Urban traffic congestion remains an ongoing issue that requires advanced traffic management solutions. Accurate traffic forecasting plays a crucial role in Intelligent Transportation Systems, helping to mitigate congestion and improve mobility. Traditional machine learning approaches have been widely used for prediction tasks, often relying on large volumes of historical data for training. However, real-time adaptability is essential for dynamic traffic conditions. In this study, we leverage real-time traffic data and employ ADAptive GRAdient Descent, an online learning method that adaptively adjusts learning rates, allowing efficient updates as new data become available. To evaluate its performance, we implemented our approach on traffic data from a network of streets in Muscat, Oman, demonstrating its ability to provide accurate and timely congestion forecasts.

Traffic congestion remains a critical challenge in urban mobility, necessitating adaptive and intelligent regulation strategies. This paper presents a novel, generic framework that integrates a Multi-Agent System (MAS) layer within a traffic simulator to optimize traffic flow through simulation-driven decision-making. The proposed system allows Dynamic Message Signs (DMS) to function as autonomous agents that assess congestion levels, exchange information, and collaboratively determine rerouting strategies. Unlike conventional approaches, our framework enables agents to negotiate multiple rerouting strategies offline, test them in simulation, and store the most effective ones in a learning module. When congestion occurs in real-world conditions, the agents perform a similarity check within their learning module; if a matching scenario is found, the best prevalidated strategy is immediately applied, eliminating the need for real-time simulations. Otherwise, new rerouting strategies are generated and tested offline for future optimization. The framework ensures that only reliable and effective strategies are deployed, leading to more adaptive and data-driven traffic regulation. Preliminary simulation results demonstrate significant reductions in total waiting time, highlighting the efficiency of integrating MAS-based learning into traffic management systems.

The integration of Machine Learning (ML) into education is transforming traditional learning environments by enabling personalized, data-driven instructional methods. As conventional models increasingly fall short in addressing diverse student needs, ML technologies offer adaptive learning solutions that tailor content, pace, and teaching strategies to individual learners. In this paper, we investigate the application of ML in education. Although global progress has been notable, specific challenges persist in regions such as the Gulf Cooperation Council, including concerns about data privacy, infrastructure limitations, resistance to pedagogical change, and the need for culturally localized AI tools. Through a mixed-method research design, this study evaluates the effectiveness of a ML-based personalized learning system compared to traditional approaches. Surveys, academic performance metrics, and teacher observations are analyzed to assess outcomes related to student engagement and satisfaction.