Social recommendation via deep neural network-based multi-task learning

作者：

Highlights：

• A social recommendation model (SoNeuMF) based on deep learning is proposed.

• SoNeuMF simultaneously models social and item interactions by neural networks.

• User representation in two tasks is shared to enhance the social recommendation list.

• Experiments show the improvement over baselines especially on cold-start users.

摘要

•A social recommendation model (SoNeuMF) based on deep learning is proposed.•SoNeuMF simultaneously models social and item interactions by neural networks.•User representation in two tasks is shared to enhance the social recommendation list.•Experiments show the improvement over baselines especially on cold-start users.

论文关键词：Social recommendation,Collaborative filtering,Neural networks,Matrix factorization,Multi-task learning

论文评审过程：Received 4 February 2021, Revised 30 March 2022, Accepted 2 June 2022, Available online 14 June 2022, Version of Record 21 June 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.117755