基于Adaboost-DPSO-SVM模型的供应链金融信用风险评价研究

数字化赋能供应链金融创新是推动我国产业链、供应链持续稳定优化升级的重要内容,在全球金融风险急剧增加的背景下,供应链金融与大数据、区块链、物联网、人工智能等技术相融合,成为解决我国中小企业融资授信问题的有效方式之一。本文在对供应链金融信用风险评价指标进行特征选择的基础上,采用一种动态变异的粒子群算法(DPSO)和AdaBoost算法对SVM进行协同优化和集成,建立了Adaboost-DPSO-SVM模型,并将该模型应用于我国新能源汽车行业供应链金融信用风险评价中,实验结果表明所建立的模型相对其他评价模型具有更好的分类识别性能。

Research on Supply Chain Financial Credit Risk Evaluation Based on AdaBoost-DPSO-SVM Model

Digital enabled supply chain financial innovation is an important content to promote the continuous,stable,optimization and upgrading of China.s industrial chain and supply chain. Under the background of the sharp increase of global financial risks,the integration of supply chain finance with big data,blockchain,internet of things,artificial intelligence and other technologies has become one of the effective ways to solve the financing and credit problems of small,medium-sized and micro enterprises in China. Based on the feature selection of supply chain financial credit risk evaluation indexes,this paper uses dynamic mutation particle swarm optimization algorithm and AdaBoost algorithm to jointly optimize and integrate SVM,establishes AdaBoost-DPSOSVM model,and applies the model to the supply chain financial credit risk evaluation of new energy automobile industry in China,the experimental results show that the model has better classification and recognition performance than other evaluation models.