A Reputation Ranking Method based on Rating Patterns and Rating Deviation

Jian Zhou, Yu Feng Liu, Hong Liang Sun

Research output: Chapter in Book/Conference proceedingConference contributionpeer-review

2 Citations (Scopus)

Abstract

In the e-commerce platform, user purchase behavior often depends on personal experience and the objective ratings of others. Various businesses employ a large number of spammers to obtain illegal benefits by distorting the ranking of goods, which seriously affects the market order. How to design a high- speed and effective ranking method to remove these spammers is necessary and significant. In this paper, a novel reputation ranking method is proposed based on users' Rating Patterns and Rating Deviation (RPRD) because users' rating preference and historical behavior differ significantly with spammers. We compare RPRD method with three classical methods Deviation- based Ranking (DR), Iterative Group-based Ranking (IGR) and Iterative Balance Ranking (IBR) on three real datasets. Experimental results show that the RPRD method can effectively resist spammers attack and identification, especially in detecting random spammers. On the other hand, this method always has high accuracy and robustness even if the network is relatively sparse. It can also be applied in large and sparse bipartite rating networks in a short time.

Original languageEnglish
Title of host publication2022 5th International Conference on Data Science and Information Technology, DSIT 2022 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781665498685
DOIs
Publication statusPublished - 2022
Event5th International Conference on Data Science and Information Technology, DSIT 2022 - Shanghai, China
Duration: 22 Jul 202224 Jul 2022

Publication series

Name2022 5th International Conference on Data Science and Information Technology, DSIT 2022 - Proceedings

Conference

Conference5th International Conference on Data Science and Information Technology, DSIT 2022
Country/TerritoryChina
CityShanghai
Period22/07/2224/07/22

Keywords

  • Fraud Detection
  • Rating Patterns
  • Spammers Attack

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Computer Vision and Pattern Recognition
  • Information Systems
  • Decision Sciences (miscellaneous)
  • Information Systems and Management
  • Statistics, Probability and Uncertainty

Fingerprint

Dive into the research topics of 'A Reputation Ranking Method based on Rating Patterns and Rating Deviation'. Together they form a unique fingerprint.

Cite this