A deep bi-directional prediction model for live streaming recommendation

作者：

Highlights：

• We propose a model for live streaming recommendation to improve performance.

• We address the problem of how to simultaneously make predictions from both viewer and anchor sides.

• We develop multiple mechanisms to help the two sides learn from each other to enhance mutual learning.

• Experiment study demonstrates the effectiveness of the proposed model and mechanisms.

摘要

•We propose a model for live streaming recommendation to improve performance.•We address the problem of how to simultaneously make predictions from both viewer and anchor sides.•We develop multiple mechanisms to help the two sides learn from each other to enhance mutual learning.•Experiment study demonstrates the effectiveness of the proposed model and mechanisms.

论文关键词：Live streaming,Recommendation system,Deep learning

论文评审过程：Received 27 July 2020, Revised 26 October 2020, Accepted 26 November 2020, Available online 15 December 2020, Version of Record 15 December 2020.

论文官网地址：https://doi.org/10.1016/j.ipm.2020.102453