VOLUME 14, ISSUE 2, FEBRUARY 2025

Abstract: Recommendations extracted from generative artificial intelligence tools such as large language models via prompt engineering for the design of a system for the automated diagnosis of heart disease are followed through to construct suitable artificial intelligence models for the automated diagnosis of heart disease using clinical measurements. The resulting artificial intelligence models are trained, tested and validated on a clinically validated and publicly accessible heart disease dataset. Observed system performance was reasonable compared to the performance of systems developed by artificial intelligence experts by adopting a custom synthesis approach. The artificial intelligence models could be further refined using inputs such as expert and domain knowledge and ultimately incorporated as an automated heart disease diagnosis module in a comprehensive artificial intelligence-driven healthcare system.

Keywords: Heart Disease, Generative Artificial Intelligence (AI), Large Language Model (LLM), ChatGPT, DeepSeek, Artificial Neural Network (ANN), Deep Learning (DL), TensorFlow, Healthcare System, Disease Diagnosis and Prediction.

Abstract: This study explores the transformative potential of artificial intelligence (AI) in strategic management, focusing on its capacity to drive innovation, enhance efficiency, and address complex ethical and regulatory challenges. A key objective is to establish best practices that help firms mitigate risks associated with AI adoption while ensuring sustainable growth. AI’s role in decision-making, operational efficiency, and competitive advantage, highlighting its impact across various business functions such as marketing, supply chain, and risk management. It also addresses the broader socio-economic implications of AI beyond organizational benefits. The study employs a secondary research approach with a descriptive research design, utilizing thematic analysis of scholarly articles, industry reports, and empirical studies. A positivist philosophy guides the research, ensuring an objective evaluation of AI’s role in strategic management. The findings indicate that AI-driven automation is reshaping business operations, influencing workforce dynamics, and raising societal concerns. Through systematic analysis of twelve relevant studies, the research underscores the pivotal role of data in adapting to an evolving business landscape powered by emerging technologies. The conclusion and recommendations emphasize the need for robust AI governance frameworks and continuous learning environments within organizations. Establishing structured AI implementation strategies and fostering innovation will be crucial for businesses to leverage AI effectively while addressing ethical and regulatory challenges.

Keywords: Artificial Intelligence (AI), Strategic Management, AI-driven Automation, AI-Challenges

Abstract: Most of the manufacturing industries perform batch-wise production where different jobs undergo various machining operations. More number of the machining operations at different machining stations requires scheduling in order to minimize the cycle, delay time and energy consumption. There are various solutions available with different combinations, such kind of problems are termed as “NP Hard Problems”. Various algorithms and software’s are available for solving this type of problems in manufacturing industries, but it provides solution by using traditional algorithms and are expensive too for medium-scale industries. Therefore, there is need of technique which will provide global, unique and feasible solution for complex scheduling problems. Solutions provided by the algorithms are automatically generated, they can be executed by the automatic system like Robotics. Robotics System itself consist of multiple path planning combinations.This project focuses on the automatic material handling and scheduling of machining processes using a six-degree-of-freedom (6-DOF) Kawasaki robot in an industrial setup with two CNC machines. The robot loads a job into the first CNC machine, flips the job, places it into the second CNC machine, and finally unloads it into the finished pallet. The proposed system improves workflow efficiency, minimizes delays, and ensures consistent job handling accuracy. This also enhances worker safety by minimizing human interaction in hazardous areas. The case study conducted in the industrial setup demonstrates the effectiveness of robotic automation in material handling and scheduling, making it adaptable for various manufacturing applications while offering a cost-effective solution for reducing labor expenses. The use of robots for loading and unloading jobs in industries offers significant advantages in terms of reducing labor costs and increasing productivity. By automating the loading and unloading process, industries can reduce dependency on manual labor, minimizing the risk of human errors, inconsistencies and workplace injuries associated with repetitive tasks. Robots ensure precise and accurate placement of jobs, preventing misalignment issues that could lead to defects or machine downtime. Additionally, robotic systems can work in hazardous environments, reducing the need for human workers to be exposed to high temperatures, heavy loads, or sharp tools. With improved efficiency and reduced labor costs, industries experience higher throughput, lower operational expenses, and increased overall profitability. The reliability and speed of robotic loading and unloading contribute to a streamlined manufacturing process, making automation an essential investment for modern industries aiming for higher productivity and consistent quality.

Keywords: CNC machine, Material Handling, labor expenses, productivity.

Abstract: Text-to-Speech (TTS) systems for low-resource languages like Igbo face significant challenges due to dialectal diversity. This research presents a dual-language TTS system for English and Igbo, specifically designed to harmonize the diverse pronunciations and linguistic features across different Igbo dialects. Leveraging a custom model trained on a curated dataset, the system aims to generate natural-sounding speech for both languages. The system is implemented as a web application, providing a user-friendly interface with features like pitch and rate adjustment for English. The system's performance is evaluated using Word Error Rate (WER) on diverse sentences, demonstrating its ability to handle various linguistic complexities within Igbo. This research contributes to enhancing accessibility for Igbo speakers, promoting language preservation, and advancing TTS technology for low-resource languages.

Keywords: Igbo TTS, Dialectal TTS, Low-Resource Language TTS, Multilingual TTS, TTS for Igbo.

Abstract: Training deep neural networks often relies on fixed learning rates and static hyperparameters, which can lead to inefficiencies and suboptimal results [1, 2]. This paper introduces Adaptive Learning via Dynamic Variable Integration (ALDVI), a novel method that dynamically adjusts learning parameters during training. By incorporating auxiliary variables that adapt based on loss and accuracy trends from prior iterations, ALDVI enhances the optimization process and reduces dependence on manually tuned hyperparameters [3]. This adaptive mechanism refines convergence behavior and improves generalization, addressing challenges in training efficiency and robustness [4]. Experimental evaluations on widely used benchmark datasets demonstrate substantial improvements in convergence speed, accuracy, and resistance to hyperparameter sensitivity [5, 6]. These findings highlight ALDVI’s potential as a valuable augmentation to conventional training strategies for deep neural networks.

Keywords: Adaptive Learning, Dynamic Variable Integration, Neural Network Optimization, Hyperparameter Tuning, Convergence Efficiency, Generalization Performance, Deep Neural Networks, Loss and Accuracy Trends, Benchmark Datasets, Robust Training Strategies, Parameter Adjustment, Model Convergence, Training Efficiency, Hyperparameter Sensitivity, Optimization Process

Abstract: LoRaWAN, each node transmits packets autonomously, so packet collisions occur when multiple nodes transmit packets at the same time and frequency. However, the clocks of inexpensive LPWAN nodes are generally not highlyaccurate, and synchronization errors can occur between devices over time. In addition, in single-hop LoRaWAN, it is not possible to achieve high data rates and a wide communication range at the same time. By using multi-hop communication, it is possible to achieve a wide communication range while maintaining a high data rate. However, LoRaWAN multi-hop communication suffers from the hidden node problem, throughput degradation due to the inability to transmit and receive packets simultaneously, and power consumption due to the need to constantly open the receive window. This paper proposes an autonomous distributed adaptive resource allocation method to solve the above problems. Specifically, we show that by assigning packet transmission slot decisions based on LoRaWAN packet information, the transmitting and receiving sides can share transmission and reception locations, avoid packet collisions, and reduce power consumption by avoiding unnecessary opening of reception windows

Keywords: Transmission Timing, Synchronization Control, Energy Efficiency, Multi-Hop LoRaWAN, Wireless Communication

Abstract: GPS uses ECCs to see if an error occurs when the data sent from the satellite reaches the user. Each message structure uses ECCs such as Hamming Code, CRC, BCH Code, and LDPC Code. If the satellite contains all of the en-coders, it has a negative impact to the area and power consumption. Therefore, in this paper, we propose a CRC-BCH unified encoder for GPS, which is efficient in terms of space and power consumption. Since both the CRC and BCH encoders use shift registers, the design was made using this part. To replace the existing encoder, the CRC-BCH en-coder must have the same output. To validate this, we used individual CRC and BCH encoders and confirmed that the generated output was identical to the output of the proposed encoder. The proposed CRC-BCH unified encoder was synthesized at an operating frequency of 400 MHz using the CMOS 28nm process. The synthesis results showed that it used 16.67% less area and consumed 19.68% less power than the existing encoder. Therefore, the proposed CRC-BCH unified encoder offers advantages in terms of satellite weight and energy efficiency.

Keywords: GPS, ECC, CRC, BCH, Encoder.

