Empirical Study on Financial Fraud Identification Model Based on Data Mining

The application of data mining technology promotes the establishment of financial fraud identification models to improve the accuracy of financial fraud detection. On this basis, a set of “financial fraud identification” model based on multi-dimensional indicators is constructed, and an empirical study is carried out with the sample of listed companies from 2018 to 2022. The empirical analysis shows that in different industries and different enterprises, the detection rate of fraud cases is above 85%, while the detection rate of non-fraud cases is above 90%. In terms of the main influencing factors, the variables of gearing ratio, accounts receivable turnover ratio, equity concentration, and earnings per share predicted financial fraud and tested the model.