A Data Mining Framework for the Acquisition of Procedural Knowledge in Intelligent Tutoring Systems

Domain knowledge acquisition is a very important and time-consuming process in intelligent tutoring system design. Furthermore, for many ill-defined domains, domain knowledge is hard to define explicitly. To address this problem, we proposed a data mining framework for extracting partial task models from recorded user interactions. This article describes improvements made to the framework and its application to extract a knowledge base that supports useful tutoring services in a tutoring system. A preliminary experiment with the tutoring system is presented.