Knowledge Representation Requirements for Intelligent Tutoring Systems

In this paper, we make a first effort to define requirements for knowledge representation (KR) in an ITS. The requirements concern all stages of an ITS’s life cycle (construction, operation and maintenance), all types of users (experts, engineers, learners) and all its modules (domain knowledge, user model, pedagogical model). We also briefly present and compare various KR formalisms used (or that could be used) in ITSs as far as the specified KR requirements are concerned. It appears that various hybrid approaches to knowledge representation can satisfy the requirements in a greater degree than that of single representations. Another finding is that there is not a hybrid formalism that can satisfy the requirements of all of the modules of an ITS, but each one individually. So, a multi-paradigm representation environment could provide a solution to requirements satisfaction.

[1]  Stéphanie Jean-Daubias,et al.  The Ambre ILE: How to Use Case-Based Reasoning to Teach Methods , 2002, Intelligent Tutoring Systems.

[2]  Emilia Pecheanu,et al.  A Hybrid Aproach to Dynamic Course Generation on the WWW , 2000 .

[3]  Roger Nkambou.  Managing inference process in student modelling for intelligent tutoring systems , 1999, Proceedings 11th International Conference on Tools with Artificial Intelligence.

[4]  Paul S. Rosenbloom,et al.  Improving Accuracy by Combining Rule-Based and Case-Based Reasoning , 1996, Artif. Intell..

[5]  Julita Vassileva,et al.  Dynamic Courseware Generation on the WWW , 1998, Br. J. Educ. Technol..

[6]  KURT VANLEHN.  Bayesian student modeling, user interfaces and feedback : A sensitivity analysis , 2001 .

[7]  George D. Magoulas,et al.  Neuro-fuzzy synergism for planning the content in a web-based course , 2001, Informatica.

[8]  Stephen I. Gallant,et al.  Neural network learning and expert systems , 1993 .

[9]  Ioannis Hatzilygeroudis.  Using a hybrid rule-based approach in developing an intelligent tutoring system with knowledge acquisition and update capabilities , 2004, Expert Syst. Appl..

[10]  Ioannis Hatzilygeroudis,et al.  Neurules: improving the performance of symbolic rules , 1999, Proceedings 11th International Conference on Tools with Artificial Intelligence.

[11]  Ioannis Hatzilygeroudis,et al.  A Web-Based Intelligent Tutoring System Using Hybrid Rules as Its Representational Basis , 2002, Intelligent Tutoring Systems.