A rough-set-based two-class classifier for large imbalanced dataset

Abstract

The objective of this paper is to provide a rouch-set-based two-class classifier approach to classifying samples in large and imbalanced dataset. A database has plenty of hidden knowledge, which can be used in decision making to support commerce, research and other activities. Prediction is another form of expanding data analysis. It enables us to establish a data model using existing data and to predict the trend of data in future. In this paper, a method consists of data scaling, rough sets analysis and support vector machine with radial basis function (SVM-RBF), which is used to classify a large and imbalanced data set obtained in semiconductor industry. © Springer-Verlag Berlin Heidelberg 2010.