Incorporating priori knowledge into linear programming support vector regression

Abstract

In order to obtain an accurate regression model from a small dataset, a novel linear program support vector regression with priori knowledge is presented in the paper. The algorithm incorporates the data that is possible biased from a priori simulator into the existing linear programming support vector regression by modifying optimization objectives and inequality constraints. Moreover, multiple kernels are introduced to achieve an accurate modeling for complex and changeful problems. Synthetic examples show that the proposed algorithm is effective, and that the obtained model is sparse and accurate.