Self-Regulated Learning (SRL) strategies are essential for learners to effectively plan, monitor, and reflect on their learning progress, ultimately enhancing problem-solving skills. This research focuses on real-time detection of SRL strategies within a Computer-Based Learning Environment (CBLE) by analyzing click-stream and log data. Using process mining and machine learning, we aim to identify key programming actions that indicate SRL strategies and develop adaptive interventions that provide personalized guidance. The study employs a mixed-methods approach, combining qualitative and quantitative techniques such as surveys, interviews, and statistical modelling to gain a comprehensive understanding of SRL in programming contexts. The goal is to design intelligent learning systems that deliver real-time feedback and scaffold learners’ SRL strategies, enabling them to become more independent and efficient problem solvers. This research contributes to AI in Education, Learning Sciences, and Computer Science by advancing SRL detection and fostering meaningful, data-driven educational interventions.