A support vector approach to detecting manipulated reviews

Kelvin Kizito King, Francis Kofi Andoh-Baidoo

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Despite the abundance and successes of studies on manipulated reviews, there have been notable limitations acknowledged by academia and industry alike. Some of which include the lack of rich datasets for feature extraction and the inability to marry verbal and nonverbal features for model development. Furthermore, prior studies have always relied on existing literature in order to select relevant features for analysis. The limitations mentioned have rendered recent attempts at combating review manipulations quite unreliable and they lack the efficacy to target complex manipulated reviews, for example fake reviewers whose profiles are set to private. We attempt to bridge this gap by proposing a hybrid framework that incorporates econometrics and machine learning in feature extraction, engineering and review detection. Furthermore we utilize these methods in tandem and derive new sets of features for future analysis. This emergent study has important implications for research and practice.

Original languageEnglish (US)
Title of host publication26th Americas Conference on Information Systems, AMCIS 2020
PublisherAssociation for Information Systems
ISBN (Electronic)9781733632546
StatePublished - 2020
Externally publishedYes
Event26th Americas Conference on Information Systems, AMCIS 2020 - Salt Lake City, Virtual, United States
Duration: Aug 10 2020Aug 14 2020

Publication series

Name26th Americas Conference on Information Systems, AMCIS 2020

Conference

Conference26th Americas Conference on Information Systems, AMCIS 2020
Country/TerritoryUnited States
CitySalt Lake City, Virtual
Period8/10/208/14/20

Keywords

  • Fake
  • Manipulated reviews
  • Poisson distribution
  • Support vector machine

ASJC Scopus subject areas

  • Computer Science Applications
  • Information Systems
  • Computer Networks and Communications
  • Library and Information Sciences

Fingerprint

Dive into the research topics of 'A support vector approach to detecting manipulated reviews'. Together they form a unique fingerprint.

Cite this