LAUSR

Educational process mining is now a promising method to provide decision-support information for the teaching–learning process via finding useful educational guidance from the event logs recorded in the learning management system. Existing studies mainly focus on mining students' problem-solving skills or behavior patterns and intervening in students' learning processes according to this information in the late course. However, educators often expect to improve the learning outcome in a proactive manner through dynamically designing instructional strategies prior to a course that are more appropriate to students' average ability. Therefore, in this article, we propose a two-stage problem-solving ability modeling approach to obtain students' ability in different learning stages, including the pre-problem-solving ability model and the post-problem-solving ability model. The models are trained with Gradient Boosting Decision Tree (GBDT) on the historical event logs of the prerequisite course and the target course, respectively. With the premodel, we establish the students' pre-problem-solving ability profiles that reflect their average knowledge level before starting a course. Then, the instructional design is dynamically chosen according to the profiles. After a course completes, the post-problem-solving ability profiles are generated by the postmodel to analyze the learning outcome and prompt the learning feedback, in order to complete the closed-loop teaching process. We study the modeling of coding ability in computer programming education to show our teaching strategy. The experimental results show that the generalizable problem-solving ability models yield high classification precision, while most students' abilities have been significantly improved by the proposed approach at the end of the course.