CPIN: Comprehensive present-interest network for CTR prediction

作者：

Highlights：

• Taking different kinds of interests into consider.

• Building the ancillary MLP to improve the training of the model.

• Experiments on public and industrial datasets are conducted.

摘要

•Taking different kinds of interests into consider.•Building the ancillary MLP to improve the training of the model.•Experiments on public and industrial datasets are conducted.

论文关键词：Recommender system,Deep neural network,Interest representation

论文评审过程：Received 11 August 2020, Revised 26 November 2020, Accepted 5 December 2020, Available online 13 December 2020, Version of Record 16 December 2020.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.114469