Heterogeneous collaborative filtering contrastive learning for social recommendation

Chaojun Meng, Changfan Pan, Hongji Shu, Qing Wang, Hanghui Guo, Jia Zhu

Abstract: Highlights•A representation learning model based on heterogeneous collaborative signal mining.•The proposed model mines group feature among users or items.•A transfer network is proposed to personalized transfer group feature.•Contrastive learning captures the consistency of heterogeneous collaborative signals.