id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-144033-pmchx05r	Shin, Dongmin	SAINT+: Integrating Temporal Features for EdNet Correctness Prediction	2020-10-19		.txt	text/plain	3491	209	52	We propose SAINT+, a successor of SAINT which is a Transformer based knowledge tracing model that separately processes exercise information and student response information. SAINT is a Transformer [25] based knowledge tracing model that separately processes information of exercise and student response. Also, the experimental results show that incorporating the temporal features into the decoder input achieves the best AUC compared to incorporating them into the encoder input, and both the encoder and decoder input, verifying the hypothesis that separately processing exercise information and student response information is appropriate for knowledge tracing. SAINT [2] is the first Transformer based knowledge tracing model which leverages encoder-decoder architecture composed of stacked self-attention layers. In this paper, we proposed SAINT+, a Transformer based knowledge tracing model that processes exercise information and student response information separately, and integrates two temporal feature embeddings into the response embeddings: elapsed time and lag time.	./cache/cord-144033-pmchx05r.txt	./txt/cord-144033-pmchx05r.txt