Research On The Identification Of Online Shopping Fraud Based On Case Based Reasoning

As a result of the rapid development of Internet, the whole global village steps into an era of Internet family, at the same time, a new economic model-- Network Shopping comes into being accordingly, through which people can not only buy all kinds of items they want, but also do commodity trading over regions and nations without boundaries, covering an unprecedented wide scope. This new commodity trading model is providing people with huge convenience, yet it is also causing new problems. In order to obtain high illegal interests, some criminals create endless fraudulent cases that are haunting customers. To control the situation, some developed countries such as the United States, the European Union, Japan and other countries, which have developed electronic commerce earlier, have adopted a series of preventive measures. As for our country, which is in the initial development stage of online shopping, there is still a long way to go. Therefore, it is particularly important and has strong realistic meaning to study online shopping fraud so as to promote the healthy development of electronic commerce in China and maintain social stability.This paper had firstly studied massive domestic and foreign documents and found that the issue had been widely studied and attracted scholars’ great attention from home and abroad.Most of the research studied around the security of personal information, online recommendation and other issues, with research methods mostly focusing on data analysis, evidence theory etc. However, this paper, for the first time, proposed the idea of constructing model to analyze and judge network shopping fraud, and applied the scientific research method of case based reasoning(CBR) to the research of this problem. Case Based Reasoning is a highly-valued reasoning technology which rises in the field of artificial intelligence in recent years and is used in a wide range.Taking the four steps(R4) of case-based reasoning as the main line, this paper mainly established the model based on CBR to analyze network shopping fraud cases, and further provided the preventive measures. This paper firstly used framework to present case bases, then calculated the similarity of the cases by the rule of “nearest neighbor”, and used AHP method to design prevention and make some adjustment when it involved the proportion. If it is identified as a fraud case, it will output the preventing strategies, if not, it runs the process of correction or case learning, eventually providing a powerful support to identify the fraud of network shopping.