Rough Set Methods for Constructing Support Vector Machines

Analyzed the generalities and specialties of Rough Sets Theory (RST) and Support Vector Machines (SVM) in knowledge representation and process of regression, a minimum decision network combining RST with SVM in intelligence processing is investigated, and a kind of SVM information process system on RST is proposed for forecasting. Using RST on the advantage of dealing with great data and eliminating redundant information, the system reduced the training data of SVM, and overcame the disadvantage of great data and slow training speed. The experimental results proved that the presented approach could achieve greater forecasting accuracy and generalization ability than the BP neural network and standard SVM.