Abstract: This paper tells the basic concept of the cheat and analyses the procreant reason of resulting in the cheat.The method of identification,analyzing and estimation of the cheat risks are discussed.It points out the existent problems of the tradition analyzing models in evaluating cheat risks.In the end,this paper puts forward to a research method of cheat risks based on distance and validates this method’s validity by a simulation experiment.The research work supplies a basis for further study of applying data mining for the enterprise cheating risks anglicizing.

Key words: cheating, distance, exceptional rules, outlier data, k-th nearest neighbor point