Researchers from the University of Illinois at Chicago (UIC) and Google have developed software that can identify groups of fraudulent online reviews that attempt to steer the sentiment of products or businesses.
In the past, researchers have used Amazon's the Mechanical Turk to assess reviews, but UIC's Arjun Mukherjee and Bing Liu and Google's Natalie Glance decided to use paid experts, and then train software to differentiate between spammers and genuine reviewers. The researchers say the behavior of groups who collude to kill or hype products stuck out like sore thumbs. Spamming groups often file their reviews in quick bursts and the language they use is very similar, considering each reviewer is briefed by a contracting agency.
"Although labeling individual fake reviews and reviewers is very hard, to our surprise labeling fake reviewer groups is much easier," the researchers say. Deceptive reviews are a growing problem, but the research suggests that moderators of sites will be able to identify them more easily in the future.
From New Scientist
View Full Article
Abstracts Copyright © 2012 Information Inc., Bethesda, Maryland, USA
No entries found