epsilon-Tube Based Pattern Selection for Support Vector Machines

Dongil Kim, Sungzoon Cho

Abstract: The training time complexity of Support Vector Regression (SVR) is O(N 3 ). Hence, it takes long time to train a large dataset. In this paper, we propose a pattern selection method to reduce the training time of SVR. With multiple bootstrap samples, we estimate ε-tube. Probabilities are computed for each pattern to fall inside ε-tube. Those patterns with higher probabilities are selected stochastically. To evaluate the new method, the experiments for 4 datasets have been done. The proposed method resulted in the best performance among all methods, and even its performance was found stable.