Integrating nonlinear graph based dimensionality reduction schemes with SVMs for credit rating forecasting |
| |
| Authors: | Shian-Chang Huang |
| |
| Affiliation: | 1. Non-Traditional Security Center, Huazhong University of Science and Technology, China;2. The State Key Laboratory of ISN, Xidian University, China;3. The Department of Comnet, Aalto University, Finland;4. The School of Computer Science, Huazhong University of Science and Technology, China;5. The Department of Computer Science, St. Francis Xavier University, Canada;1. Department of Biosystems, Faculty of Bioscience Engineering, Katholieke Universiteit Leuven – KULeuven, Kasteelpark Arenberg 30, B-3001 Heverlee, Belgium;2. Departamento de Quimica, Universidade Federal da Para?ba, CCEN, Caixa Postal 5093, CEP 58051-970 Joao Pessoa, PB, Brazil;3. Divisao de Engenharia Eletronica, Instituto Tecnologico de Aeronautica, CEP 12228-900 Sao Jose dos Campos, SP, Brazil;4. Analytical Chemistry and Pharmaceutical Technology, Center for Pharmaceutical Research (CePhaR), Vrije Universiteit Brussel (VUB), Laarbeeklaan 103, 1090 Brussels, Belgium;1. Université catholique de Louvain, Louvain School of Management, Place des Doyens 1, 1348 Louvain-la-Neuve, Belgium;2. Université de Lille – SKEMA Business School, Université Lille 2, Rue de Mulhouse 2 – BP 381, 59020 Lille Cédex, France;1. Department of Informatics, Kaunas Faculty, Vilnius University, Muitines Str. 8, Kaunas, Lithuania;2. Currently works in: Center of Information Systems Design Technologies, Department of Information Systems, Kaunas University of Technology, Studentu Str. 50-313a, Kaunas, Lithuania;1. BW PARTNER, Hauptstrasse 41, 70563 Stuttgart, Germany;2. Cranfield School of Management, Cranfield, Bedford MK43 0AL, United Kingdom |
| |
| Abstract: | By integrating graph based nonlinear dimensionality reduction with support vector machines (SVMs), this study develops a novel prediction model for credit ratings forecasting. SVMs have been successfully applied in numerous areas, and have demonstrated excellent performance. However, due to the high dimensionality and nonlinear distribution of the input data, this study employed a kernel graph embedding (KGE) scheme to reduce the dimensionality of input data, and enhance the performance of SVM classifiers. Empirical results indicated that one-vs-one SVM with KGE outperforms other multi-class SVMs and traditional classifiers. Compared with other dimensionality reduction methods the performance improvement owing to KGE is significant. |
| |
| Keywords: | Kernel graph embedding Dimensionality reduction Support vector machine Multi-class classification Credit rating |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
