Application of Backpropagation Neural Network Based on Levenberg-Marquardt Algorithm in Detection of Fraudulent Financial Statements

摘要

Neural network, a sort of nonlinear model, has been successfully applied in many fields. Considering the features of fraudulent financial statements, six financial indexes are selected as recognition variables according to an index similar to signal-to-noise ratio. In addition, a BP neural network model based on Levenberg-Marquardt algorithm is used to detect fraudulent financial statements of public companies. The experimental result shows that the model is of high accuracy and effective in the detection of fraudulent financial statements.