A graph attention network with contrastive learning for temporal review-based recommendations

Open Webpage

Peilin Yang, Yingyuan Xiao, Wenguang Zheng, Yanli Liu, Ching-Hsien Hsu

Published: 01 Jan 2024, Last Modified: 05 Nov 2025Appl. Soft Comput. 2024EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•A novel graph neural network can integrate time-series review and ratings.•The graph attention network and gating mechanism of users and items can construct more accurate feature representation.•A new self-supervised contrastive learning strategy can strengthen node information.