Eugen Harinda

This review paper provides a comprehensive analysis of recent advancements in leveraging reinforcement learning (RL) and deep reinforcement learning (DRL) techniques for optimizing traffic signal control systems within urban environments. Traditional traffic signal management methods often fail to address the inefficiencies resulting in congestion, delays, environmental impacts, and driver dissatisfaction. In contrast, RL methodologies empower traffic signals to dynamically learn and refine signal timing strategies by iteratively interacting with real-world traffic conditions. DRL builds upon RL by integrating deep neural networks to better capture and respond to intricate traffic patterns. The paper investigates the potential of RL and DRL approaches in addressing diverse urban traffic challenges across peak and non-peak periods, as well as in areas with irregular traffic distributions. It delves into the strengths, weaknesses, underlying principles, and advanced architectures of current RL/DRL methodologies proposed for this domain. Furthermore, it evaluates practical applications and simulation studies, showcasing performance enhancements over traditional signal control methods. In conclusion, the review consolidates key findings and outlines future research directions for implementing RL/DRL systems in real-world scenarios to effectively alleviate urban traffic congestion.

Internet of things (IoT) field is quickly developing, with billions of devices deployed worldwide to provide IT solutions. It is an intersection point between computer science and electronic engineering, meaning that learning the fundamentals of electronic engineering is critical for computing students pursuing a career path in this field. This research aims to investigate how simulation-based learning affects computing students' confidence in developing IoT solutions. A mixed-methods approach was used, utilising pre and post-surveys with both closed and open-ended questions. Results indicate a significant increase in student confidence after the intervention, with a paired samples t-test showing a statistically significant improvement (p = 0.007). Data reliability was confirmed using Cronbach's alpha, yielding values of 0.924 and 0.859 for the pre and post-surveys, respectively. The findings have been validated using the triangulation approach. The findings suggest that the use of simulation-based learning significantly improved students' confidence and practical skills in IoT development. Reference to this paper should be made as follows: BenMubarak, M., Harinda, E. and Ihsan, M. (xxxx) 'Investigating the impact of simulation-based learning on computing students' confidence in developing internet of things solutions', Int.). His research interests include networks, wireless communication, IoT, AI and machine learning. He has international teaching experience in Yemen, Malaysia, Saudi Arabia, and the UK and has published in peer-reviewed journals and conferences. Eugen Harinda has a PhD and is a Lecturer at the University of Bolton, specialising in IoT, Data Science, and Artificial Intelligence. He is also a Fellow of Advance HE. His academic interests span applied machine learning, 2 M. BenMubarak et al. with a current focus on computer vision and deep learning techniques. His recent research explores intelligent systems capable of visual recognition and automated decision-making. He actively contributes to interdisciplinary projects that harness AI to solve real-world problems. With a background in both academic teaching and research, he is committed to advancing innovation in data-driven technologies and mentoring the next generation of AI professionals. Mansoor Ihsan is a Lecturer in Computing at the University of Bolton and holds a PhD degree in Computer Sciences and MSc in Data Telecom and Networks. He is also a Fellow of Advance HE. He is a computer networks researcher with expertise in sensor networks, cybersecurity and network security. He has experience in teaching a range of subjects including Linux OS, programming, cybersecurity and computer networks. He has a particular research interest in IoT, sensor networks, cybersecurity and cloud security. His current research area involves security in IoT and cloud leveraging machine learning technologies.

As Large Language Models (LLMs) continue to advance in their ability to process natural language, their potential to transform industries and reshape the future of human-computer interaction becomes increasingly evident. This study evaluates the application of Large Language Models (LLM) to automate SQL query generation from natural language inputs in enterprise environments. We investigated the feasibility of using open-source LLMs, including Mistral, CodeLlama, Phi-3, and DeepSeek Coder, by fine-tuning them with a custom dataset reflecting company-specific data tables. This dataset was iteratively constructed using a baseline LLM that allows fine-tuning to address unique enterprise data structures and use cases. A two-stage filtering and refinement mechanism was implemented to improve query accuracy. The first stage identifies relevant tables and the second stage adds an iterative error correction step to improve the SQL query generation process. The resulting system significantly reduced query errors and increased accuracy by 84%,88%,81%, and 90% for Mistral, CodeLlama, Phi-3, and DeepSeek Coder LLM. However, challenges remain in resource consumption and logical error handling. Future work will focus on improving contextual understanding and integrating advanced AI techniques to further enhance robustness and applicability.