ORFEL: super-fast detection of defamation and illegitimate promotion in online recommendation.
Gabriel P. Gimenes, Jose Fernando Rodrigues Junior, Robson L. F. Cordeiro · arXiv (Cornell University) · 2015
What if a successful company starts to receive a torrent of low-valued (one or two stars) recommendations in its mobile apps from multiple users within a short (say one month) period? Is it legitimate evidence that the apps have lost quality, or an intentional plan (via lockstep behavior) to steal market share through defamation? In case of a systematic attack to one's reputation, it might not be possible to manually discern between legitimate and fraudulent interaction in the immense universe of possibilities of user-product recommendation. Previous works have focused on this issue, but none of them has considered the context, modeling, and scale that we work with in this paper. We propose one novel method named Online-Recommendation Fraud ExcLuder (\ORFEL) to detect defamation and/or illegitimate promotion of online products using vertex-centric asynchronous parallel processing of bipartite (users-products) graphs. With an innovative algorithm, our results demonstrate efficacy -- detecting over $95\%$ of potential attacks; and efficiency -- at least two orders of magnitude faster than the state-of-the-art. Over our new methodology, we introduce three contributions: (1) a new algorithmic solution; (2) a scalable approach; and (3) a novel context and modeling of the problem, which now addresses both defamation and illegitimate promotion. Our work deals with relevant issues of the Web 2.0, potentially augmenting the credibility of online recommendation to prevent losses to both customers and vendors.