A graph-based recommendation approach for highly interactive platforms

作者：

Highlights：

• Ranking recommendation approach for data streams recommendation.

• Heterogeneous information network modeling psychological factors that drive users’ decisions.

• Incremental learning at scale and with low latency.

• A graph ranking measure to infer online consumers’ perceived value.

• Framework for benchmarking recommender systems under streaming settings.

摘要

•Ranking recommendation approach for data streams recommendation.•Heterogeneous information network modeling psychological factors that drive users’ decisions.•Incremental learning at scale and with low latency.•A graph ranking measure to infer online consumers’ perceived value.•Framework for benchmarking recommender systems under streaming settings.

论文关键词：Recommender systems,Stream analysis,Consumption behavior modeling and prediction,Ranking recommendation approach,Knowledge graphs

论文评审过程：Received 30 April 2020, Revised 6 May 2021, Accepted 2 July 2021, Available online 10 July 2021, Version of Record 24 July 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115555