作者: Guangrong Tong , Siwei Li
DOI: 10.1155/2015/197258
关键词: Credit risk 、 Support vector machine 、 Creditor 、 Isomap 、 Dimensionality reduction 、 Relevance vector machine 、 Computer science 、 Data mining 、 Key (cryptography)
摘要: Credit assessment is the basis and premise of credit risk management systems. Accurate scientific great significance to operational decisions shareholders, corporate creditors, management. Building a good reliable model key assessment. Traditional models are constructed using support vector machine (SVM) combined with certain traditional dimensionality reduction algorithms. When constructing such model, algorithms first applied reduce dimensions samples, so as prevent correlation samples’ characteristic index from being too high. Then, learning samples will be conducted SVM, in order carry out classification To further improve accuracy methods, this paper has introduced more cutting-edge algorithms, isometric feature mapping (Isomap) for reduction, used relevance (RVM) classification. It an Isomap-RVM it conduct financial analysis China's listed companies. The empirical shows that significantly higher than Isomap-SVM slightly PCA-RVM model. can correctly identify risks