Rough sets method for SVM data preprocessing

Abstract

To improve the generalization performance and structure of SVM classifiers (SVCs), we introduce rough sets theory to the data preprocessing of SVCs. Three measures arc taken: removing duplicate samples from thc dataset, finding a reduct and then multiplying every attribute with its corresponding significance factor which equals to the dependency of decision attribute with respect to thc attribute. Experiment results on a UCI benchmark dataset and a practical steam turbine failure diagnosis problem show that the presented approach is feasible.