基于深度学习的舞弊识别模型缺陷自动检测方法

Deep Learning Based Automatic Detection of Fraud Recognition Model Defects

财务舞弊是当今商业领域中的一个严峻问题,为了预防和发现此类行为,此次研究利用深度学习中的极致梯度提升树算法(Extreme Gradient Boosting,XGBoost)搭建了财务舞弊识别模型。首先,研究利用国泰安数据库中的财务报表数据,建立了综合评价指标体系。其次,采用XGBoost算法搭建了财务舞弊识别模型。研究结果表明,利用XGBoost搭建的舞弊识别模型对于两所企业的财务数据舞弊情况识别精度分别高达9.11和9.58。此外,XGBoost舞弊模型能够准确识别对于企业舞弊现象影响较大的财务指标,并且其识别时间均在2 min以内。综上,此次研究所搭建的舞弊识别模型具有一定的应用价值,对于财务舞弊识别领域的发展具有重要的理论和实践意义。

Financial fraud is a serious problem in today's business field,in order to prevent and detect such behaviors,this study builds a financial fraud recognition model using Extreme Gradient Boosting(XGBoost)tree algorithm in deep learning.First,the study utilized the financial statement data in the Cathay Pacific database to build a comprehensive evaluation index system.Second,the financial fraud identification model was built using the XGBoost algorithm.The results of the study show that the fraud identification model constructed by XGBoost has a high accuracy of 9.11 and 9.58 respectively for the two enterprises'financial data fraud identification;moreover,the XGBoost fraud model can accurately identify the financial indicators that have a greater impact on the fraud phenomenon of the enterprise,and its identification time is within 2 min.In conclusion,the fraud identification model constructed in this study has certain application value and is of great theoretical and practical significance for the development of financial fraud identification.