Measuring service quality from unstructured data: A topic modeling application on airline passengers’ online reviews

作者：

Highlights：

• Measuring quality is challenging due to preference heterogeneity among experts.

• Online reviews offer a solution through quality features extracted from review text.

• Using structural topic modeling we couple review text with numerical ratings.

• An experimental application to airline passengers’ reviews is demonstrated.

• Quality features better predict variations in passenger preferences and competition.

摘要

•Measuring quality is challenging due to preference heterogeneity among experts.•Online reviews offer a solution through quality features extracted from review text.•Using structural topic modeling we couple review text with numerical ratings.•An experimental application to airline passengers’ reviews is demonstrated.•Quality features better predict variations in passenger preferences and competition.

论文关键词：Electronic WOM,Unstructured data,Service quality,Correspondence analysis,Structural topic model

论文评审过程：Received 26 March 2018, Revised 20 August 2018, Accepted 16 September 2018, Available online 17 September 2018, Version of Record 24 September 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.09.037