Volume 13 | Issue 5 |

 Abstract

This project focuses on developing an emotion recognition model that simplifies the process into key steps like data collection, feature extraction, and real-time deployment, while also considering ethical implications and user-friendliness. By accurately interpreting emotions from speech, facial expressions, and body language, such a model can enhance digital interactions by providing emotionally aware feedback, crucial for decision-making and improving user experience in various applications. To ensure the model performs effectively across various scenarios, a multimodal approach is utilized, combining audio-visual information and contextual data for enhanced emotion recognition. The system is trained using deep learning and machine learning algorithms on an extensive dataset that captures diverse emotional expressions. Implementing this model in real-time applications can be particularly beneficial in areas such as virtual communication tools, educational platforms, and healthcare systems, where emotional intelligence is essential for meaningful and responsive interactions.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Emotion Detection, Deep Learning, NLP, Multimodal, Real-Time Processing

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Explainable Artificial Intelligence (XAI) is a field of AI that focuses on making machine learning models transparent, interpretable, and understandable to humans. As AI systems become increasingly complex and integral to decision-making in areas like healthcare, finance, and autonomous systems, the need for interpretability grows to ensure trust, fairness, and accountability. XAI techniques aim to provide insights into model predictions, helping users understand the rationale behind AI-driven decisions. Methods such as feature importance analysis, SHAP (Shapley Additive Explanations), LIME (Local Interpretable Model-Agnostic Explanations), and decision trees enable a balance between model performance and interpretability. The adoption of XAI not only improves user trust but also ensures compliance with ethical and regulatory standards like GDPR. This paper explores various XAI techniques, their applications, challenges, and future directions in bridging the gap between AI's predictive power and human interpretability.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Explainable AI (XAI), Transparency in AI, Post-Hoc Methods, SHAP and LIME, AI in Healthcare and Finance.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

India's foreign investment in Africa has emerged as a vital component of its broader diplomatic and economic engagement with the continent. This paper examines the growing role of Indian foreign direct investment (FDI) in Africa, highlighting key sectors such as pharmaceuticals, information technology, agriculture, infrastructure, and energy. It explores how these investments are driven by mutual interests, including access to natural resources, expanding markets, and strategic partnerships in the Global South. The study also assesses the impact of Indian investment on African development, technology transfer, employment generation, and capacity building. Moreover, it situates India's approach within the broader context of South-South cooperation, comparing it with China's investment model. Through an analysis of bilateral agreements, institutional mechanisms, and private sector initiatives, the paper underscores India's evolving role as a development partner in Africa.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Foreign Direct Investment (FDI), India-Africa Relations, South-South Cooperation, Economic Diplomacy, Infrastructure Development, Technology Transfer, Sustainable Development, Private Sector Engagement, Strategic Partnership, Resource Mobilization, Capacity Building, Bilateral Trade, Development Cooperation, Pharmaceutical Industry, Energy Security, Agricultural Investment, Digital Economy, Employment Generation, Geopolitical Influence, Multilateral Institutions.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The AI-Based Career Guidance System is an intelligent platform designed to assist students and recent graduates in making informed career decisions by offering personalized recommendations. It addresses the common challenges faced by individuals who are uncertain about their future paths due to limited awareness, rapidly changing job markets, and a lack of proper guidance. By utilizing machine learning and natural language processing techniques, the system analyzes a wide range of user data, including academic background, personal interests, acquired skills, and long-term aspirations. This data is collected through a user-friendly web-based interface that simplifies the process of inputting relevant information. The AI models then process this data to identify patterns and correlations between the user's profile and successful career trajectories in the current job market. Based on this analysis, the system suggests tailored career options that align with both the individual's strengths and market demands. Additionally, a feedback mechanism allows users to rate and review the recommendations, enabling the system to continuously improve its predictive accuracy and relevance through iterative learning. This adaptive approach ensures that the guidance provided remains up-to-date and aligned with evolving industry requirements.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Machine Learning Algorithm, Google ML Kit, Cosine Similarity, Natural Language Processing, Career Recommendation System, Pattern Recognition

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Federated Learning (FL) is an emerging decentralized machine learning approach that enables model training across multiple devices or clients without transferring raw data to a central server. By maintaining data locality, FL enhances data privacy and reduces the risks associated with data breaches. This method is particularly suitable for sensitive domains like healthcare, finance, and IoT applications. The report explores the architecture, core mechanisms, advantages, challenges, and security strategies of FL, including techniques such as differential privacy and secure multi-party computation. The paper also discusses different types of FL such as Horizontal, Vertical, and Federated Transfer Learning and highlights its applications and future research directions

