From textual reviews to Individual Reputation Rankings: Leaving ratings aside solving MPC task

作者：

Highlights：

• Multiway Polarity Classification task solved with ratings entails problems.

• It is explored estimating users’ preferences from reviews leaving ratings aside.

• Performance evaluated on real datasets from IMDb site and Android based Apps.

• It is more accurate to map reviews to preferences than to numerical ratings.

• It is independent of the sentiment algorithm executed.

摘要

•Multiway Polarity Classification task solved with ratings entails problems.•It is explored estimating users’ preferences from reviews leaving ratings aside.•Performance evaluated on real datasets from IMDb site and Android based Apps.•It is more accurate to map reviews to preferences than to numerical ratings.•It is independent of the sentiment algorithm executed.

论文关键词：On-line review systems,Multiway polarity classification,Rating systems,Reputation evaluation

论文评审过程：Received 3 April 2018, Revised 9 July 2018, Accepted 16 July 2018, Available online 18 July 2018, Version of Record 23 July 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.07.037