In this paper, a novel hybrid approach to extract rules from support vector machine and support vector regression (SVM/SVR) is presented. The hybrid has three phases: Extensive experiments are conducted on three benchmark classification problems, four bank bankruptcy prediction problems and five benchmark regression problems. We conclude that the rules obtained after feature selection perform comparably to those extracted from all features. Further, comprehensibility is also improved after feature selection.
International Journal of Information and Decision Sciences – Inderscience Publishers
Published: Jan 1, 2011