SAINT+: Integrating Temporal Features for EdNet Correctness Prediction

TL;DR

SAINT+ is a Transformer-based knowledge tracing model that enhances prediction accuracy by incorporating temporal features like elapsed and lag time, achieving state-of-the-art results on the EdNet dataset.

Contribution

SAINT+ introduces temporal feature embeddings into a Transformer-based knowledge tracing model, improving predictive performance over previous models.

Findings

SAINT+ outperforms SAINT with a 1.25% AUC improvement.

Incorporating temporal features enhances model accuracy.

SAINT+ achieves state-of-the-art results on EdNet dataset.

Abstract

We propose SAINT+, a successor of SAINT which is a Transformer based knowledge tracing model that separately processes exercise information and student response information. Following the architecture of SAINT, SAINT+ has an encoder-decoder structure where the encoder applies self-attention layers to a stream of exercise embeddings, and the decoder alternately applies self-attention layers and encoder-decoder attention layers to streams of response embeddings and encoder output. Moreover, SAINT+ incorporates two temporal feature embeddings into the response embeddings: elapsed time, the time taken for a student to answer, and lag time, the time interval between adjacent learning activities. We empirically evaluate the effectiveness of SAINT+ on EdNet, the largest publicly available benchmark dataset in the education domain. Experimental results show that SAINT+ achieves state-of-the-art performance in knowledge tracing with an improvement of 1.25% in area under receiver operating characteristic curve compared to SAINT, the current state-of-the-art model in EdNet dataset.