Quaternion-based knowledge graph neural network for social recommendation

作者：

Highlights：

• We propose a novel quaternion-based knowledge graph neural network for social recommendation (QSoR).

• We design a double-end attention mechanism to obtain user preferences and user social relationships in the quaternion space.

• We propose a quaternion-based user explicit and implicit social relationships integration algorithm.

摘要

•We propose a novel quaternion-based knowledge graph neural network for social recommendation (QSoR).•We design a double-end attention mechanism to obtain user preferences and user social relationships in the quaternion space.•We propose a quaternion-based user explicit and implicit social relationships integration algorithm.

论文关键词：Quaternion-based recommenders,Attentional mechanism,Graph neural networks,Social recommendation

论文评审过程：Received 11 April 2022, Revised 21 September 2022, Accepted 23 September 2022, Available online 29 September 2022, Version of Record 17 October 2022.

论文官网地址：https://doi.org/10.1016/j.knosys.2022.109940