<p>The rapid expansion of the Internet of Things (IoT) has made managing user mobility and resource allocation in distributed computing environments a critical challenge. Traditional cloud computing systems face inherent latency and bandwidth limitations that hinder real-time data processing and resource efficiency, especially in dynamic mobile environments. To address these challenges, this paper explores the use of machine learning techniques to optimize user mobility management in a mobile fog computing environment. Specifically, we use the K-means and Self-Organizing Maps (SOM) clustering algorithms to enhance the classification of mobile users, minimize energy consumption, and improve resource allocation at the network edge. By leveraging fog computing’s proximity to end users, our approach reduces the energy demands of traditional cloud systems while improving performance in user mobility management. The experimental results show that SOM outperforms K-means and both algorithms deliver superior performance compared to the random selection method. The findings underscore the potential of ML-driven strategies in addressing mobility and resource allocation challenges in mobile fog networks, offering scalable and adaptive solutions for user mobility in highly dynamic environments.</p>