The document summarizes research on detecting fake reviewer groups on consumer review sites. The researchers developed a method to discover groups of reviewers working collaboratively to write fake reviews to promote or demote products. The method uses frequent itemset mining to find candidate groups and behavioral models to detect abnormal behaviors between group members. It was shown to outperform other detection methods and a labeled dataset of fake reviewer groups was created for evaluation. The researchers provide an example of a detected fake reviewer group coordinating positive reviews for the same products.