Advanced Knowledge Tracing: Incorporating Process Data and Curricula Information via an Attention-Based Framework for Accuracy and Interpretability
Year: 2024
Type: article
Abstract: Knowledge tracing aims to model and predict students’ knowledge states during learning activities.Traditional methods like Bayesian Knowledge Tracing (BKT) and logistic regression have limitationsin granularity and performance, while deep knowledge tracing (DKT) models often suffer from lackingtransparency. This paper proposes a Transformer-based framework that emphasizes both accuracyand interpretability. It captures the relationship between student behaviors and learning outcomes consideringthe associations between exam and exercise problems. We participated in the EDM Cup 2023Contest using the proposed framework and achieved first place on the task of predicting students’ performanceon end-of-unit test problems using clickstream data from previous assignments. Furthermore,the framework provides meaningful insights by analyzing user actions and visualizing attention weightmatrices. These insights enable targeted interventions and personalized support, enhancing online learningexperiences. We have uploaded our code, saved models, and predictions to an OSF repository:https://osf.io/mdpzc/.
Authors: Lu, Yikai, Tong, Lingbo, Cheng, Ying
Institution: University of Notre Dame
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: diamond
OpenAlex ID: W6949248418