Abstract: In order to improve energy efficiency and lessen its negative effects on the environment, sustainable agriculture mostly depends on renewable energy sources. Solar energy is one of the most useful and efficient of these. However, the varying angle and intensity of sunshine during the day affects how efficient solar panels are. Our solution includes an automatic solar tracking device that continuously modifies the panel's position to optimize energy absorption. The panel tilts toward the sun using a single-axis solar tracker, greatly increasing energy production and overall efficiency. Our approach creates a hybrid renewable energy system by combining wind and solar electricity to further improve energy reliability. This combination makes it perfect for agricultural applications including irrigation systems, water pump motors, and other crucial farming operations since it guarantees a steady and continuous power supply. The hybrid approach is a more resilient and sustainable option since it minimizes reliance on a single power source while simultaneously optimizing energy generation. Furthermore, a Wireless Sensor Network (WSN) is integrated into our system to effectively monitor and control energy distribution. The Connected Dominating Set (CDS) algorithm is used to optimize the WSN, increasing communication efficiency, reducing sensor node energy consumption, and extending the network's lifespan. Our suggested solution intends to transform smart agriculture by combining advanced WSN-based optimization, hybrid energy harvesting, and intelligent tracking to improve environmental sustainability, lower operating costs, and increase energy efficiency. In addition to meeting the growing demand for sustainable food production, this research advances contemporary farming methods, opening the door for a more eco-friendly and energy-efficient agricultural industry.

Keywords: WNS, CDS, Energy efficiency, Network lifespan

Abstract: The Smart City Project Model is an advanced, integrated solution designed to address the growing challenges of urbanization through the application of IoT (Internet of Things) and automation technologies. This project encompasses a wide range of subsystems, including an automatic entry gate, a dual-axis solar panel tracking system, smart street lighting, waste management, parking, pollution monitoring, and water management. The ESP32 microcontroller serves as the central processing unit, enabling seamless communication and control across all subsystems. The project emphasizes energy efficiency, sustainability, and real-time monitoring, making it a scalable and practical solution for modern urban environments. A web-based dashboard provides a user-friendly interface for monitoring and controlling all systems, ensuring optimal resource utilization and improved quality of life for citizens..

Keywords: Smart City, IoT, Automation, ESP32, Solar Tracking, Waste Management, Pollution Monitoring, Smart Parking, Water Management, Energy Efficiency, Urban Sustainability.

Abstract: Present era is Artificial Intelligence (AI) era. A lot of Machines today mimics human beings. A social person is facilitated from pattern recognition to language understanding, speech recognition, and even visual perception. At its core, AI combines data, algorithms, and computational power to mimic cognitive functions, enabling machines to learn from experience and adapt to new situations without explicit programming. A large number of Machine Learning (ML) Frameworks and natural language processing (NLP) tools exist which supports scientists of different disciplines. This paper aims to provide insight to young researchers about the famous facilities that exist today in terms of ML Frameworks and NLP tools in technology. It also provides a look for future challenges in this field.

Keywords: Machine Learning Frameworks, Natural Language Processing, AI Applications

Abstract: Artificial Intelligence (AI) has transformed industries and everyday life with its ability to automate complex tasks and make predictions based on large datasets. However, one of the biggest challenges with AI, particularly with advanced models such as deep learning, is the lack of transparency. These models, often referred to as "black boxes," provide predictions and decisions, but the reasoning behind them is not immediately clear to users. This lack of inter-pretability has led to the development of Explainable AI (XAI), a set of techniques that aim to make AI systems more transparent, understandable, and trustworthy. XAI is crucial for building confidence in AI, especially in high-stakes areas like healthcare, finance, law, and autonomous vehicles. The aim of this work is to provide a comprehensive guide that delves into the components, methods and techniques, future scope and applications of XAI. It concludes by providing a detailed understanding about XAI, how it enhances AI models by making them more interpretable and accountable. So, it provides a new vision to researchers how they can justify decision making and results with AI.

Keywords: LIME, SHAP, Post-Hoc Explainability, Intrinsic Explainability

Abstract: Artificial Intelligence (AI) is today’s hot field of computer science. It focused on creating systems and ma-chines which have the capability of performing tasks of human intelligence. In research, these tasks range from prob-lem definition, problem-solving, decision-making to result interpretation. A lot of research work is done in this disci-pline by the scientists. This paper aims to provide a look to current research publication in this field. For a road ahead, it also discusses future applications and challenges in AI.

Keywords: Healthcare, Education, Industry revolution, Robotic Machines.

Abstract: Liver disease is diagnosed automatically using artificial intelligence (AI) models trained, tested and validated on liver disease datasets representing clinical or diagnostic measurements encapsulated in biochemical markers like albumin as well as enzymes implicated in metabolic processes. The responses of the trained AI models to new clinical diagnostic input could drive clinical decision-making support. Ultimately, the trained AI models could be packaged into an automated liver disease diagnosis module and merged with a repertoire of modules for the automated diagnosis of a wide range of health conditions within the context of a comprehensive AI-driven healthcare system.

Keywords: Deep Learning (DL), Artificial Intelligence (AI), Liver Disease, Cirrhosis of the Liver, Albumin, Artificial Neural Network (ANN), TensorFlow, Healthcare System, Disease Diagnosis and Prediction.

Abstract: The risk of accidents is very high due to highways, cross-roads, hill station, bunds of the lake and human negligence. Committing small mistakes when driving in these areas may lead to fatality. Despite using signboards, accidents continue to occur because of people’s negligence. Researchers found that 57% of accidents are caused solely by driver related factors. The chances of occurrence of landslide in ghat section are high, especially during the rainy season and during the road construction. The objective of the paper is to build a communication between vehicle using Light Fidelity system and indicating the hazardous signals when the landslide occurs. In this work, we have developed a tool using an embedded system design approach, that incorporates hardware components like Arduino Uno, LCD, Buzzer, Moisture Sensor, DHT11, ServoMotor, Li-Fi transmitter and Li-Fi Receiver, Solar Panel, Potentiometer, ADXL345 and software system like Arduino IDE. The tool is tested in an experimental setup for a scenario that warn potential vehicles of different contexts, such as “LAND SLIDE OCCURRED STAY ALERT” along with the temperature, moisture and humidity of the location.

Keywords: Landslide detection, Li-Fi, Road-accidents, Driving

Abstract: Suspicious transactions in mobile money systems have become a growing concern due to the increasing volume of digital transactions and the sophistication of fraudulent activities. Mobile money platforms, widely used for quick and secure financial transfers, are vulnerable to various types of fraud, such as identity theft, unauthorized access, and transaction manipulation. Detecting fraudulent transactions in real-time is a challenge due to the vast amounts of data and the dynamic nature of fraudulent behaviors. Existing systems often struggle with high false-positive rates or fail to catch advanced fraud patterns, leading to financial loss and security breaches. This paper proposes a secure model for mobile money applications, incorporating two layers of authentication that include passwords, One-Time Passwords (OTP), and a machine learning-based approach. Specifically, the Random Forest classifier is employed to accurately detect suspicious mobile money transactions. Results show that the system ensures a robust and scalable solution to efficiently identify fraudulent transactions with minimal user intervention, achieving 100% training accuracy and 99% testing accuracy, enhancing financial security. This innovation plays a crucial role in safeguarding mobile money platforms globally.

Keywords: Mobile money, One-Time Passwords, Random Forests, Suspicious Transactions

Abstract: The integration of Artificial Intelligence (AI), Blockchain, and the Internet of Things (IoT) has revolutionized multiple industries, with healthcare and automation being at the forefront. This paper provides a comprehensive review of recent advancements in these fields, highlighting innovative solutions, challenges, and future prospects. It discusses the role of AI in personalized patient care, robotics in automation, and blockchain for secure healthcare data management. The research synthesizes findings from multiple studies and presents insights into the impact of these technologies on healthcare accessibility, operational efficiency, and data security. Furthermore, this study explores the synergistic relationship between these technologies and their potential in revolutionizing patient care, hospital management, and secure data exchange.

Keywords: Artificial Intelligence, Blockchain, Internet of Things, Healthcare, Automation, Robotics, AI-powered IoT, Smart Hospitals, Secure Data Management, AI-driven Wearables, Predictive Analytics.

Abstract: This project proposes an IoT-based smart garbage monitoring and segregation system that utilizes sensors, robotic arms and real-time notification system to optimizes waste management. The system monitors garbage bin fill levels, temperature , humidity and gas emissions Real-time data is transmitted to the Municipal corporation via Wi- Fi/GSM , ensuring timely emptying and maintenance. This system aims to reduce manual labor and provide data-driven insight for efficient waste management.

Keywords: IoT, Smart Garbage Monitoring, Waste Segregation, Sensors, Robotic Arm, Real-time Notification, Wi-Fi, GSM, Municipal Corporation, Data-driven Insights, Waste Management Optimization.

