A Novel Knowledge-Based Twin Support Vector Machine

In this paper we proposed a novel knowledge based twin support vector machine (TWSVM), in which the prior knowledge in the form of multiple polyhedral sets, each belonging to one of two categories, is incorporated into the Linear TWSVM. Different with the existing approaches, we applied the regularized TWSVM model and changed the regularization term to be 1-norm, which resulted in a linear programming that can be solved efficiently, furthermore we amended the constraints corresponding to the knowledge for some special cases. Experiments proved the efficiency and effectiveness of our new model.