CoCNN: Co-occurrence CNN for recommendation

作者：

Highlights：

• To improve recommendations, we designed a CNN-based network to learn user preference.

• To explore the relationships among items, we designed a co-occurrence pattern.

• Our model is simultaneously modeled on user–item and item–item interactions.

摘要

•To improve recommendations, we designed a CNN-based network to learn user preference.•To explore the relationships among items, we designed a co-occurrence pattern.•Our model is simultaneously modeled on user–item and item–item interactions.

论文关键词：Collaborative Filtering,CNN,Deep learning,Neural network,Co-occurrence pattern,Multi-task

论文评审过程：Received 20 November 2021, Revised 4 January 2022, Accepted 19 January 2022, Available online 5 February 2022, Version of Record 7 February 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.116595