Knowledge-aware fine-grained attention networks with refined knowledge graph embedding for personalized recommendation

Wei Wang, Xiaoxuan Shen, Baolin Yi, Huanyu Zhang, Jianfang Liu, Chao Dai

Abstract: Highlights•An end-to-end recommendation model KFGAN is proposed.•KFGAN construct a fine-grained attention network to exploring user preferences.•We learns from contrastive learning approach to reduces redundant KG signal.•KFGAN mines the latent semantic information to improve representation learning.