A ν-twin support vector machine based regression with automatic accuracy control

Abstract

This paper presents an efficient ź-Twin Support Vector Machine Based Regression Model with Automatic Accuracy Control (ź-TWSVR). This ź-TWSVR model is motivated by the celebrated ź-SVR model (Schlkoff et al. 1998) and recently introduced źź-TSVR model (Shao et al., Neural Comput Applic 23(1):175---185, 2013). The ź-TSVR model can automatically optimize the parameters źź1 and źź2 according to the structure of the data such that at most certain specified fraction ź1(respectively ź2) of data points contribute to the errors in up (respectively down) bound regressor. The ź-TWSVR formulation constructs a pair of optimization problems which are mathematically derived from a related ź-TWSVM formulation (Peng, Neural Netw 23(3):365---372, 2010) and making use of an important result of Bi and Bennett (Neurocomputing 55(1):79---108, 2003). The experimental results on artificial and UCI benchmark datasets show the efficacy of the proposed model in practice.