Discovering Student Models with a Clustering Algorithm Using Problem Content

One of the key factors that affects automated tutoring systems in making instructional decisions is the quality of the student model built in the system. A student model is a model that can solve problems in various ways as human students. A good student model that matches with student behavior patterns often provides useful information on learning task difficulty and transfer of learning between related problems, and thus often yields better instruction on intelligent tutoring systems. However, traditional ways of constructing such models are often time consuming, and may still miss distinctions in content and learning that have important instructional implications. Automated methods can be used to find better student models, but usually require some engineering effort, and can be hard to interpret. In this paper, we propose an automated approach that finds student models using a clustering algorithm based on automaticallygenerated problem content features. We demonstrate the proposed approach using an algebra dataset. Experimental results show that the discovered model is as good as one of the best existing models, which is a model found by a previous automated approach, but without the knowledge engineering effort.