Detecting positive and negative deceptive opinions using PU-learning
作者:
Highlights:
• Detection of negative deceptive opinion spam.
• Improved PU-learning approach.
• Compares the performance of the proposed approach and the original PU-learning method.
• The role of opinions’ polarity in the detection of deception.
• Reports experimental results on a set of negative deceptive opinions.
摘要
•Detection of negative deceptive opinion spam.•Improved PU-learning approach.•Compares the performance of the proposed approach and the original PU-learning method.•The role of opinions’ polarity in the detection of deception.•Reports experimental results on a set of negative deceptive opinions.
论文关键词:Opinion mining,Opinion spam,Deceptive opinions,PU-learning
论文评审过程:Received 24 September 2013, Revised 28 October 2014, Accepted 9 November 2014, Available online 30 November 2014, Version of Record 6 June 2015.
论文官网地址:https://doi.org/10.1016/j.ipm.2014.11.001