Online product reviews are subject to opinion spamming. Previously, a heterogeneous review graph ... more Online product reviews are subject to opinion spamming. Previously, a heterogeneous review graph was proposed to detect fake store reviewers. In this paper, we propose a new product review graph model to detect fake product reviewers in a storeless shopping environment. The model not only captures the relationships among reviews, reviewers and products, but also associates each node with abundant features, which can significantly improve the detection precision. An efficient algorithm is designed to speedup the convergence by eliminating a certain portion of reviewers during each iteration. Experiments verify the effectiveness and efficiency of the method.
Online product reviews are subject to opinion spamming. Previously, a heterogeneous review graph ... more Online product reviews are subject to opinion spamming. Previously, a heterogeneous review graph was proposed to detect fake store reviewers. In this paper, we propose a new product review graph model to detect fake product reviewers in a storeless shopping environment. The model not only captures the relationships among reviews, reviewers and products, but also associates each node with abundant features, which can significantly improve the detection precision. An efficient algorithm is designed to speedup the convergence by eliminating a certain portion of reviewers during each iteration. Experiments verify the effectiveness and efficiency of the method.
Uploads
Papers by Zhuo Wang