BP Neural Network Algorithm in Accounting Audit Recognition

In the face of the rapid development of the capital market, we have to reflect on how to effectively identify audit risks, and how the CSRC and accounting firms should perform their functions of supervision and review. The purpose of this paper is to study the application of BP neural network algorithm in accounting audit recognition. The related theory of audit risk is introduced, and the theoretical principles of BP neural network and neighborhood rough set method used in this paper are explained. Construct sample data and indicator system, screen 35 financial indicators and non-financial indicators to construct indicator system, and select qualified A-share listed companies from 2010 to 2019 for data collection as the data input of the audit risk identification model. The BP neural network is used to build an audit risk identification model, and then the training accuracy of the model is calculated. Finally, the reduced index of the neighborhood rough set is used to compare and analyze the identification ability of the model. The experimental results show that the comprehensive accuracy of the BP neural network recognition model reduced by the neighborhood rough set index is 94%.