Using a hybrid content-based and behaviour-based featuring approach in a parallel environment to detect fake reviews
作者：
Highlights：
• We study fake review detection using machine learning.
• A hybrid approach combining content and behaviour-based features is used.
• Resampling methods are utilised to overcome imbalanced data issues.
• Results show that our proposed approach is promising for fake review detection.
• A parallel cross validation method is introduced to speed up the validation process.
摘要
•We study fake review detection using machine learning.•A hybrid approach combining content and behaviour-based features is used.•Resampling methods are utilised to overcome imbalanced data issues.•Results show that our proposed approach is promising for fake review detection.•A parallel cross validation method is introduced to speed up the validation process.
论文关键词：Fake review detection,Featuring approach,Machine learning,Deep learning,Imbalanced data,Parallel processing
论文评审过程：Received 14 March 2020, Revised 28 January 2021, Accepted 26 March 2021, Available online 31 March 2021, Version of Record 23 May 2021.
论文官网地址：https://doi.org/10.1016/j.elerap.2021.101048
