Research Of Commercial Fraud Base On Data Mining

Commercial fraud as an important research topic in business ethics, in recent years by the enterprise managers more attention, because with the development of socialist market economy, the growing phenomenon of commercial fraud, has become increasingly complex, harm to enterprises and society is becoming more and more serious, so we need to put commercial fraud on this important topic of business ethics research increase. But the traditional research methods such as the triangle fraud theory, GONE theory, fraud risk factors theory are all focused on the qualitative analysis, lack of quantitative and accurate analysis. With the development of information technology, data mining technology has become more mature, for quantitative analysis of commercial fraud provides useful tools, analysis the conclusion can be very good for the enterprise managers a reasonable decision, avoid commercial fraud, to achieve healthy and sustainable development to help enterprises.he feasibility analysis of application of data mining in APRIRO correlation analysis and ID3 decision tree of commercial fraud on the basis of this, commercial fraud data mining and modeling process in Matlab with M script, and in Wenzhou city Longwan District People's Procuratorate in handling commercial fraud cases are simulated by mining the historical data commercial fraud, combined with business ethics principle, found:(1) Data mining can quickly realize the commercial fraud mining analysis,points out that it has some characteristics of enterprises tend to be relatively large commercial fraud,the conclusion is true and reliable,can guide our practical work.(2) VAT invoice crime is often accompanied by commercial fraud to defraud the import and export tax rebates,tax deductible invoice crime.(3) Corporate culture and corporate responsibility of the person in charge of the enterprise to avoid the impact of the most important business fraud.