Self-Attention Based Time-Rating-Aware Context Recommender System

Table 3

Comparison of recommended performance of the SATRAC model and other baselines. The best results are in bold. The underlined numbers are suboptimal results.

Dataset	Metric	BPR	GRURec	Caser	MARank	SASRec	HGN	TiSASRec	SATRAC
Movies-1M	Hit@10	0.5922	0.5781	0.7487	0.7622	0.8493	0.8123	0.8529	0.8591
	NDCG@10	0.3401	0.3781	0.5123	0.5169	0.5712	0.5448	0.5767	0.6107
Amazon-movies	Hit@10	0.5517	0.5368	0.4938	0.6546	0.7563	0.7429	0.7647	0.7856
	NDCG@10	0.3464	0.3317	0.3201	0.4562	0.5238	0.5167	0.5334	0.5519
Amazon-CDs	Hit@10	0.5627	0.5364	0.6865	0.6981	0.7015	0.7045	0.7001	0.7141
	NDCG@10	0.3626	0.3267	0.4585	0.4589	0.4635	0.4662	0.4604	0.4876