Ensemble learning algorithm - research analysis on the management of financial fraud and violation in listed companies

Abstract

In recent years, despite the strict "zero tolerance" crackdown on the financial fraud and violation behavior of listed companies, the cases of financial fraud, revenue and profit overstatement, and suspected fraud have continued to be exposed. This study first established a financial fraud index system and used the XGBoost algorithm to construct a prediction model for financial fraud and violations of listed companies. The indicators were selected and input into the model. A dataset was obtained for experiments. The XGBoost algorithm was compared with two other algorithms. The receiver operator characteristic (ROC) curves showed that the XGBoost algorithm had the best prediction performance among the three algorithms. It was found that the precision of the XGBoost algorithm was 93.17%, the recall rate was 92.23%, the value was 0.9270, and the area under the curve was 0.90, indicating a better performance than the prediction models based on the Gradient Boosted Decision Tree (GBDT) algorithm and the Logistics algorithm. Considering the data of various evaluation indicators, it is found that the predictive effect of the financial fraud and violation prediction model built by the XGBoost algorithm is the best.