A Convolutional Attention Network for Unifying General and Sequential Recommenders

作者：

Highlights：

• A unified framework to model the users’ purposes and personal preferences.

• Purpose specific Attention Unit to attend important items according to user purposes and preferences.

• Use PSAU in both long- and short-term interacted itemset to generate high-level hybrid user representation.

• Conduct extensive experiments on two real-world datasets.

摘要

•A unified framework to model the users’ purposes and personal preferences.•Purpose specific Attention Unit to attend important items according to user purposes and preferences.•Use PSAU in both long- and short-term interacted itemset to generate high-level hybrid user representation.•Conduct extensive experiments on two real-world datasets.

论文关键词：General recommenders,Sequential recommenders,User purpose modeling,Personal preference modeling,Attention mechanism,Convolutional neural network

论文评审过程：Received 27 April 2021, Revised 28 August 2021, Accepted 5 September 2021, Available online 7 October 2021, Version of Record 7 October 2021.

论文官网地址：https://doi.org/10.1016/j.ipm.2021.102755