A study of human flesh search with epidemic models

Abstract

Human Flesh Search (HFS) is a phenomenon of collaborative researching with a purpose of exposing personal information details of a target who has committed some misbehaviors. With the steady growth of Internet population and the communication convenience brought by forums, SNS and micro-blogs, HFS is becoming more and more powerful and able to fulfill tasks which are mission impossible by other conventional means.Most existing research work on HFS focuses on legal or privacy issues, while we aim at building a mathematical model to understand the evolution of the search process and hence to evaluate and quantify the capability of this massive collaboration intelligence. By borrowing ideas from epidemics, we build a mathematical model which views the process of HFS as an analogy to the process of infectious disease spread. Experimental results show that our model matches real HFS cases very well.The contribution of this paper is as follows. First, it is the first attempt to develop mathematical models to quantitatively describe the capability of HFS. Second, it is the first time to examine the relationship between target information entropy and its impact on netizen response, which cast new light on the success/failure prediction of future HFS campaigns.