Research on detecting technique of financial statement fraud based on Fuzzy Genetic Algorithms BPN

In recent years, the phenomenon of financial statement fraud what happened in listed companies becomes a global focus, which seriously affects economic development. To avoid the huge harm brought by financial statement fraud, to reduce the heavy work of the auditors, to increase the efficiency and precision of auditing and detecting,it is extremely urgent to research detecting technique, which is efficient, convenient and practical. This paper studies on the financial statements of fraud companies and paired companies. It explores the two aspects characteristic signals both from finance and corporate governance, and establishes a set of more perfect feature indicators for detecting the fraud. Then it designs the Fuzzy Genetic Algorithms BPN (FGABPN) model to detecting fraudulent financial reporting for the first time. It is found by test that discrimination of the model is higher.