Publication

Abstract : Support Vector Machine (SVM) which is used to solve regression problems is referred to as Support Vector Regression (SVR). This work mainly focuses on improving the robustness of SVR along with the study of various kernels such as linear, polynomial, Radial Basis Function (RBF) and bessel. A novel idea, the Hampel’s weight function has been used to increase the accuracy of the SVR, and the name of the proposed procedure coined as Robust Weighted Support Vector Regression (RWSVR). The performance of the RWSVR has been studied with the help of various kernels by checking their efficiencies under the various error measures such as Mean Absolute Error (MAE), Median Absolute Error (MDAE), Mean Absolute Percent Error (MAPE) and Root Mean Square Error (RMSE). Further, the real and simulation study concluded that RWSVR along with the RBF kernel performs well over the other kernels. The RWSVR may be applied in statistical learning in order to get more accuracy than the existing procedures, even the data with contamination.