Memory-aware gated factorization machine for top-N recommendation

作者：

Highlights：

• We propose a memory-aware gated factorization machine (MAGFM) method.

• We propose user memory components to facilitate representation learning.

• We propose gated filtering units to filter out noises in auxiliary data.

• Experiments on real datasets confirm the superior performance of MAGFM.

摘要

•We propose a memory-aware gated factorization machine (MAGFM) method.•We propose user memory components to facilitate representation learning.•We propose gated filtering units to filter out noises in auxiliary data.•Experiments on real datasets confirm the superior performance of MAGFM.

论文关键词：Factorization machine,Memory-ware,Gated filtering,Neural network

论文评审过程：Received 2 December 2019, Revised 14 May 2020, Accepted 16 May 2020, Available online 21 May 2020, Version of Record 22 May 2020.

论文官网地址：https://doi.org/10.1016/j.knosys.2020.106048