At King Abdulaziz University, I developed a self-supervised framework for learning from MOOC clickstream data and predicting student dropout. My models achieved 80.91% accuracy on KDDCup and 72.70% on XuetangX.
I used SHAP to explain model outputs and K-Means to identify behavioral patterns. I also integrated student learning-preference profiles with the Gemini API to generate personalized feedback and instructor reports.
My research includes a peer-reviewed IEEE Access publication on student navigational behavior in MOOCs, and a manuscript on personalized feedback is under peer review. I hold an M.Sc. in Computer Science and Microsoft Certified: Azure AI Fundamentals.
