Supply chain finance credit risk assessment using support vector machine–based ensemble improved with noise elimination

• DOI: 10.1177/1550147720903631
• Publication Date: 2020-02-03T11:57:31Z
• Authors: Ying Liu; Lihua Huang
• Keywords: Artificial intelligence; Support vector machine; Economics; Structured support vector machine; Bankruptcy Prediction and Credit Scoring Models; Social Sciences; Handling Imbalanced Data in Classification Problems; FOS: Mechanical engineering; Business, Management and Accounting; Noise (video); 02 engineering and technology; Engineering; Cluster analysis; Artificial Intelligence; Accounting; Support Vector Machines; Machine learning; Image (mathematics); 0202 electrical engineering, electronic engineering, information engineering; Ensemble Methods; Data mining; Credit risk; Ensemble Learning; Credit Scoring; Mechanical Engineering; Particle swarm optimization; Comminution in Mineral Processing; QA75.5-76.95; Computer science; Algorithm; Electronic computers. Computer science; Computer Science; Physical Sciences; Finance

Recently, support vector machines, a supervised learning algorithm, have been widely used in the scope of credit risk management. However, noise may increase complexity algorithm building and destroy performance classifier. In our work, we propose an ensemble machine model to solve assessment supply chain finance, combined with reducing noises method. The main characteristics this approach include that (1) novel filtering scheme avoids noisy examples based on fuzzy clustering principal component analysis is proposed remove both attribute class achieve optimal clean set, (2) classifiers, improved particle swarm optimization are seen as classifiers. Then, obtained final classification results by combining finally individual prediction through AdaBoosting new sample set. Some experiments applied financial China’s listed companies. Results indicate accuracy can be increased applying approach.