A novel framework for deep knowledge tracing via gating-controlled forgetting and learning mechanisms. Issue 1 (January 2023)

• Record Type: Journal Article
• Authors: Zhao, Weizhong; Xia, Jun; Jiang, Xingpeng; He, Tingting
• Is Part Of: Information processing & management. Volume 60:Issue 1(2023)
• Publication Date: 2023-01
• Subjects: Deep knowledge tracing -- Forgetting and learning mechanisms -- Intelligent education; Information storage and retrieval systems -- Periodicals; Information science -- Periodicals; Systèmes d'information -- Périodiques; Sciences de l'information -- Périodiques; Information science; Information storage and retrieval systems; Periodicals; 658.4038
• DOI: 10.1016/j.ipm.2022.103114
• Languages: English
• ISSNs: 0306-4573

Abstract

Abstract: In this paper, we propose a framework called Gating-controlled Forgetting and Learning mechanisms for Deep Knowledge Tracing (GFLDKT for short). In GFLDKT, two gating-controlled mechanisms are designed to model explicitly forgetting and learning behaviors in students' learning process. With the designed gating-controlled mechanisms, both the interaction records and students' different backgrounds are combined effectively for tracing the dynamic changes of students' mastery of knowledge concepts. Results from extensive experiments demonstrate that the proposed framework outperforms the state-of-the-art models on the KT task. In addition, the ablation study shows that designed forgetting and learning mechanisms contribute clearly to the performance improvement of GFLDKT. Highlights: A novel deep learning-based KT model is proposed, which explicitly utilizes the theories of learning and forgetting curves in updating knowledge states. Two gating-controlled mechanisms are designed for learning and forgetting curves, by which the interaction records and students' distinctive backgrounds are considered simultaneously. Results from extensive experiments demonstrate the effectiveness of the proposed model, which outperforms the SOTA KT models.