摘要
金融风险影响实体经济发展,甚至可能引发金融危机,加强我国金融系统风险预警研究具有重要意义。研究根据金融系统性原则设计了一套金融系统风险指标体系,并基于Convolutional Neural Networks(CNN)卷积神经网络构建了一种金融系统风险预警模型。实验结果表明,与K-Nearest Neighbor(KNN)、Extreme Gradient Boosting(XGBoost)、Back Propagation Neural Networks(BPNN)神经网络模型相比,CNN卷积神经网络模型预警准确率高、性能优,是一种更有效的金融系统风险预警方法。
Financial risk affects the development of the hypostatic economy and may even trigger financial crisis.It is significant to strengthen the research on financial system risk forewarning in China.This paper designs a set of financial risk index system and constructs a financial system risk forewarning model based on CNN(Convolutional Neural Networks).The experimental results show that compared with neural network models such as KNN(K-Nearest Neighbor),XGBoost(Extreme Gradient Boosting)and BPNN(Back Propagation Neural Networks),the CNN modelis withhigher accuracy,better performance and more effectivefor financial system risk forewarning.
作者
罗俊霞
丁邦旭
查道懂
LUO Junxia;DING Bangxu;ZHA Daodong(School of Economics,Tongling University,Tongling 244061,China;School of Mathematics and Computer,Tongling University,Tongling 244061,China;Library,Tongling University,Tongling 244061,China)
出处
《宿州学院学报》
2020年第4期25-29,共5页
Journal of Suzhou University
基金
安徽省教育厅高校人文社会科学研究基地重点项目(SK2016A0925)。