Abstract: This research paper introduces a novel smart hybrid drone system integrated with a 5G base station to enhance land survey capabilities, real-time data transmission, and surveillance applications. The proposed drone leverages hybrid propulsion technology to optimize flight efficiency and endurance while integrating edge computing and AI-driven analytics for intelligent decision-making. The study evaluates the feasibility, architecture, and potential use cases of the system across various domains, including infrastructure assessment, security surveillance, and geographic mapping. Experimental results demonstrate the effectiveness of the hybrid drone system in maintaining high-speed, low-latency communication and efficient power management, making it a viable solution for next-generation autonomous aerial networks.

Keywords: Smart Drone, Hybrid Propulsion, 5G Base Station, Edge Computing, Land Survey, Aerial Surveillance, Real-time Communication

Abstract: This paper offers a thorough examination of how User-Centered Design (UCD) can greatly enhance website usability and performance. It starts by outlining the significance of usability in web design and providing an overview of UCD, laying the groundwork for a thorough analysis of its function in producing user-friendly digital experiences. The theoretical background explores usability and performance metrics, the fundamental principles of user-centered design (UCD), and a comprehensive analysis of recent literature on identifying usability issues using this approach. The discussion then delves into different methods and approaches for identifying usability issues, showcasing both qualitative and quantitative techniques. The text provides an in-depth exploration of various user research methods, including surveys, interviews, and usability testing. It also delves into the use of tools and frameworks such as eye-tracking and think-aloud protocols. Various data analysis techniques are explored to effectively synthesize user feedback. The paper explores the benefits of incorporating usability feedback into the design and development process to achieve iterative improvements and enhance website performance through UCD. The paper features case studies that demonstrate the successful implementation of UCD and how it has improved performance. The analysis covers the challenges of eliciting usability issues, such as obstacles to obtaining reliable feedback, ethical concerns, and balancing conflicting stakeholder needs. It explores future trends and innovations in usability testing, including the impact of AI and machine learning, emerging user-centered approaches, and the significance of predicting future user behavior. Ultimately, the paper underscores the importance of consistently enhancing usability and performance, while also acknowledging the crucial role of user-centered design (UCD) in driving progress in web development. Designers and developers can enhance the effectiveness, engagement, and responsiveness of websites by embracing a user-centered approach.

Keywords: User-Centered Design (UCD), Usability, Website Performance, Usability Testing, Iterative Design, User Research, Emerging Technologies

Abstract: In today's fast-paced digital era, businesses are rapidly transforming by embracing electronic transactions to streamline their operations. With the increasing reliance on digital platforms, companies are actively leveraging social networking apps to attract and engage with customers. This shift has given rise to a new form of commerce known as "social commerce," where social media networks facilitate buying and selling activities. However, the adoption of new technologies in various business processes introduces a plethora of security challenges. One significant concern is the rampant use of social media for product marketing, which has led to the proliferation of fake information and counterfeit product reviews. These deceptive practices can significantly influence consumer purchasing decisions, complicating the ability of business organizations to provide accurate and trustworthy marketing information. Addressing these security issues requires robust models and frameworks that can effectively safeguard against cyber threats. This paper delves into the realm of cybersecurity, examining its frameworks and strategies for protecting personal information in the digital landscape. By conducting a systematic review of existing literature, the author aim to uncover the cybersecurity challenges faced by business organizations and their customers. The study explores various prevention methods to mitigate cybercrime and ensure customer safety. Additionally, it offers valuable insights and recommendations for future research to further enhance cybersecurity measures. Keywords: Cybersecurity Frameworks, Cyber Threats, Risk Management, Malware

Abstract: From early rule-based systems to sophisticated deep learning architectures supporting today's most advanced applications, artificial intelligence (AI) has fast progressed. This work offers a thorough investigation of the several forms of artificial intelligence depending on their functionality and cognitive capacity. From reactive machines and restricted memory systems to theoretical concepts including artificial general intelligence (AGI) and artificial superintelligence (ASI), we explore the historical development of artificial intelligence. After a synthesis of the most important results, the research approach consists in a thorough evaluation of foundational papers and the most recent investigations. Results show that even if restricted artificial intelligence rules contemporary applications (e.g., natural language processing, computer vision, and autonomous systems), major obstacles still exist for the evolution of AGI and self-aware systems. Furthermore, influencing the upcoming generation of intelligent systems are developing themes include neuro symbolic integration, edge artificial intelligence (XAI) and explainable artificial intelligence (XAI). As artificial intelligence develops, the conversation emphasizes ethical, transparent, and biassed problems that have to be resolved. At last, we suggest future directions of research to close present gaps and guarantee that artificial intelligence develops in a way that is both technically strong and morally sound.

Keywords: Artificial Intelligence, Narrow AI, Artificial General Intelligence, Superintelligence, Machine Learning, Deep Learning, Natural Language Processing, Computer Vision, Neuro symbolic AI, Edge AI, Explainable AI

Abstract: Globally, pests are responsible for destroying up to 20–40% of annual crop yields, resulting in economic losses exceeding 1.5 lakh crores. Excessive pesticide use to combat pests not only increases farming costs but also contributes to environmental degradation. This project presents an AI-Powered Pest Detection System to address these challenges. Utilizing advanced computer vision, the tool identifies harmful insects in agricultural fields with high accuracy, enabling farmers to take early action and prevent infestations. Real-time detection and integration with drones or cameras enhance surveillance and support precision agriculture. By targeting pest issues promptly, the system reduces pesticide reliance, supports crop health, and maximizes yields. Its efficiency and adaptability make it ideal for large-scale farms, promoting sustainable farming practices and contributing to global food security through proactive crop management.

Keywords: Pest Detection, Deep learning, YOLOV8, Open CV, AI in Agriculture

Abstract: Driver fatigue is a well-documented risk factor for road accidents worldwide. Prolonged driving hours and inadequate rest significantly impact a driver's cognitive alertness, leading to a decline in reaction time and decision-making ability. This study provides an extensive review of current advancements in intelligent anti-sleep eyewear designed to detect and mitigate drowsiness in drivers. The paper examines key technologies such as Electroencephalography (EEG), Electrooculography (EOG), Photoplethysmography (PPG), eye-tracking, and head pose estimation, highlighting their efficacy, limitations, and feasibility for real-time application. Additionally, machine learning techniques for data analysis and classification are explored, focusing on their role in improving detection accuracy. The review further outlines future directions in multimodal sensing integration, adaptive learning algorithms, and human factors research to enhance the reliability and usability of these systems. Rigorous testing and real-world implementation remain crucial for ensuring the efficacy of these solutions in practical driving environments.

Keywords: Driver fatigue, anti-sleep eyewear, drowsiness detection, physiological signals, machine learning, wearable sensors, road safety.

Abstract: The integration of data analytics, artificial intelligence (AI), and cloud computing has significantly transformed the healthcare industry, particularly in optimizing Medicaid and enhancing patient outcomes. This paper delves into the latest advancements in predictive analytics, AI-driven diagnostics, blockchain integration, and cloud-based healthcare solutions aimed at improving efficiency. Through a comprehensive literature review, this study examines the role of AI and data-driven decision-making in enhancing data quality, interoperability, and workforce training. The findings highlight emerging trends, challenges, and future directions, paving the way for continued research and innovation in healthcare technology.

Keywords: Artificial Intelligence, Data Analytics, Cloud Computing, Healthcare Optimization, Predictive Analytics, AI-Driven Diagnostics, Blockchain, Interoperability, Medicaid, Healthcare Innovation.

Abstract: Many technical advancements have led to the formation of the Internet of Things (IoT), which facilitate the worldwide networking of internet-enabled objects. These gadgets, which are often referred to as "smart gadgets," have the ability to send, receive, and manage data. It is a widely held belief that the IoT is a technology seeing rapid growth and user acquisition. The flawless transfer and receipt of data is essential to the Internet of Things' functionality. Moreover, it is essential to provide an outstanding level of Quality of Service (QoS) and avoid severe energy constraints for battery-powered devices. One emerging trend in the IoT is the proliferation of networked gadgets. To safeguard the gadgets and the data they generate, stringent security measures are being implemented. The purpose of this work is to examine approaches that have been developed to protect source and sink nodes in order to prevent data breaches and unauthorized access. The purpose of this study is to examine the algorithms, with a special emphasis on the Ad-hoc On-Demand Distance Vector (AODV) protocol, on the QoS of IoT -connected networks. Selective data recognition is made possible by QoS systems, which maximize network traffic use. By using this technique, the network's reach is increased, information is used as efficiently as possible, and prompt delivery of the best internet service is ensured. The goal of this project is to put the K-Nearest Neighbors (KNN) method to use in detecting and eliminating malware, or malicious software. The accuracy of forecasts was greatly improved by using these models cautiously. MATLAB software was used in the development of the models. By using these algorithms, the study provides information that can be used to progress the resilience and security of Internet of Things networks against possible attacks.

