Course generation as a hierarchical task network planning problem

Course generation automatically assembles courseware adapted to the learner’s competencies and learning goals. Previous course generators offered only a limited representation a pedagogical knowledge and did not take full advantage of serviceoriented architectures. This dissertation proposes a course generation framework (called Paigos) that enables the formalization and application of complex and realistic pedagogical knowledge. Compared to previous course generation, this approach generates structured courses that are adapted to a variety of different learning goals and to the learners’ competencies. Additionally, Paigos is service-oriented, thus allowing the integration of learning-support services into generated courses in a generic and pedagogically sensible way and enabling third-party learning environments access to its functionality using a Web-service interface. Paigos was evaluated by technical, formative and summative evaluations. The technical evaluation primarily investigated the performance to Paigos; the formative and summative evaluations targeted the users’ acceptance of Paigos and of the generated courses.