Abstract
Recently, we have seen a rapid rise in usage of online educational platforms. The personalized education became crucially important in future learning environments. Knowledge tracing (KT) refers to the detection of students' knowledge states and predict future performance given their past outcomes for providing adaptive solution to Intelligent Tutoring Systems (ITS). Bayesian Knowledge Tracing (BKT) is a model to capture mastery level of each skill independently with psychologically meaningful parameters and widely used in successful tutoring systems. However it has lower efficiency in student performance prediction and is unable to detect learning transfer across skills. While recent KT models based on deep neural networks shows impressive student performance prediction but it came with a price. Ten of thousands of parameters in neural networks are unable to provide psychologically meaningful interpretation that reflect to cognitive theory. In this paper, we proposed an efficient knowledge tracing model (BKT-LSTM) that is able to provide meaningful interpretation (where individual \textit{skill mastery} of a student is learned by BKT and \textit{learning transfer} (across skills) is detected by k-means clustering) and then \textit{problem difficulty} is taken into account in student performance prediction by leveraging predictive power of LSTM. BKT-LSTM outperforms state-of-the-art student models in term of predictive power whilst considering skill mastery with psychologically meaningful interpretation of BKT.
Abstract (translated)
URL
https://arxiv.org/abs/2012.12218
