Modeling Individualization in a Bayesian Networks Implementation of Knowledge Tracing

The field of intelligent tutoring systems has been using the well known knowledge tracing model, popularized by Corbett and Anderson (1995) to track individual users’ knowledge for 15 years. Surprisingly, models currently in use do not allow for individual learning rates nor individualized estimates of student background knowledge. Corbett and Anderson, in their original articles, were interested in trying to add individualization to their model which they accomplished but with mixed results. Since their original work, the field has not made significant progress towards individualization of knowledge tracing models in fitting data. In this work, we introduce an elegant way of formulating the individualization problem entirely within a Bayesian networks framework that learns individualized as well as skill specific parameters in a single step. With this new model we are able to show a reliable improvement in knowledge estimation and data prediction. The novelty of the model is the ability to learn model parameters with the assumption that different students have different initial background knowledge probabilities. We evaluate the individualized model and standard knowledge tracing model with empirical results of predicting real student data. The implication of this work is the ability to enhance existing intelligent tutoring systems to more accurately estimate when a student has reached mastery of a skill. Adaptation of instruction based on individualized knowledge and learning speed is discussed as well as the open research questions facing those that wish to exploit student and skill information in their user models.