Knowledge Tracing: A Review of Available Technologies

As a student modeling technique, knowledge tracing is widely used by various intelligent tutoring systems to infer and trace the individual’s state during learning process. In recent years, models were proposed get accurate easy-to-interpret results. To make sense of wide Knowledge (KT) landscape, this paper conducts systematic review provide detailed nuanced discussion relevant KT techniques from perspective assumptions, data, algorithms. The results show that most existing consider only fragment assumptions relate components within items student’s cognitive Almost all types take “quize data” as input, although it insufficient reflect clear picture students’ Dynamic Bayesian network, logistic regression deep are main algorithms models. Some open issues identified based on analytics reviewed works discussed potential future research directions.