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Artificial Intelligence , Federated Learning , Decentralized Training , Data privacy , Edge Devices

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The rapid advancement of Internet of Things (IoT) technology has revolutionized home automation by enabling seamless remote control and monitoring of smart devices. This project focuses on designing and implementing a smart home automation system using the Blynk IoT platform, which provides an intuitive interface for managing various household appliances via smartphones or other internet-enabled devices. The system is built around the ESP8266 Wi-Fi module, known for its low power consumption, integrated TCP/IP stack, and efficient processing capabilities. With a 32-bit Tensilica L106 processor operating at 80-160MHz, 16MB flash memory support, and Wi-Fi Direct (P2P) connectivity, the ESP8266 ensures reliable real-time communication between devices. Additionally, its GPIOs, ADC input, and multiple concurrent TCP connections allow for the integration of various sensors and actuators, making the system highly adaptable. This project explores the setup, configuration, and implementation of a smart home automation system that controls devices such as lights, thermostats, and security cameras. The integration of Blynk enhances the system's flexibility, scalability, and ease of use, allowing users to monitor and manage their home environment efficiently. By leveraging IoT-based automation, this project demonstrates how smart homes can improve convenience, energy efficiency, and security. The findings and implementation steps serve as a valuable resource for IoT enthusiasts, researchers, and homeowners looking to adopt smart home solutions.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Internet of Things (IoT)

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

As the production of electric vehicle has scaled up in the recent years in order to meet the objective of lowering the carbon footprint and eco-friendly transportation. The raise in the production of electric vehicles in accordance with hike in the price of petroleum and diesel has shifted the huge market share of the automobile industry from ICE engines to the battery powered engines. The shift in turn pushes the demand for installation of charging stations for electric vehicles at most locations. But the installation of such EV charging base stations requires high capital and sophisticated spatial infrastructure in densely populated area. Therefore, the paper proposes the AIS based mobile wireless charging system for electric vehicles which is cost effective and reliable. The system is best suited for densely populated areas, parking arenas at theatres, malls, parks etc., The Wireless Sensor Network is implemented to effectuate adaptive intelligent system, therefore leading to better accuracy and modularity.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Battery Management, Wireless Charging, Electric Vehicles, IOT, Battery Performance Optimization

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The increasing integration of technology in education has paved the way for more dynamic and accessible learning environments. This project presents the development of an EdTech website, a comprehensive online learning platform designed to enhance educational experiences for students, instructors, and administrators. Leveraging the MERN stack (MongoDB, Express.js, React.js, Node.js), the system offers interactive tools such as quizzes, personalized content, and real-time analytics. Key features include secure user authentication, course management, responsive UI, and scalable cloud deployment using services like Vercel Render, and MongoDB Atlas. Agile development methodologies guided the project's progress, with continuous stakeholder engagement and iterative feedback. The result is a secure, engaging, and user-centric platform. This initiative highlights the potential of EdTech solutions in revolutionizing modern education.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

EdTech, MERN Stack, Online Learning, Agile Development, MongoDB Atlas, Vercel

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Abstract: This study explores the use of machine learning techniques, focusing on the Random Forest Classifier, to predict the directional shifts of the S&P 500 stock market index. The research frames the problem as a binary classification challenge, aiming to determine whether the market will rise or fall on the subsequent day. By analyzing historical OHLCV (Open, High, Low, Close, Volume) data collected over a period of three decades, we evaluate the model's reliability and predictive capabilities. The approach incorporates rolling-window backtesting, with performance assessed through precision and confusion matrix metrics. Our results suggest that the Random Forest model demonstrates competitive effectiveness, especially in low volatility environments, and establishes a solid foundation for more sophisticated ensemble and hybrid methodologies.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Stock Market Prediction, S&P 500, Random Forest, Machine Learning, Time Series, Financial Forecasting, Backtesting, OHLCV

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Organizational commitment is vital in determining the long-term stability and success of any company. This study aims to evaluate the level of commitment and workforce involvement in a private finance company. The results reveal significant relationships between motivation and employee performance, with specific areas identified for improvement. Various factors such as motivation, job satisfaction, leadership, and performance were studied using both primary and secondary data. Statistical tools like Chi-Square Test, Mann-Whitney U, Kruskal-Wallis H Test, Spearman Correlation were used for analysis. The findings show that performance and motivation are strongly correlated, as well as areas that need work.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Keywords: (Motivation, Job Satisfaction, Employee Performance, Workforce involvement)

  License

Creative Commons Attribution 4.0 and The Open Definition