Keywords: K–Nearest Neighbor, Internet of Things (IoT), Quality of Service (QoS), Malware detection, Wireless Network, Security

Abstract: Sign language acts as a key communication method for individuals with hearing difficulties, but obstacles often arise in conversations between sign language users and the general population. People engage with one another to express their ideas, feelings, and experiences, but this is not true for those who are disabled or mute. Sign language enables individuals who are mute to communicate without relying on vocal sounds. The goal of this project is to create a system that can recognize sign language, facilitating communication between individuals with speech impairments and those without, thus bridging the communication divide between them. In comparison to other forms of gestures like those made with the arms, face, head, and body; hand gestures hold significant importance as they convey a person's thoughts more swiftly. The system employs machine learning and computer vision methodologies to identify and understand hand gestures linked to sign language. The Sign Language Translator system is designed to close this communication gap by converting sign language gestures into text or speech instantly.

Keywords: hand gestures, machine learning, hearing impairments, deaf.

Abstract: As sensors become smaller, channel bandwidth (BW) increases, and internet connectivity speeds increase, Wireless Body Area Networks, or WBANs, are becoming more and more significant. An internetwork is a vast network of networks with thousands or even millions of nodes and links. Any biological stimulus from the human body is transformed into an electrical signal, standardized, and sent to the internetwork. This study examines the possibility of a human body implanted with biomedical sensors that measure many physiological parameters and run wireless protocols with various frequencies. This study examines the possibility of a human body implanted with biomedical sensors that measure many physiological parameters and run wireless protocols with various frequencies. A wireless body area network is made up of a number of nodes that are connected to one another to create a network of biomedical or other sensors positioned at the nodes. Military personnel stationed in remote areas require constant health monitoring, and the packets must be transmitted to the base station. The headquarters (HQ) must be connected to each base station. Additionally, the data must be encrypted and authorized. The enemy will get an advantage from any incursion or information breach. The research paper focuses on network construction, authentication, and encryption. To determine the optimal way and anticipate the path between the sender and the recipient, we employ specific routing protocols. EIGRP (Enhanced Interior Gateway Routing Protocol) and OSPF (Open Shortest Path First) are recommended. For traffic, we also use two-way authentication. By combining data from several patients, this configuration helps medical professionals make better clinical decisions and promote collaborative care models. To better understand trends in the prevalence of chronic diseases, for example, researchers might use aggregated data. Podiatric abnormalities due to improper gait of Army men with and without Diabetes Mellitus is also dealt with using Artificial Intelligence (AI) models developed using supervised classification algorithms- Support Vector Machine (SVM).

Keywords: Wireless Body Area Network (WBAN), internetwork, authentication, encryption, OSPF (Open Shortest Path First) and EIGRP (Enhanced Interior Gateway Routing Protocol), Podiatric abnormalities, improper gait, Diabetes Mellitus, Artificial Intelligence models (AI), Supervised Classification Algorithms, Support Vector Machine (SVM).

Abstract: Emotion detection in text is a complex yet vital component of natural language processing, significantly contributing to enhanced human-computer interaction. This research investigates the effectiveness of various word embedding techniques—Fast Text, RoBERTa, and Glo Ve—when combined with Convolutional Neural Networks (CNN) for detecting emotions. The study categorizes emotions into five types: happiness, anger, sadness, fear, and surprise. Three datasets are analyzed: one comprising movie reviews, another consisting of customer feedback from e-commerce platforms, and a hybrid dataset merging the two. Results indicate that RoBERTa+CNN outperforms other combinations, achieving accuracy rates of 89.45%, 90.12 % and 89.87% on the respective datasets. FastText+CNN is the second-best performer, while GloVe+CNN achieves the lowest accuracy. Additionally, evaluation metrics such as Precision, Recall, and F1-Score highlight the superior performance of RoBERTa+CNN in text-based emotion detection. This study underscores the value of contextual embeddings like RoBERTa in improving the reliability of emotion recognition models.

Keywords: Emotion Detection, Text Classification, Word Embeddings, RoBERTa, FastText, GloVe, Convolutional Neural Networks.

Abstract: Our initiative presents a centralized platform that streamlines participation and improves data-driven insights in response to the many obstacles university students encounter while accessing extracurricular activities. In addition to offering students simple access to information about different student clubs, such as club types and event registrations, the platform acts as a comprehensive center and includes a secure payment method for paid events. Gathering student interests allows student organizations/clubs as well as university management to understand what students are interested in technically as well as in cultural type of events. By utilizing data visualization specific to each event’s category and student interests, the platform derives actionable insights that successfully inform university management and student organizations/clubs by exposing trends of student interest. In conclusion, the goal of our initiative is to facilitate club involvement, maximize event participation for students through having their preferred events in the university.

Keywords: Centralized Platform, Event Management, Event registration, Student Clubs/Organizations, University Management, Students Interests, Data Visualization, Data-Driven Insights, React+Vite, Firebase, D3.js, Razor-Pay API

Abstract: Weed management is a critical challenge in agriculture, affecting crop yields and sustainability. Traditional methods, such as manual weeding and blanket herbicide spraying, are labor-intensive and environmentally harmful. This paper presents an AI-driven Weed Detection and Management System that utilizes deep learning models like YOLOv8 and Convolutional Neural Networks (CNNs) to accurately detect and classify weeds in real time. By integrating computer vision, precision agriculture techniques, and automated herbicide application, the system minimizes chemical usage and improves farming efficiency. Experimental results demonstrate over 92% accuracy in weed detection, making this system a scalable solution for modern agriculture.

Keywords: Weed Detection, Deep Learning, YOLO, Precision Agriculture, Machine Learning, AI in Farming

Abstract: This paper presents a facial emotion recognition The main aim of this is Human facial expressions are an important medium for communication and a signal for emotions like happiness, sadness, anger, and surprise. The objective of this project is to develop a Facial Emotion Recognition (FER) system that can detect a person’s emotions and create a matching bitmoji correlated with the detected expression. The way the system works is that, from a camera placed in front of the user, it will capture the face and the expression of the user through deep learning. When the system will recognize the emotion, it will create a bitmoji representing that detected feeling. This makes online communication more fun and interesting. This technology, thus, can be implicated into social media, gaming, virtual meetings, customer service, and even mental health applications. This technology increases personalization and thus improves human-computer interaction. Future improvements of this system can include more emotions, improved accuracy, and merging along with augmented reality (AR) for an engaging experience.

Keywords: Deep Learning, Computer vision, CNN, Emotion based avatar

Abstract: In the modern era, balancing professional responsibilities and parenting is a significant challenge for many parents. Due to demanding work schedules or the absence of adequate support systems, parents often struggle to provide their infants with the consistent attention and care they require. The proposed work presents the development of a smart cradle system that combines real-time monitoring, automation, and safety features to support both home infant care and neonatal care units. By integrating advanced IoT technologies, the cradle addresses key challenges faced by caregivers, such as detecting baby cries, monitoring mattress wetness, maintaining optimal environmental conditions, and ensuring proper illumination in low-light settings. The system also includes automatic rocking, soothing music with rotating toys, live video streaming for remote supervision, and an alarm system for prolonged crying. Furthermore, motion detection algorithms distinguish between a sleeping and awake baby, enabling better insights into the infant’s activity patterns. In neonatal care units, this smart cradle reduces the workload of caregivers by automating routine tasks while ensuring the infants’ comfort and safety. Through its user-friendly design and intelligent features, the smart cradle offers a holistic approach to modern baby care, demonstrating its potential to enhance parental confidence and optimize neonatal care environments.

Abstract: Voting is an essential part of any democracy but traditional methods based on paper ballots often face risks like security breaches, fraud, and voter intimidation. The opportunity to solve these issues comes with applying blockchain technology where voting is decentralized, secure, and transparent. This document analyzes how block chain can facilitate e-voting while dealing with problems such as voter impersonation, ballot secrecy, and election fraud. Utilizing crypto security, smart contracts, and decentralization, trust and accessibility of the electorate can surely be improved. We evaluated blockchain e-voting systems and their pros and cons as well as innovations that will follow in these systems. Voter impersonation and tampering have been long standing issues in electioneering and voting and recently emerged electronic voting scheme provides a possible solution to these problems. However, it has introduced complicated issues regarding security, credibility, transparency, functionality and most importantly, reliability. Using blockchain technology in e- voting improves security by effectively addressing threats of fraud and vote tampering. Such systems are virtually impossible to hack as the vote ledger is decentralized and can be verified by anybody.

