A user similarity-based Top-N recommendation approach for mobile in-application advertising

作者：

Highlights：

• An effective top-N model is designed for large-scale mobile in-app advertising.

• User feature is divided into stable context feature and dynamic implicit feedback.

• Similarity aggregation is performed to obtain the refined online user similarity.

• An elastic factor is proposed to improve the accuracy of top-N recommendation.

摘要

•An effective top-N model is designed for large-scale mobile in-app advertising.•User feature is divided into stable context feature and dynamic implicit feedback.•Similarity aggregation is performed to obtain the refined online user similarity.•An elastic factor is proposed to improve the accuracy of top-N recommendation.

论文关键词：Neighborhood-based recommendation,User similarity,Top-N preference,Mobile in-application advertising

论文评审过程：Received 24 April 2017, Revised 7 February 2018, Accepted 8 February 2018, Available online 8 February 2018, Version of Record 29 July 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.02.012