Improving Question Embeddings with Cognitive Representation Optimization for Knowledge Tracing

TL;DR

<3-5 sentence high-level summary> CRO-KT tackles the gap in knowledge tracing by explicitly optimizing cognitive representations to account for distractors such as slips and guesses. It introduces two synergistic modules: a coordination module based on dynamic programming to align records with cognitive patterns, and a collaboration module using co-optimization to enforce synergy among related questions, complemented by bipartite-graph learned relations. The approach fuses optimized records with relation embeddings and demonstrates state-of-the-art performance on three public KT datasets, highlighting its potential for more accurate modeling of student cognition. The work also provides ablation and parameter-sensitivity analyses, underscoring the contribution of each module and the robustness of the method.

Abstract

Designed to track changes in students' knowledge status and predict their future answers based on students' historical answer records. Current research on KT modeling focuses on predicting future student performance based on existing, unupdated records of student learning interactions. However, these methods ignore distractions in the response process (such as slipping and guessing) and ignore that static cognitive representations are temporary and limited. Most of them assume that there are no distractions during the answering process, and that the recorded representation fully represents the student's understanding and proficiency in knowledge. This can lead to many dissonant and uncoordinated issues in the original record. Therefore, we propose a knowledge-tracking cognitive representation optimization (CRO-KT) model that uses dynamic programming algorithms to optimize the structure of cognitive representation. This ensures that the structure matches the student's cognitive patterns in terms of practice difficulty. In addition, we use a synergistic optimization algorithm to optimize the cognitive representation of sub-target exercises based on the overall picture of exercise responses by considering all exercises with synergistic relationships as one goal. At the same time, the CRO-KT model integrates the relationship embedding learned in the dichotomous graph with the optimized record representation in a weighted manner, which enhances students' cognitive expression ability. Finally, experiments were conducted on three public datasets to verify the effectiveness of the proposed cognitive representation optimization model.

Figures (13)

• Figure 1: Taking the three parts of the figure from left to right as an example. Figure (a) shows the process of student working on the problem. Figure (b) shows the forecasting process of student performance. Figure (c) shows the process of predicting student performance based on the relationship between different questions.
• Figure 2: Schematic framework diagram of the CRO-KT model.
• Figure 3: Algorithm flowchart.
• Figure 4: Algorithm flowchart.
• Figure 5: Bar graphs show the performance of each model.