Keywords: Blockchain; E-Voting; Decentralization; Smart Contracts; Cryptographic Security

Abstract: Customer feedback analysis plays a crucial role in helping organizations improve their products and services. Using Natural Language Processing (NLP) techniques, this project converts unstructured customer feedback into meaningful insights. The analysis utilizes tools such as NLTK, along with models like Bag of Words (BoW) and advanced deep learning frameworks such as Transformers. The process starts with data preprocessing steps like tokenization, removal of stop words, and lemmatization, efficiently handled by the NLTK library. The Bag of Words model transforms text into numerical data for sentiment classification and topic identification, though it lacks the ability to grasp context. To overcome this, Transformers are employed, offering contextual understanding and accurate sentiment detection. By combining traditional methods like BoW with the sophisticated capabilities of Transformers, this project ensures precise and scalable analysis of customer feedback. This integration enables companies to address user concerns promptly and enhance customer satisfaction. 

Keywords: Automated Review System, Semantic Analysis, Natural Language Processing (NLP), NLTK (Natural Language Toolkit), Bag of Words (BoW), Transformers, Tokenization

Abstract: In today’s competitive retail environment, accurate demand forecasting is essential for effective inventory management and customer satisfaction. This research explores a machine learning approach for retail demand forecasting that not only uses historical sales data but also integrates social media trends and local event data to capture real world demand influencers. Unlike many existing models that focus solely on past sales patterns, this approach leverages big data sources to refine predictions and address fluctuating consumer preferences. The proposed model employs algorithm such as Random Forest and XGBoost to analyze a broad array of data and enhance forecasting accuracy. A key feature of this system is its ability to automatically adjust forecasts based on real-time social sentiment, allowing retailers to respond dynamically to shifts in customer interest, such as sudden demand spikes for trending products. The forecasting results are visualized through an interactive React-based dashboard, enabling retailers to quickly access demand insights. With a cloud-based backend for data processing and storage, this solution ensures scalability and timely data handling, helping businesses make data-driven inventory and supply chain decisions.

Keywords: Demand forecasting, retail, machine learning, big data, real-time analytics, social media trends, React, Random Forest, XGBoost.

Abstract: Nova is a voice-controlled desktop assistant designed to simplify user interactions and enhance productivity. Utilizing advanced speech recognition, prompt engineering, and data-driven decision-making, Nova facilitates hands-free control of applications and provides real-time feedback via text-to-speech. Unlike traditional NLP-based solutions, Nova leverages structured datasets and JSON-based prompts to streamline interactions, reduce processing overhead, and prioritize user privacy. Developed using PyQt5 for a native graphical user interface (GUI) and python PyQT5 functions for backend Connection Nova offers a comprehensive, cross-platform solution tailored for accessibility, multitasking, and efficient computing environments. Keywords: Speech Recognition, Desktop Assistant, Datasets, TTS, Automation, Prompt Engineering.

Abstract: This study examines the evolving role of artificial intelligence (AI) in the field of human resource management (HRM),shedding light on its transformative capabilities and obstacles. A comprehensive literature analysis spanning from 2010 to 2020 revealed a disjointed body of knowledge regarding AI integration in HRM and underscored the necessity for additional research into human-AI synergy. Principal discoveries suggest that while AI can boost productivity and enable data-informed decision-making, it also generates concerns about managerial power disparities, ethical quandaries, and potential job losses. Our investigation demonstrates that AI's impact on HRM is multidimensional and shaped by contextual elements such as organizational leadership and employee competencies. As a result, we propose a balanced strategy that emphasizes enhancing human capabilities through AI rather than substituting them. This study advocates for thorough research on the characteristics that promote successful AI adoption among workers and organizations, as well as the ramifications of AI on employee autonomy. By addressing current research gaps, we aim to contribute to a more nuanced comprehension of AI's bla in HRM and emphasize the crucial role of HR professionals in guiding its responsible implementation.[1]

Keywords: Human Resource Management Artificial Intelligence (AI), Human-AI Collaboration, AI Adoption, Organizational Culture, Employee Skills, Ethical Considerations, Data-Driven Decision Making, AI Bias, Performance Enhancement

Abstract: Ensuring infant safety and well-being is a primary concern for parents and caregivers. This paper presents an AI driven baby monitoring system that leverages computer vision and deep learning techniques to track infant sleep patterns, detect anomalies (crying, woke up), and provide real-time alerts. Our proposed approach integrates convolutional neural networks (CNNs) and recurrent neural networks (RNNs)to analyze video feeds for crying, sleeping, and facial expressions, enhancing monitoring accuracy.The system aims to provide a reliable solution for reducing risks associated with sleep disorders and sudden infant health issues. Our experiments demonstrate the effectiveness of our approach in detecting abnormal movements and sleep condition with high precision.

Keywords: Baby Monitoring, AI for Healthcare, Sleep Tracking, Deep Learning, Infant Safety

Abstract: This Paper presents a Strain Analysis Based on Eye Blinking This cloud of a world has splattered towards more than normal screen hours, leading towards executable eye strain plus exhaustion. This system focuses on resolving eye strain by “monitoring eye blinking patterns.” Staring passively is a behavior that can be recorded from live video feeds of individuals, like how many times a person blinks and how long each time the individual looks at the screen. By watching these movements, the system determines if the user struggles with visual fatigue or exhaustion. It is possible for a person who suffers a lot from eye strain. In this step, the accuracy of the system will be determined. In the future, expanding the datasets and fine-tuning the accuracy of user comfort levels for more than just different lighting conditions or variations plus activities performed on the screen can make this more efficient. Another step we can take is the cross-device implementation so every device can use it. Overall simple eye care goals were achieved- making life easy in this world of digitization and screens surrounding us everywhere.

Keywords: Electrooculography, Blink Duration, Frequency, Deep learning techniques, Convolutional neural network (CNN), Fatigue.

Abstract: Diplomate is your one-stop solution for excelling in your studies. We offer a offer a comprehensive platform designed to help you succeed in your diploma studies and score the marks that you want. Study Materials and Notes: Diplomate offers well-organized and curated study materials, tailored for diploma courses. These materials include summaries, detailed notes, and other essential resources to help students grasp complex concepts easily. Organized Content: The app structures the content in a user-friendly manner, ensuring that students can quickly find what they need without overwhelming them. Topics are grouped logically to align with the course syllabus, making study sessions more efficient. Study Schedules and Reminders: Diplomate helps students plan their study time with customizable schedules. It also sends timely reminders for upcoming exams, assignments, and revision sessions to keep students on track. Resource Library and References: The app includes an extensive library of additional learning resources like textbooks, journal articles, and reference materials, all in one place, so students don’t have to search elsewhere. Career Guidance and Counseling: Diplomate provides personalized career advice and counseling, helping students make informed decisions about their post-diploma career paths, whether they aim for further studies or entering the workforce. Personalized Learning Experience: The app adapts to the unique needs of each student by offering tailored recommendations and study plans based on individual progress and goals, creating a more focused and effective learning journey.

Abstract: The Computerized Examination Engine (CEE) is an online exam monitoring system designed to ensure a fair and secure examination process. This system allows teachers to create and schedule exams while providing students with a platform to take tests efficiently. CEE enhances the traditional online examination system by integrating advanced security features to prevent cheating. It includes a tab change detection system, which warns students if they switch tabs or minimize the browser. After multiple warnings, the exam is either terminated or restarted. Additionally, camera access is mandatory to monitor students during the exam. If a student does not grant camera permissions, they cannot start the test. The system also captures screenshots and records screen activity to further prevent dishonest practices. Developed using HTML, JavaScript, and MySQL, CEE ensures a user-friendly interface for both teachers and students. The platform is efficient, reliable, and provides a seamless examination experience while maintaining integrity and fairness in online assessments.

Keywords: Secure examination, teachers, students, schedule exams, take tests, security features, prevent cheating, tab change detection, warnings, exam termination, exam restart, camera access, monitor students, camera permissions, screenshots, screen recording

Abstract: This project focuses on developing a fully functional, secure, and user-friendly donation platform that connects individual donors and orphanages seamlessly. Key features include robust user authentication and registration processes, efficient donation management systems for donors and orphanages, enhanced backend security using Supabase, and an improved user experience with a clean and intuitive interface. The platform addresses security concerns by implementing proper data encryption, verification checks, and access controls, ensuring a safe environment for all users. Performance optimization measures are in place to enhance speed and responsiveness, with bug fixes and code cleanup contributing to a smooth user experience. By meeting these project goals and delivering a well-rounded solution, the platform aims to streamline the donation process while prioritizing security, usability, and efficiency.

Abstract: Customer analytics is crucial for data-driven decision-making in today’s cutthroat business environment, but there isn’t a single platform that combines sales forecasting, churn prediction and Customer lifetime value (CLV) assessment. We provide an AI-driven framework that integrates these three essential features into a single system in order to close this gap. Our method uses a Bidirectional Long Short-Term memory (BLSTM) network for sales forecasting to capture intricate temporal patterns, XGBoost for churn prediction to identify at-risk customers, and the Gamma-Gamma model for CLV estimation to predict future customer spending. Through the integration of various models, our system offers a thorough and precise understanding of consumer behavior, empowering companies to maximize customer engagement and revenue expansion. Experimental results demonstrate superior predictive performance over traditional approaches, making this a valuable tool for organizations seeking to enhance their customer analytics capabilities.

