Revisiting Applicable and Comprehensive Knowledge Tracing in Large-Scale Data

DKT2 is a deep learning-based Knowledge Tracing model that uses xLSTM architecture to predict student performance in Intelligent Tutoring Systems by modeling knowledge states from interaction sequences. The paper introduces a benchmark with published code and datasets, evaluating the model across three large-scale educational datasets against 18 baseline models.

Knowledge Tracing (KT) is a fundamental component of Intelligent Tutoring Systems (ITS), enabling the modeling of students' knowledge states to predict future performance. The introduction of Deep Knowledge Tracing (DKT), the first deep learning-based KT (DLKT) model, has brought significant advantages in terms of applicability and comprehensiveness. However, recent DLKT models, such as Attentive Knowledge Tracing (AKT), have often prioritized predictive performance at the expense of these benef

Study Type

Source

Framework Categories

Tool Types

Tags