基于粒子群协同优化算法的供应链金融信用风险评价模型  

Financial Credit Risk Evaluation Model of Supply Chain Finance Based on Particle Swarm Cooperative Optimization Algorithm

作　　者：刘颖 张丽娟[4] 韩亚男[5] 庞丽艳 王帅 

机构地区：[1]吉林财经大学管理科学与信息工程学院,长春130117 [2]吉林省物流产业经济与智能物流重点实验室,长春130117 [3]吉林财经大学互联网金融重点实验室,长春130117 [4]长春工业大学计算机科学与工程学院,长春130012 [5]长春工业大学马克思主义学院,长春130012

出　　处：《吉林大学学报（理学版）》2018年第1期119-125,共7页Journal of Jilin University:Science Edition

基　　金：国家自然科学基金(批准号:61402193;61572225);长春市地院(校;所)合作专项基金(批准号:17DY009);国家社会科学基金(批准号:16BGL180);物流产业经济与智能物流吉林省高校重点实验室开放基金(批准号:201702);吉林省教育厅"十二五"社会科学研究规划项目(批准号:2015363)

摘　　要：针对供应链金融模式下信用风险评价精度受信用特征子集与模型参数影响的问题,提出一种粒子群协同优化信用风险评价模型.该模型在充分论证供应链金融风险特征指标体系的基础上,利用二进制粒子群算法优选特征子集,并对支持向量机(SVM)参数协同优化.对供应链金融信用风险评估进行实验,并与传统径向基支持向量机和主成分分析特征抽取方法对比,结果表明,该模型优选的特征子集和SVM参数能显著提高信用风险评价精度.Aiming at the problem that the accuracy of credit risk evaluation of supply chain finance mode was affected by credit feature subset and model parameters, we proposed a credit risk evaluation model with particle swarm cooperative optimization. On the basis of fully demonstrating the characteristic index system of supply chain financial risk, we used the binary particle swarm algorithm to optimize the feature subset and optimize parameters of support vector machines. We carried out an experiment on the risk evaluation of supply chain financial credit, and compared it with traditional radial basis support vector machines and feature extraction method of principal component analysis. The results show that the selected feature subset and SVM parameters of the proposed model can significantly improve the accuracy of credit risk evaluation.

关 键 词：供应链金融 信用风险评价 粒子群算法 支持向量机 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]