Keywords: Customer Lifetime Value (CLV), Churn Prediction, Sales Forecasting, AI-driven analytics, Gamma-Gamma model, XGBoost, Bidirectional Long Short-term memory (BLSTM), Customer retention, Predictive modelling, Business Intelligence.

Abstract: E-waste is a pressing global challenge, with improper disposal threatening the environment and public health. In 2022, an estimated 62 million tons of e-waste were generated worldwide, and this number is projected to surpass 75 million tons annually by 2030. Countries like China, the United States, and India are among the largest contributors. Existing solutions have limitations in addressing this growing crisis. We propose a novel "E-Waste Reward Club" to integrate collection, public engagement, and authorized recycling in an automated and user-friendly manner. By incentivizing proper e-waste disposal and integrating technology with public participation, this approach aims to revolutionize e-waste management.

Keywords: E-waste, eco RC Machines, Sustainability, Recycling, Environmental Management, Public Engagement.

Abstract: The integration of Artificial Intelligence (AI) into education is transforming traditional teaching and learning paradigms, offering personalized and adaptive learning experiences. AI allows educators to customize teaching to meet each student's unique needs, enhancing engagement and boosting results. The advancement of artificial intelligence, algorithms, and methods has made personalized learning (PL) solutions an effective way to improve learning performance.These include machine learning techniques to predict learner preferences, adaptive learning systems that adjust content dynamically, and natural language processing tools for real-time feedback. AI-powered language learning tools, automated grading systems, and immersive virtual learning environments represent other promising applications in education. These systems utilize Data Analytics to track learner progress, identify knowledge gaps, and provide targeted support to help students achieve their objectives. The objective of this review is to know the impact of AI-driven personalization on learner’s performance. And to explore how the implementation of AI-powered adaptive learning platforms influences academic achievement, engagement levels, and overall satisfaction among learners.

Keywords: Artificial Intelligence(AI), Personalized Learning, Machine Learning, Data Analytics and Adaptive learning systems.

Abstract: Enhanced CAT-DM is an advanced virtual try-on system that combines diffusion models with GAN-based initialization to attain greater realism, efficiency, and controllability. It builds on the Garment-Conditioned Diffusion Model (GC-DM) and incorporates DINO-V2, a top-performing self-supervised vision model, for fine-grained, pixel-level garment representations. ControlNet is also used to improve conditioning accuracy such that garments fit naturally onto body shapes. In order to speed up the normally time-consuming sampling process of diffusion models, we propose a truncation-based acceleration method that leverages a GAN-synthesized coarse image as an initial guess. This largely minimizes the number of sampling steps needed without compromising high-fidelity garment details. In addition, Poisson blending is employed to blend the synthesized garments into the target person's image with seamless transitions and realistic texture conservation. Extensive assessments on benchmark datasets show that Enhanced CAT-DM beats current virtual try-on techniques in terms of higher LPIPS, SSIM, and CLIP-I scores, which confirm its superiority in retaining high-level details, structural features, and garment semantics. All these innovations render Enhanced CAT-DM highly appropriate for real-time, high-fidelity virtual try-on scenarios, filling the gap between AI-based garment synthesis and real-world usability in fashion and e-commerce sectors.

Keywords: Virtual Try-On, Diffusion Models, Generative Adversarial Networks (GANs), Garment-Conditioned Diffusion Model (GC-DM), ControlNet, DINO-V2, Truncation-Based Acceleration, Poisson Blending.

Abstract: The quest for immortality has captivated humanity for millennia, from religious beliefs to scientific advancements. Today, technology is redefining the concept of eternal existence through the emergence of AI cloud consciousness. This research paper explores the complex landscape of digital immortality, delving into the science behind mind uploading, the ethical dilemmas, and the role of cloud computing and blockchain in preserving a person's digital legacy. The human desire to transcend mortality has been a driving force throughout history and across diverse cultures. Traditional methods, such as religious beliefs, cryonics, and genetic legacy, have long been the primary avenues for pursuing immortality. However, the rapid advancement of technology is now reshaping this age-old quest, giving rise to the concept of AI cloud consciousness as a new frontier of eternal existence.

Keywords: Artificial Intelligence, Cloud Computing, Data Centers, Self-Adaptive Systems, Autonomous Operations, Quantum, Cloud Networks, Supercomputing

Abstract: This paper presents the research and development of a solar power inverter as an alternative energy solution. With increasing power outages in rural and suburban areas, there is a dire need for reliable and renewable energy sources. This project focuses on designing a solar-powered UPS system that can provide backup energy during grid failures. The proposed system utilizes a solar panel to convert solar energy into electrical energy, stores it in a battery, and uses an inverter circuit to convert DC to AC, making it suitable for household and industrial applications.

Keywords: Solar Power, Inverter, Renewable Energy, UPS System, Sustainable Energy.

Abstract: Fire hazards in industrial environments can lead to catastrophic losses. An early detection system can significantly mitigate risks. This paper presents an IoT-based industrial fire detection system using Arduino Nano, OLED display (128×64), MQ-2 gas sensor (for LPG detection), Flame Sensor (for fire detection), DHT11 (for temperature and humidity monitoring), and a Buzzer. The system provides real-time monitoring and alerts users with distinct buzzer sounds for different alerts. The OLED display operates in a slideshow format to accommodate all sensor readings within the limited display space. The proposed system ensures early fire detection with improved accuracy and efficiency.

Keywords: IoT, Fire Detection, Industrial Safety, Arduino, Sensors, Embedded System.

Abstract: This project aims to develop an online learning platform focused on programming education, addressing the challenges students face in acquiring coding skills. The platform will support multiple programming languages, including Python, Java, JavaScript, and C++, with personalized learning paths for beginner, intermediate, and advanced learners. Key features include interactive learning modules with video tutorials, quizzes, and coding exercises, as well as coding challenges and competitions to reinforce learning through practical application. A mentorship system will connect students with industry professionals for guidance and career insights. Additionally, the platform will offer project-based learning opportunities, allowing students to build a portfolio of real-world applications. A progress tracking system will help learners monitor their achievements, with certifications provided upon course completion. The platform will continuously improve through student and mentor feedback, ensuring relevance and engagement. Built with React for the frontend, Node.js and Express for the backend, and MongoDB for data management, the platform will provide a comprehensive and engaging learning experience tailored to the needs of aspiring programmers.

Keywords: Programming Education, Interactive Learning, Mentorship, Project-Based Learning

Abstract: Automated Machine Learning (Auto ML) is transforming network and service management by making it easier to apply machine learning methods to intricate network settings. In order to facilitate quicker deployment and more effective network operations, auto ML platforms automate a number of machine learning workflow processes, including as data pretreatment, model selection, and hyperparameter customization.

Abstract: This project presents an interactive coding quiz website designed to enhance programming knowledge through a structured question-answer format. The quiz features multiple levels, a timer, dark mode, and navigation buttons for a user-friendly experience. Developed using HTML and CSS in VS Code, the platform includes a welcome page, a scoring system, and a final result display. By integrating a dynamic and engaging interface, this project aims to make coding practice more efficient and enjoyable for learners.

Keywords: Coding Quiz, HTML, CSS, Web Development, Interactive Learning, Timer, Dark Mode, User-Friendly Interface, Programming Education, Quiz Navigation, Scoring System.

Abstract: Object detection is a crucial field in computer vision that allows machines to identify and track objects in real time. This project develops a live object detector using yolov8, integrated with tkinter to provide an interactive and user-friendly graphical interface. The system captures live video through a webcam, processes frames with yolov8, and delivers real-time object detection with voice feedback using pyttsx3. The application features a graphical user interface (gui) that enables users to start and stop detection, toggle voice alerts, and view detected objects dynamically. The yolov8 model is chosen for its high efficiency and accuracy, while opencv and pillow (pil) are used for real-time image processing and display. The system also employs a queue-based text-to-speech mechanism to provide audio notifications when objects are detected. A movement detection feature ensures real-time tracking of detected objects, updating the interface accordingly. This project has practical applications in security surveillance, inventory management, accessibility support, and smart home automation. The ability to detect and identify objects in real time makes it useful for various domains, including education, healthcare, and transportation. The voice feedback feature enhances accessibility, making it beneficial for visually impaired users. Future enhancements may include custom training of yolo models, multi-object tracking, integration with iot devices, and edge computing for improved performance. The project demonstrates the potential of ai-driven object detection in real-world applications, showcasing its effectiveness in developing intelligent vision-based systems.

