A Bipartite Graph Model and Mutually Reinforcing Analysis for Review Sites

A number of methods have been proposed for detecting spam reviews in order to obtain credible summaries. These methods, however, could not be uniformly applied to various forms of reviews and are not suitable for a product or service which has been evaluated by few reviewers. In this paper, we propo...

Full description

Saved in:
Bibliographic Details
Published inDatabase and Expert Systems Applications pp. 341 - 348
Main Authors Tawaramoto, Kazuki, Kawamoto, Junpei, Asano, Yasuhito, Yoshikawa, Masatoshi
Format Book Chapter
LanguageEnglish
Japanese
Published Berlin, Heidelberg Springer Berlin Heidelberg 2011
SeriesLecture Notes in Computer Science
Subjects
anomally detection
graph model
mutually reinforcing
Online AccessGet full text
ISBN9783642230875
3642230873
ISSN0302-9743
1611-3349
DOI10.1007/978-3-642-23088-2_25

Cover

More Information
Summary:A number of methods have been proposed for detecting spam reviews in order to obtain credible summaries. These methods, however, could not be uniformly applied to various forms of reviews and are not suitable for a product or service which has been evaluated by few reviewers. In this paper, we propose a bipartite graph model of review sites and a mutually reinforcing method of summarizing evaluations and detecting anomalous reviewers. Our model and method can be applied to reviews of various forms, and is suitable for a subject with few reviewers. We ascertain the effectiveness of our method using reviews of three forms on Yahoo! Movie web site.
ISBN:9783642230875
3642230873
ISSN:0302-9743
1611-3349
DOI:10.1007/978-3-642-23088-2_25