.png){kind=icon}
.png){kind=icon}
Return
Deceptive opinion spam detection approaches: a literature survey
.png){kind=icon}
PRE
AI
S
D
A
DOI：10.1007/s10489-022-03427-1.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Nowadays, a large number of customers purchase products and services online. Customers can write their opinions in reviews to express the value and quality of purchased goods and services. These opinions are used to make purchase decisions by customers and design market strategies by sellers. The trustiness of online reviews highly affects a company's reputation and economic benefit. That is why online sellers hire people to write deceptive opinions to recommend their products or defame competitors' products. Detecting deceptive opinion spam has emerged as a challenging task. The article describes the publicly available review datasets and explores their deficiencies in finding deceptive opinion spam. This literature systematically unwinds prominent features and approaches that have been introduced to extricate the problem of deceptive opinion spam detection. Our primordial objective is to confer a solemnity analysis of recent papers on deceptive opinion spam detection that describes methodologies' features, strengths, and constraints. Finally, this work presents some crucial challenges and shortcomings of existing methodologies and introduces promising directions for further works. This paper presents a comprehensive review of recent research in deceptive opinion spam detection, which may prove helpful to researchers' best knowledge.
Keywords:
Deceptive opinion
Machine learning
Deep learning
Spammer
Spam detection
Journal
IF:
3.5
Papers:
7.5K
Citations:
1.7W
Organization
Cited Papers