AN ARCHITECTURE FOR KNOWLEDGE MANAGEMENT IN INTELLIGENT TUTORING SYSTEM Simone Riccucci Department of Computer Science, University of Bologna Via Mura Anteo Zamboni 7, 40121, Bologna, Italy Antonella Carbonaro Department of Computer Science, University of Bologna Via Mura Anteo Zamboni 7, 40121, Bologna, Italy Giorgio Casadei Department of Computer Science, University of Bologna Via Mura Anteo Zamboni 7, 40121, Bologna, Italy ABSTRACT This paper discusses a general framework for knowledge acquisition and management in an intelligent tutoring system. This system is based on “Learning by performance errors” theory stating that in a given domain knowledge there is a set of constraints that must be satisfied in order to provide the correct solution to the problem. This paper addresses the issues of representing complex and generic information that applies to multiple domains. The proposed solution provides guidelines for both the system knowledge acquisition and management based on the natural language processing platform GATE (General Architecture for Text Engineering), inductive logic programming and constraint based paradigm. KEYWORDS Intelligent tutoring system, Constraint based tutor, Information Extraction System, Knowledge acquisition, Inductive Logic Programming 1. INTRODUCTION ITSs are tools for assisted learning process. They leverage the knowledge of the domain and student models to implement adaptive tutoring that can approach the benefits of teacher to student interaction. The deployment of ITSs such as these ones (Anderson 1985, Koedinger 1997, Mitrovic 2001) have been demonstrating the efficacy of this technology. The scenario is composed by two actors, the teacher and the student, each of which interacts with a ITS module. The teacher provides the necessary knowledge to the ITS by means of an authoring tool and the student gets some questions from the ITS responding with an answer. ITS gives to a student the feedback on her solution showing the errors she made. This process is iterated until the student provides the right solution. While ITSs have been proven useful in the learning process, they suffer from the disadvantage of being difficult to setup and tune for the following reasons: • They require an expert in knowledge engineering in order to transfer the specific domain knowledge from teacher to computer; • Currently ITSs are still complex to setup and manage because usually they require to implement ad hoc mechanisms depending on the domain they are made for. The proposed architecture addresses the complexity of customizing the ITSs for specific knowledge domains by partitioning the knowledge codification and manipulation tasks in workable subtask some of IADIS International Conference on Cognition and Exploratory Learning in Digital Age (CELDA 2005) 473