Using the Wisdom of Crowds to Prevent Internet Frauds

Tao Zheng,Yong Jiang,Lei Zhang

2013-01-01

Abstract:With the rapid growth of the netizen population in China, more and more internet frauds are committed. Many people suffer from internet frauds by losing wealth or other valuable things. To prevent internet frauds, we first need to discover the methods in which internet frauds are conducted. In this paper, we investigate and categorize the internet frauds in China. So far, there are typically six kinds of internet frauds, including email fraud, website fraud, e-commerce fraud, virus fraud, password fraud, and message fraud. To prevent these internet frauds, many approaches, such as in- trusion detection and access control, have been proposed to help users. However, most of these methods are limited to detecting a small volume of internet frauds. To address this issue, we propose a methodology to use the wisdom of crowds, with the help of Semantic Web and Web 2.0 tech- nologies, to detect a large volume of internet frauds. The proposed framework is composed of eight modules: internet fraud report module, key element extraction module, linked data generation module, linked data repository, query inter- face, query interpretation module, SPARQL module, and answer generation module. Based on the framework, users are able to input the internet fraud reports in a controlled natural language. The internet fraud reports are converted into linked data automatically. Then, the users can query the linked data in a semantic fashion. A case study and a survey preliminarily indicate that the proposed method is able to help users share and identify the internet frauds effectively.