Research On Recognition Model Of Corporate Financial Fraud Based On Data Mining

China's stock market has gone through 40 years since its establishment in 1989.During this period,the number and overall scale of listed companies have grown rapidly,which has injected strong impetus into the economic development of our country.However,due to the relatively short development time of China's securities market and the imperfect regulatory policies,the financial fraud of listed companies is still high,which damages the vital interests of investors and affects the healthy and stable development of China's securities market.Therefore,how to accurately identify and effectively warn the financial fraud of listed companies has become a common concern of regulators,institutional investors and individual investors.In recent years,with the development of information technology,data mining technology plays an increasingly prominent role in financial risk prevention.This paper focuses on the comparative analysis of data mining models and the effect test in Chinese market.The accuracy of data mining technology discrimination is studied by using the financial fraud information database of listed companies and the financial information database of listed companies in Chinese market from 2000 to 2016,and the introduction of data mining model into Chinese market is discussed.At the same time,this paper tries to build a comprehensive recognition mechanism on the basis of several known models.Based on the screening results of decision tree,neural network and other models,the possibility of financial fraud of listed companies is evaluated comprehensively through the optimized weight settings,so as to improve the accuracy of financial fraud identification and reduce the omission rate.To give risk tips to listed companies that may have financial fraud problems,so as to provide investment basis for market investors and decision-making reference for securities regulatory authorities.