Research On The Application Of Improved BP Neural Network In The Accounting Malpractices Detecting

Normal information disclosure is the fundamental guarantee to maintenance the open, fair and just of stock market. But accounting malpractices is the hinder of the stock market healthy development in China.With the increasing perfection of securities market in China and the frequent occurrence of accounting malpractices, the detecting is being paid more and more attention to. In this paper, the research status of accounting malpractices detecting in China and abroad is firstly analyzed. Then basing on 116 listed companies, a system of accounting malpractices detecting indicators is constructed by analyzing the cases and methods of malpractices. In the end, a model of accounting malpractices detecting is constructed basing on Principal Component Analysis and BP neural network improved by Genetic Algorithm. An Empirical Analysis has been conducted. According to the study, the model has a better accuracy to the accounting malpractices detecting, and is an operable detecting method.