Applying Case-Based Planning to Personalized E-learning

Sequencing of Learning Objects (LOs) has been an important issue in the last decades. From a technical perspective, we can take advantage of Artificial Intelligence (AI) planning techniques that allow us to adapt these sequences to pedagogical and students’ requirements. However, there is neither a standard way to represent and compile such LO knowledge into a planning model, nor an optimal way to deal with changes during the execution of the previously adapted learning sequences. In this paper, we propose a general and effective approach to automatically extract information from the LOs to create planning domains, which are then solved by case-based plan merging techniques that are also used as a recommender system. This way of proceeding allows the teacher to store, and reuse, the best learning routes for each student’s profile and course objectives. When discrepancies on the student’s profile or state are detected during the course execution, the system assists the teacher in readapting, repairing or improving the route to fit the new objectives.