A global contextual enhanced structural-aware transformer for sequential recommendation

Zhu Zhang, Bo Yang, Xingming Chen, Qing Li

Abstract: Highlights•We propose to utilize global contextual information to enrich item representations.•We propose the Jaccard attention to capture structural dependencies between items.•Based on the above findings, we propose a sequential recommendation model, GC-ST.•Experimental results illustrate that GC-ST outperforms all baseline models.