Rating prediction based on combination of review mining and user preference analysis
作者：
Highlights：
• Proposing aspect-based rating prediction methods - ARPM and ARPM-Social.
• Integrating aspect detection, sentiment analysis and social behavior analysis.
• Analyzing different aspects of users and businesses mentioned in reviews.
• Measuring user similarity, business similarity, expertise value, and influence.
• Using social matrix factorization to predict ratings.
• Improving the accuracy of rating prediction.
摘要
•Proposing aspect-based rating prediction methods - ARPM and ARPM-Social.•Integrating aspect detection, sentiment analysis and social behavior analysis.•Analyzing different aspects of users and businesses mentioned in reviews.•Measuring user similarity, business similarity, expertise value, and influence.•Using social matrix factorization to predict ratings.•Improving the accuracy of rating prediction.
论文关键词：Data mining,Opinion mining,Natural language processing,Aspect detection,Rating prediction,Matrix factorization
论文评审过程：Received 19 July 2020, Revised 18 January 2021, Accepted 9 February 2021, Available online 14 February 2021, Version of Record 23 February 2021.
论文官网地址：https://doi.org/10.1016/j.is.2021.101742
