基于 LSSVM 的供应链金融业务中企业信用风险评估研究

摘要

In response to unscientific and imperfect enterprise credit risk assessment in supply chain finance at present, on the basis of combing and completing the indicator system of enterprise credit risk assessment, the paper applies the least squares support vector machine (LSSVM) to build an enterprise credit risk assessment model, and makes an analysis on the feasibility and superiority of the model. It draws the conclusion that the LSSVM assessment model is more effective with high accuracy and powerful universalisation, which provides a new method for assessing enterprise credit risk in the supply chain finance business.%针对目前供应链金融业务中企业信用风险评估不科学与不完善的问题，在对企业信用风险评估的指标体系进行梳理和完善的基础上，利用最小二乘支持向量机(LSSVM)的基本原理尝试构建了企业信用风险 LSSVM 评估模型，并对模型的可行性和优越性做了分析和验证。研究结果表明：LSSVM 评估模型收敛速度快、准确度较高、泛化能力强，为供应链金融业务中的企业信用风险评估提供了一种新的办法和思路。