Authors:
Matej Guid
;
Martin Možina
;
Ciril Bohak
;
Aleksander Sadikov
and
Ivan Bratko
Affiliation:
University of Ljubljana, Slovenia
Keyword(s):
Intelligent Tutoring Systems, Artificial Intelligence, Education, Problem Solving, Chess Endgames, Chess.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence and Decision Support Systems
;
Authoring Tools and Content Development
;
Computer-Supported Education
;
e-Learning
;
Enterprise Information Systems
;
Information Technologies Supporting Learning
;
Intelligent Tutoring Systems
Abstract:
We present the development of an intelligent tutoring system for chess endgames, and explain in detail the system’s architecture that is comprised of five essential components. The rule-based domain model contains a conceptualized domain theory, which serves as a bridge between the basic declarative domain theory and procedural knowledge for concrete problem solving. The search engine is used to generate new chess problems and to validate students’ solutions on the fly. The tutoring model represents pedagogical knowledge and follows the standard model-tracing approach. The student model contains the knowledge about the user in the form of a skill meter, aiming to show the level of a student’s understanding of particular skills. Finally, the user interface is where the interaction between a student and the tutor takes place.