Personalized Knowledge Tracing through Student Representation Reconstruction and Class Imbalance Mitigation

Abstract

Knowledge tracing is a technique that predicts students' future performance by analyzing their learning process through historical interactions with intelligent educational platforms, enabling a precise evaluation of their knowledge mastery. Recent studies have achieved significant progress by leveraging powerful deep neural networks. These models construct complex input representations using questions, skills, and other auxiliary information but overlook individual student characteristics, which limits the capability for personalized assessment. Additionally, the available datasets in the field exhibit class imbalance issues. The models that simply predict all responses as correct without substantial effort can yield impressive accuracy. In this paper, we propose PKT, a novel approach for personalized knowledge tracing. PKT reconstructs representations from sequences of interactions with a tutoring platform to capture latent information about the students. Moreover, PKT incorporates focal loss to improve prioritize minority classes, thereby achieving more balanced predictions. Extensive experimental results on four publicly available educational datasets demonstrate the advanced predictive performance of PKT in comparison with 16 state-of-the-art models. To ensure the reproducibility of our research, the code is publicly available at https://anonymous.4open.science/r/PKT.

Figures (6)

• Figure 1: Illustration of learning scenarios within knowledge tracing tasks.
• Figure 2: Class imbalance ratios in publicly available datasets within knowledge tracing.
• Figure 3: An overview of the proposed PKT model. $h$, $u_s$, and $u_{c,j}$ represent the hidden state, student representation, and capsule representation, respectively. $p$, $r$, and $sim$ correspond to performance prediction, reconstructed student representation, and similarity. $\sigma$, $+$, and $\times$ denote the sigmoid, concatenation and multiplication functions, respectively.
• Figure 4: Performance comparison between setting maxlen to the average student sequence length and a fixed value of 200.
• Figure 5: Visualization of the reconstructed and original student representations using t-SNE, showing clustering of similar representations and dispersion of outliers.