Researchers Develop Algorithm to Detect Fake Reviewers

Researchers at the University of Illinois at Chicago have designed an algorithm to protect against online review fraud that’s capable of damaging businesses’ reputation.

The GSRank algorithm can seek out organized groups of fake reviewers and automate the process of shutting them down, according to the recent study.

“Opinionated social media such as product reviews are now widely used by individuals and organizations for their decision making,” the researchers said. “However, due to the reason of profit or fame, people try to game the system by opinion spamming (e.g., writing fake reviews) to promote or demote some target products.”

Researchers believe a group of reviewers who work together to write fake reviews does more damage than individual spammers. Due to its size, it can control public opinion on the target product. The key to identifying such organized spam groups is their behavior, researchers said.

Their solution includes key fingerprints like the similarity of review/rating, the group size, and when the reviews were posted. For instance, comments posted too quickly in the life of an app tend to be fake. Also, members of an organized group all post similar ratings. The degree to which the group’s reviews deviate from “genuine” reviews is a hint that someone’s trying to game the system.

User reviews have become a particularly popular tool for mobile devices. Frauds have started to spread with social-style operations like Yelp and TripAdvisor. The spam review study was financed by Google.