基于BP-Adaboost的公司财务预警系统设计  

Design of Corporate Financial Early Warning System Based on BP-Adaboost

作　　者：张瑞华 吴吉 ZHANG Ruihua;WU Ji(School of Physics and Information Engineering,Jianghan University,Wuhan 430000,China)

机构地区：[1]江汉大学人工智能学院,武汉430000

出　　处：《移动信息》2025年第1期346-348,共3页Mobile Information

基　　金：“通信原理”虚拟仿真实验教学应用研究(2019012)。

摘　　要：在快速变化的市场环境中,上市公司的财务稳定性对国家经济和金融市场的健康发展具有重要影响。因此,构建有效的财务危机预警系统显得非常重要。文中设计了一种基于BP-Adaboost的公司财务预警系统,采用1000组公司财务数据对模型进行训练和测试,通过比较经典BP-Adaboost模型和经过遗传算法优化的BP-Adaboost模型的预测效果,发现后者在财务危机预警中具有较高的准确度和鲁棒性。实践结果显示,采用10个网络结构为10-12-1的BP神经网络构成的BP-Adaboost并结合遗传算法,可为上市公司财务危机预警提供一种新的视角和解决方案。In the rapidly changing market environment,the financial stability of listed companies has an important impact on the healthy development of national economy and financial market.Therefore,it is very important to construct an effective financial crisis early warning system.This paper establishes a corporate financial early warning system based on BP_Adaboost,and uses 1000 sets of corporate financial data to train and test the model.By comparing the prediction effect of classical BP_Adaboost model and genetic algorithm optimized BP_Adaboost model,it is found that the latter has higher accuracy and robustness in financial crisis early warning.The practice results show that BP_Adaboost,which is composed of 10 BP neural networks with 10-12-1 network structure and combined with genetic algorithm,can provide a new perspective and effective solution for financial crisis warning of listed companies.

关 键 词：财务预警 机器学习 BP-Adaboost 遗传算法 

分 类 号：TN391.9[电子电信—物理电子学]