditional Tabular Generative Adversarial Networ

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Artificial intelligence (AI) is playing a critical role in wildlife conservation by enabling species monitoring, poaching prevention, and habitat restoration efforts [1] . Due to habitat loss, poaching, and climate change, wildlife conservation is a major concern. Habitat assessment and resource conservation involve AI-powered image analysis, which aids in assessing forest health, detecting deforestation, and identifying areas in need of restoration [2] . The AI-Powered Wildlife Conservation System improves animal conservation and monitoring by utilizing artificial intelligence. With the help of sophisticated image recognition algorithms, users can submit photos or scan animals in real time. Additionally, it offers vital conservation status data, showing, based on international databases, if an animal is vulnerable or endangered. The technology also uses the Google Maps API to find local physicians and animal rescue facilities, guaranteeing prompt assistance for wildlife that is hurt or in danger. This project intends to assist wildlife researchers, conservationists, and the general public in preserving biodiversity by fusing AI with geolocation services.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Artificial Intelligence (AI), Wildlife Conservation, Animal Identification, Endangered Species Protection, Rescue and Rehabilitation, Conservation Technology, Smart Conservation System, Google Map API, Image Recognition, Environmental Sustainability

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Steganography based on web technologies has improved in recent years from simple HTML and CSS manipulation to advanced techniques using artificial intelligence and advanced web APIs with cross-platform implementations. This review article analyses the progress, trends now, issues, and ways forward in web-based steganography. Classical steganography conceals data in images or sound, while web-based steganography extends this to web technologies. We review recent papers on methods such as HTML, CSS, JavaScript, HTTP headers, and web storage. Our research shows an increasing trend in web-based steganography applications of deep learning, especially those that utilize browser-native functionalities. Important research shortcomings are cross-browser compatibility, the absence of standardized metrics for evaluation, and few studies on steganalysis specific to the web. This review will be an asset to information security, data hiding, and web technology researchers and practitioners.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Web-Based Steganography, HTML Data Hiding, Web Page Steganography, Browser-Based Information Hiding, Network Security, Web Technology Encryption, Data Protection Strategies.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Research often lacks an incentive mechanism, and traditional Academic Publishing lacks transparent and equitable mechanism to reward the researchers and creates hindrances instead; this paper introduces the platform "Raktbeej", a blockchain-based platform that is inspired by the retroactive public goods funding which solves this problem by allowing authors to define royalty distribution percentages for cited works. Using smart contracts, Raktbeej automates the distribution of royalties to the cited authors whenever a donation is made to an author. We evaluate the potential of the platform to transform academic publishing for the better.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Blockchain, Academic Publishing, Decentralised Science, Smart Contracts, Citations, Ethereum, Retroactive Public Goods Funding.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Fingerprints are essential for blood group identification. to diagnose a patient without waiting for traditional medical formalities and to anticipate the patient's blood type in emergency situations. concentrating on obtaining accurate and good results for everyday medical use in hospitals. It will be more beneficial to avoid time-consuming techniques throughout the critical time of medication. In this study, the only correlation between gender and finger print patterns was that females were more likely than males to have loops and arches, and males were more likely than females to have whorls. For a long time, fingerprint identification has been considered one of the most reliable ways to identify someone, especially in court. Fingerprints are believable because, with the exception of severe skin injuries, the patterns we create while still in the womb don't change throughout our lives.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Blood Group, Machine Learning, Finger print, Medical Field.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Due in large part to the difficulties of learning programming, engagement and retention are ongoing issues in computer science education. Adopting cutting-edge teaching techniques that improve learning results and maintain student interest is essential as the need for coding abilities expands across all industries. Gamification is one such strategy that introduces game-like components into educational environments, including badges, leaderboards, points, and accomplishment milestones. Programming-related gamification turns routine coding tasks into engaging and participatory experiences. Through increasingly difficult assignments, this approach fosters critical thinking, increases student engagement, and promotes problem-solving. Learning and skill improvement are reinforced by immediate rewards and real-time feedback. Additionally, gamification fosters a growth mentality by assisting kids in accepting difficulties, growing from mistakes, and persevering through hardship. Gamified learning environments provide a potent tool to boost motivation and academic achievement in computer science education by making coding more accessible and pleasurable.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Game-based learning, motivation, engagement, gamification, educational technology

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Bioactivity forecast may be a basic errand in sedate revelation and advancement, empowering the recognizable proof of potential medicate candidates with tall viability and negligible poisonous quality. By leveraging endless chemical datasets, ML models can learn complex structure-activity connections (SARs) and make exact expectations almost compound intelligent with natural targets Different ML strategies, counting profound learning, irregular woodlands, bolster vector machines, and gathering models, are utilized to improve prescient exactness. Also, progressions in logical AI (XAI) contribute to way better show interpretability, helping chemists in levelheaded medicate plan. This paper investigates later improvements in ML-based bioactivity forecast, challenges such as information quality and show generalizability, and future headings, counting the integration of generative AI and multi-omics information. In this field, machine learning (ML) has grown as an effective tool for promoting data-driven methods that predict the unplanned behaviour of chemical molecule.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

BIO-ACTIVITY PREDICTION USING MACHINE LEARNING

  License

Creative Commons Attribution 4.0 and The Open Definition