Keywords: Yolov8, Tkinter, OpenCV, PIL (Pillow), Pyttsx3, Threading, Queue, Real-time Object Detection, GUI, Webcam, Text-to-Speech, Machine Learning, Computer Vision, Deep Learning, Python.

Abstract: The increasing prevalence of lifestyle-related health issues has underscored the necessity for individuals to take proactive measures toward their fitness and well-being. This project presents the development of a Health Fitness App designed to assist users in monitoring and improving their health through a comprehensive approach to fitness management. The app integrates various features, including workout tracking, nutrition logging, personalized fitness plans, and progress analytics. Utilizing modern technologies such as cloud computing and data analytics, the app provides users with a user-friendly interface that allows for seamless interaction and real-time data synchronization across devices. Users can set fitness goals, log their daily activities, and track their dietary intake, thereby gaining insights into their overall health and fitness levels. The app employs advanced algorithms to recommend personalized workout routines and meal plans tailored to individual user profiles, including age, weight, fitness level, and dietary preferences.

Keywords: Health Fitness App, Health issues, workout Tracking, Progress Analytics, AI based workout recommendations, Fitness goal, Smart habit tracking

Abstract: Agriculture is the foundation of any prosperous society. The newest revolution caused by improved technology in this sector would bring about a significant change; that is the proposed idea, which presents a platform for the farmers to rent agricultural equipment necessary for that moment to help enhance resource utilization and wastage. Some important features of this platform include multilingual access, a strong database for transactions, and a mechanism for feedback and rating driven by the user. The platform also works with predictive models for pest and disease management, weather forecasting, crop selection, and market price estimation. It further facilitates transportation assistance and loans as additional benefits. Such a great solution is at its best in addressing the underutilization of machines and promoting efficient practices for sustainable agriculture. The intuitive interfaces and innovative features also make it a one-stop solution for farmers to improve resource management and achieve sustainable solutions in the agricultural domain.

Keywords: Farming, Agriculture, Rental, Next JS, Web App, Farmer Multilingual Support, Agricultural Machinery Rental.

Abstract: Artificial Intelligence has revolutionized human-computer interaction, and Character AI is at the forefront of this transformation. This project explores the development of Character AI, an advanced AI-powered conversational system capable of understanding, adapting, and engaging users in meaningful conversations. The system leverages Natural Language Processing (NLP), deep learning, and contextual awareness to generate human-like responses, making digital interactions more immersive and personalized. The project involves creating an interactive AI-driven system with key features such as personality customization, adaptive learning, real-time sentiment analysis, multimodal communication (text, voice, and visual expressions), and AI-driven moderation to ensure ethical and safe interactions. Character AI can be utilized across various domains, including entertainment, customer support, virtual companionship, and education. This research demonstrates how AI-driven characters can bridge the gap between technology and human emotions, enhancing user engagement in a dynamic digital environment.

Abstract: Geofencing combines awareness of the user's current location with awareness of the user's proximity to locations that may be of interest. To mark a location of interest, you specify its latitude and longitude. To adjust the proximity for the location, you add a radius. The latitude, longitude, and radius define a geofence, creating a circular area, or fence, around the location of interest. You can have multiple active geofences, with a limit of 100 per app, per device user. For each geofence, you can ask Location Services to send you entrance and exit events, or you can specify a duration within the geofence area to wait, or dwell, before triggering an event. You can limit the duration of any geofence by specifying an expiration duration in milliseconds. After the geofence expires, Location Services automatically removes it.

Keywords: Geofencing, Location Based Services, GPS Tracking, Real Time Location Monitoring, Geo Boundaries, Virtual Parameter, Proximity Alerts, Wi-Fi Positioning System, IoT (Internet of Things), GPS (Global Positioning System), AI-powered Location Analytics.

Abstract: Cricket umpiring, particularly for Leg Before Wicket (LBW) decisions, plays a crucial role in determining match outcomes. Traditional umpiring methods rely on human judgment, which can sometimes lead to errors due to limited reaction time and viewing angles. To improve accuracy and minimize human error, this paper presents an AI-based cricket umpire system that utilizes computer vision and polynomial curve fitting for LBW decision-making. The system processes video frames, detects the cricket ball using HSV color segmentation, tracks its movement, and predicts its impact using trajectory analysis.

Keywords: Cricket, AI Umpire, LBW Decision, Computer Vision, OpenCV, Trajectory Prediction, SciPy, NumPy.

Abstract: Computer vision and artificial intelligence are transforming industries by enabling automated image analysis, object detection, and real-time decision-making. However, the complexity of existing AI vision platforms often creates barriers for developers, requiring costly hardware and extensive technical expertise. Percept AI aims to bridge this gap by offering a web-based AI vision platform that allows users to run OpenCV projects directly in a browser. By eliminating installation challenges and leveraging cloud-based execution, Percept AI makes AI vision more accessible to researchers, students, and developers. It integrates features such as real-time image and video processing, automated code generation, federated learning, and a collaborative AI community. This paper explores Percept AI’s technical innovations, its role in democratizing AI vision development, and its potential to drive future advancements in computer vision.

Keywords: Computer Vision, Percept AI, OpenCV, Federated Learning, Collaborative AI, AI Vision Tools, AI-Driven Cybersecurity.

Abstract: This paper focuses on developing a comprehensive sentiment analysis system for customer reviews, combining traditional machine learning and advanced deep learning techniques. The system classifies reviews into positive, negative, or neutral categories through robust text preprocessing, feature extraction, and model training. Traditional classifiers like Random Forest, Naive Bayes, Logistic Regression, and SVM are utilized alongside advanced NLP models such as VADER for quick analysis of short reviews and BERT for an in- depth understanding of longer, context-rich reviews. The system employs ensemble methods to enhance accuracy and consistency in sentiment classification. It evaluates performance through metrics such as accuracy, precision, recall, and F1-score to ensure reliability and scalability. A user- friendly Flask-based web application enables seamless dataset uploads, real-time analysis, sentiment visualization, and downloadable results. The project aims to provide an efficient and accurate sentiment analysis solution adaptable to diverse e-commerce platforms and customer feedback scenarios

Keywords: Sentiment Analysis; NLP; VADER; BERT; Customer Reviews; TF-IDF.

Abstract: This project develops a web-based language translator utilizing JavaScript and HTML to provide real-time translation capabilities, enabling users to communicate across language barriers. The translator supports multiple languages, including English, Spanish, French, and more, allowing users to input text in one language and receive the translated text in another language instantly. Leveraging JavaScript for client-side scripting and translation logic, and HTML for structuring and displaying the user interface, this project demonstrates the capabilities of web technologies for language translation. The project's architecture is designed to be scalable and flexible, allowing for easy integration with external translation APIs or libraries, and supporting future enhancements such as speech recognition for voice-to-text translation. With its user-friendly interface and real-time translation capabilities, this project aims to provide a convenient and accessible translation tool for individuals and organizations alike, facilitating global communication and collaboration. Overall, this project showcases the potential of JavaScript and HTML for developing innovative and practical language translation solutions.

Keywords: Language Translation, Natural Language Processing (NLP), Machine Translation, Real-time Translation, Speech Recognition, Offline Translation, Artificial Intelligence (AI)

Abstract: The rapid evolution of decentralized technologies has paved the way for innovative solutions in various sectors, including energy management and trading. This paper presents a novel Peer-to-Peer (P2P) Electricity Billing and Trading system leveraging blockchain technology to enhance transparency, security, and efficiency in energy transactions. The proposed system integrates smart contracts, a custom blockchain, and transaction management to facilitate seamless energy trading between consumers and producers. At the core of the system is a custom blockchain that ensures immutable and transparent recording of all transactions. Smart contracts automate the execution of agreements between parties, eliminating the need for intermediaries and reducing transaction costs. The system also incorporates a P2P trading mechanism, allowing customers to directly trade electricity, fostering a decentralized energy market. The architecture includes an ‘Admin module’ for system oversight, a WEB3J interface for interacting with the blockchain Smart Contracts, and a ‘Customer module’ for user engagement. This design not only simplifies the billing process but also empowers consumers to participate actively in the energy market. By leveraging blockchain's inherent security features, the system mitigates risks associated with fraud and data tampering. In conclusion, this P2P Electricity Billing and Trading system represents a significant step towards a more decentralized and efficient energy ecosystem. It offers a robust framework for future advancements in blockchain-based energy solutions, promoting sustainability and consumer empowerment.

Keywords: Blockchain Technology, Peer-to-Peer (P2P) Trading, Smart Contracts, Electricity illing, Decentralized Energy Market, Transaction Management.

