Predicting hotel review helpfulness: The impact of review visibility, and interaction between hotel stars and review ratings

作者:

Highlights:

• The interaction effect of Hotel Star Class (HSC) and Review Rating (RR) is confirmed to affect review helpfulness.

• The concept of review helpfulness is posed to include three variables. All had a varying and significant effect on review helpfulness.

• Model tree (M5P) outperformed linear regression and support vector regression.

• Certain predictors (such as badges earned) are back-traced to the time when a review was written in order to improve prediction accuracy of existing studies.

• Prediction accuracy improves after review visibility, interaction effect and improved data collection are taken into account.

摘要

•The interaction effect of Hotel Star Class (HSC) and Review Rating (RR) is confirmed to affect review helpfulness.•The concept of review helpfulness is posed to include three variables. All had a varying and significant effect on review helpfulness.•Model tree (M5P) outperformed linear regression and support vector regression.•Certain predictors (such as badges earned) are back-traced to the time when a review was written in order to improve prediction accuracy of existing studies.•Prediction accuracy improves after review visibility, interaction effect and improved data collection are taken into account.

论文关键词:Online hotel reviews,Review helpfulness,Sentiment analysis,eWOM,Review visibility

论文评审过程:Received 10 November 2015, Revised 30 March 2016, Accepted 8 June 2016, Available online 22 June 2016, Version of Record 22 June 2016.

论文官网地址:https://doi.org/10.1016/j.ijinfomgt.2016.06.003