• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

Computer Engineering & Science

Previous Articles     Next Articles

A fake review detection method
using emotional and belief function

YANG Feng-rui1,2,3,WU Xiao-hao1,2,WAN Cheng-feng1,2   

  1. (1.School of Communication and Information Engineering,
    Chongqing University of Posts and Telecommunications,Chongqing 400065;
    2.Research Center of New Telecommunication Technology Applications,
    Chongqing University of Posts and Telecommunications,Chongqing 400065;
    3 Chongqing Chongyou Information Technology (Group) Co.Ltd.,Chongqing 401121,China)
  • Received:2019-01-09 Revised:2019-03-18 Online:2019-09-25 Published:2019-09-25

Abstract:

Online comments are a basis for users to judge the quality of a product, and fake reviews seriously affect the purchase behavior of consumers. Existing fake review detection methods focus on the text and ignore the falseness of scores. Due to inaccurate and uncertain scoring, fake review detection does not work well. We propose a fake review detection method (EP-BFRD) using emotional polarity and trust function. The trust function is used to deal with the uncertainty and inaccuracy in a given reviewer's score, which are then combined with the similarity to other provided ratings to detect misleading, and to judge the consistency between the emotional polarity of the review text and  the score. A fake review is determined by both the processing results of the trust function and the consistency between the scores and text emotion. Experiments on a real database show that the proposed method is an effective solution to fake review detection problem.

 

 

 

Key words: fake review, emotional polarity, trust function, fake review detection