Weighted Support Vector Regression Algorithm Based on Data Description

Abstract

In order to overcome the overfitting problem caused by noises and outliers in support vector regression (SVR) ,a weighted coefficient model based on support vector data description (SVDD) is presented in this paper. The weighted coefficient value to each input sample is confirmed according to its distance to the center of the smallest enclosing hypersphere in the feature space. The proposed model is applied to weighted support vector regression (WSVR) for 1-dimensional data set simulation. Simulation results indicate that the proposed method actually reduces the error of regression and yields higher accuracy than support vector regression (SVR) does.