DKT2: Revisiting Applicable and Comprehensive Knowledge Tracing in Large-Scale Data
This work addresses a practical problem for educational technology developers by improving knowledge tracing models for real-world intelligent tutoring systems, though it is incremental as it builds on existing deep learning and IRT methods.
The paper tackles the challenge of maintaining applicability and comprehensiveness in deep knowledge tracing models, which often sacrifice these for predictive performance, by proposing DKT2, a model that integrates xLSTM with Rasch model and IRT for interpretability, and it outperforms 18 baseline models across three large-scale datasets.
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 benefits. While deep sequential models like DKT have shown potential, they face challenges related to parallel computing, storage decision modification, and limited storage capacity. To address these limitations, we propose DKT2, a novel KT model that leverages the recently developed xLSTM architecture. DKT2 enhances applicable input representation using the Rasch model and incorporates Item Response Theory (IRT) for output interpretability, allowing for the decomposition of learned knowledge into familiar and unfamiliar knowledge. By integrating this knowledge with predicted questions, DKT2 generates comprehensive knowledge states. Extensive experiments conducted across three large-scale datasets demonstrate that DKT2 consistently outperforms 18 baseline models in various prediction tasks, underscoring its potential for real-world educational applications. This work bridges the gap between theoretical advancements and practical implementation in KT. Our code and datasets are fully available at https://github.com/zyy-2001/DKT2.