SLPKT: A Novel Simulated Learning Process Model for Knowledge Tracing
Jingxia Zeng, Mianfan Chen, Jianing Liu, Yuncheng Jiang · Proceedings/Proceedings of the ... International Conference on Software Engineering and Knowledge Engineering · 2023
Knowledge Tracing (KT) traces students' changing knowledge states and predicts future performance based on their past performance.However, most existing methods underestimate the impact of students' learning processes on performance prediction, and thus do not model the learning process well.To address this issue, we propose a novel Simulated Learning Process Model for Knowledge Tracing, which simulates the student's learning process by reviewing historical performance and enhancing the role of related knowledge in prediction before solving exercises.We first use a state acquisition module to obtain the knowledge state.Then we mine important historical information to assist in solving the target exercise.Finally, a knowledge enhancement module is used to improve the knowledge prediction of the target exercise.Extensive experiments on four real-world datasets demonstrate that our method is effective and outperforms the state-of-the-art models.