Abstract: An innovative approach to enhance the early detection of multiple eye diseases, including glaucoma, cataract, diabetes-related eye conditions, and various infections. The significance of early diagnosis in preventing irreversible vision impairment cannot be overstated, and emerging technologies in the fields of machine learning (ML) offer unprecedented opportunities to revolutionize ophthalmic healthcare. The proposed system employs a comprehensive dataset comprising diverse instances of eye diseases to train a sophisticated ML and DL model. Leveraging state-of-the-art algorithms, including Recurrent neural networks (RNNs) and convolutional neural networks (CNNs), two of the model's methods, are designed to evaluate a variety of ocular properties extracted from medical images. These features include structural abnormalities, textural patterns, and contextual information, enabling the system to discriminate between healthy and diseased conditions with a high degree of accuracy. To validate the effectiveness of the developed model, extensive experimentation will be conducted using a diverse set of real-world eye images sourced from clinical databases. The project aims not only to achieve high accuracy in disease identification but also to optimize the model for real-time applications, ensuring its practical utility in clinical settings.

Keywords: Deep Learning, Convolutional neural networks, Machine learning, Eye diseases, Medical images

Abstract: Malware detection plays a crucial role in cybersecurity by identifying and mitigating threats posed by malicious software. Traditional detection methods rely heavily on signature-based approaches, which are often ineffective against new, evolving malware. This paper presents a deep learning-based model for malware detection, leveraging advanced neural network architectures to classify files as either benign or malicious based on their characteristics. By training the model on a comprehensive dataset, it learns to identify subtle patterns that distinguish harmful files from legitimate ones. Enhancing the accessibility and usability of the detection system, the model is integrated into a web-based interface where users can upload files and receive real-time analysis results. Experimental results demonstrate the effectiveness of the deep learning approach in achieving high accuracy and detection speed, showcasing its potential as a proactive tool for modern cybersecurity defence.

Keywords: Malware detection, Deep Learning, Cyber Security, Neural Networks, Malicious files.

Abstract: Medical identification of plants serves important functions for healthcare systems as well as pharmaceutical development and protection of biodiversity. The identification process which depends on manual techniques demands extensive time from experts so automated solutions prove to be crucial. This paper investigates deep learning techniques for medicinal plant classification through combination of ResNet and EfficientNet structures. The training of our model utilized a large database consisting of medicinal plants which incorporated EfficientNet and ResNet architectures to extract complex leaf patterns together with their textual features and color schemes in the leaves. Users can access Plantify system through its easy-to-use web interface which provides functionality for botanists and researchers along with healthcare professionals to submit plant images for instant classification. A collection of medicinal plant pictures served as input for model training and evaluation through which their main visual characteristics including leaf styles along with textures and color patterns were analysed. The experimental outcomes prove that EfficientNet surpasses traditional models both in accuracy performance and computational efficiency requirements which makes it appropriate for mobile application usage.

Keywords: Convolutional Neural Networks, EfficientNet, ResNet, Medicinal Plants, Image Processing,Machine Learning.

Abstract: Blockchain technology is a revolutionary innovation with immense disruptive potential and can change many sectors. Health care is one of the sectors that will greatly benefit from using this technology, with critical requirements for secure, transparent, and efficient data systems. Decentralization and blockchain’s tamper evident structure make patient confidential information more resilient against unauthorized access and data manipulation. Blockchain also facilitates the improvement of interoperability between health care providers, data sharing, and clinical processes. This paper will describe how blockchain addresses patient data security issues, fraud prevention, and compliance with regulatory standards. It further talks about the role of IT managers in adopting blockchain through strategic planning, team training, and robust security measures to enable seamless integration and optimal functionality.

Keywords: Blockchain, Healthcare, Data Security, Interoperability, Fraud Prevention.

Keywords: IR Sensor, Arduino, DC Motor, Water Pump.

Abstract: Hypertension, a leading risk factor for cardiovascular diseases, requires early detection to prevent severe health complications. This paper presents a hybrid machine learning model integrating Random Forest, XGBoost, Support Vector Machine (SVM), Multi-Layer Perceptron (MLP), and Logistic Regression using a voting-based ensemble method. The dataset is pre-processed with SMOTE to handle class imbalance, and features are normalized for optimal performance. The proposed model achieves an accuracy of 89%, outperforming individual classifiers. The results indicate that ensemble learning significantly enhances prediction reliability.

Keywords: Hypertension, Machine Learning, Hybrid Model, Ensemble Learning, SMOTE

Abstract: The Baby Vaccination System project aims to develop a comprehensive platform for managing child vaccination schedules efficiently. It incorporates frontend and backend development to provide users with tools for vaccine registration, tracking, and locating nearby vaccine centres. By targeting efficient management of vaccination schedules, the system contributes to public health initiatives and ensures timely immunization for children. The objectives of the study include assessing the effectiveness of the system, evaluating user satisfaction, and determining the impact on vaccination completion rates. Participants include parents or guardians of children eligible for vaccination, healthcare providers, and system administrators. Ethical considerations include obtaining informed consent, ensuring participant confidentiality, and adhering to relevant ethical guidelines. The study findings will provide insights into the usability and effectiveness of the Child Vaccination System in improving vaccination management.

Keywords: Vaccination System, Immunization, Public Health, Technology Integration , Remainder System,Chatbot integration.

Abstract: This research paper focuses on developing a text-to-image generator platform that leverages advanced artificial intelligence models to transform textual descriptions into high-quality, customizable visuals. By utilizing state-of-the-art generative models like Stable Diffusion, the platform aims to provide users with an intuitive and efficient tool for creating images tailored to their needs. Targeted at a diverse audience, including artists, designers, marketers, and casual users, the system bridges the gap between creativity and accessibility. It addresses current limitations in customization and ethical concerns by integrating enhanced user control, robust content moderation, and ethical AI practices. The platform offers cloud-based access for ease of use, as well as local deployment options for professional users with high-performance hardware. This solution democratizes visual content creation, enabling users of all skill levels to bring their ideas to life quickly, efficiently, and affordably.

Keywords: Text-to-image, AI, Stable Diffusion, Image Generator

Abstract: Medical diagnosis has seen significant advancements with the application of deep learning, offering improved solutions for healthcare challenges. This study focuses on the detection and classification of three critical diseases: heart disease, pneumonia, and diabetic retinopathy (DR). Each disease presents distinct diagnostic complexities, including variations in data representation and the need for accurate predictions. To address these challenges, the proposed system integrates CNN, ResNet, MobileNet, and DenseNet architectures, forming a robust and efficient diagnostic framework.The proposed framework incorporates CNN, Resnet, MobileNet and DenseNet architectures to build a robust system capable of addressing these challenges. Users can leverage the system through a user-friendly interface designed for healthcare professionals, providing rapid and accurate disease classification. The experimental results validate the effectiveness of the proposed deep learning framework, positioning it as a valuable tool for assisting in early diagnosis and medical decision-making. The combination of state-of-the-art architectures ensures both accuracy and computational efficiency, making the system suitable for real-time clinical applications.

Keywords: Deep Learning, MobileNet, DenseNet, ResNet, Machine Learning.

Keywords: Intrusion Detection, Machine Learning, Support Vector Machines (SVM), Random Forest, Network Security, UNSW-NB15, Malicious Network Activities Detection

Keywords: Artificial Intelligence (AI), Data Analytics, Cloud Computing, Medicaid Optimization, Predictive Analytics, Healthcare Interoperability, AI in Healthcare, Machine Learning, Healthcare Data Management, AI Ethics, Workforce Training, AI-Driven Decision-Making, Cost Optimization, Healthcare Automation, Blockchain in Healthcare, IoT in Healthcare.

Abstract: Softball pitching is a biomechanically complex skill that integrates principles of kinematics, kinetics, and energy transfer to optimize velocity, accuracy, and spin of the ball while minimizing the risk of injury. Unlike baseball pitching, which relies on an overhand motion, fast-pitch softball utilizes a windmill-style underhand motion involving a full 360° arm rotation. This unique movement pattern requires effective use of Newton’s laws of motion, torque, angular momentum, and the kinetic chain for successful execution. Over the past three decades, extensive biomechanical studies have explored the phases of softball pitching, including wind-up, stride, arm-cocking, acceleration, release, and follow-through. These studies employ motion capture systems, force plates, and electromyography (EMG) to examine kinematic and kinetic characteristics of the pitching motion. Findings consistently indicate that lower-extremity drive, trunk rotation, and shoulder angular velocity are critical determinants of ball speed, while inefficient mechanics predispose pitchers to overuse injuries, particularly in the shoulder and elbow. This review synthesizes current literature on the mechanical principles underlying softball pitching, highlighting performance determinants, injury risk factors, and training implications. By consolidating empirical evidence, it underscores how biomechanical knowledge can guide coaching strategies, athlete development, and future research directions in softball pitching.

Keywords: Softball pitching; biomechanics; windmill motion; kinematics; kinetics; Newton’s laws; kinetic chain; torque; angular momentum; injury prevention.