Model-based reasoning about learner behaviour
Artificial Intelligence
Using self-diagnosis to adapt organizational structures
Proceedings of the fifth international conference on Autonomous agents
| Hi-index | 0.01 |
New technologies associated with learning environments on the web demand more interactivity and require such environments to adapt themselves to students' individual characteristics. In this paper, we model learning tasks as a network of plans generated by a teacher, which represents alternative ways for students to reach their learning goal. Student's actions (regarding the material observed, time taken, and results achieved) are recorded and analyzed during the navigation through the learning material. A domain-independent diagnostic process compares the log of the actual navigation and the plans. The result of the process is used not only to alert students, but also to adapt existing plans or to generate new plans that enhance the library of alternative ways learning goals can be successfully achieved.