Detecting spamming stores by analyzing their suspicious behaviors

The purpose of this paper is to detect the stores with spamming behaviors. We identify suspicious behaviors of these stores to detect spamming stores. These suspicious behaviors are from the two following observations. First, spamming stores may target quantity of sale and product reviews to influence consumers´ decisions. Second, they tend to deviate from the other stores in quantity of the sale and reviews. From those observations, we propose a novel scoring methods to find spamming stores, and they are applied on Aliexpress dataset. Our experiment results show that our proposed methods are effective in finding spamming stores.