关键词: 支持向量机, 多项式核, 偏移量

Abstract: A problem for many kernel-based methods is that the amount of computation required to find the solution scales as O(N^3) , where N is the number of training examples. As an interesting version of SVM, LS-SVM reduces the complexity of standard SVM to O(N^2). Both SVM and LS-SVM are not suitable for the large scale regression problem. This paper proposes a modified LS-SVM with inhomogeneous polynomial kernel. Besides, the new LS-SVM omits the bias term. This newly proposed LS-SVM has a simpler model than the standard one. Simulations performed on artificial data sets show the modified LS-SVM speeds the training process of standard LS-SVM greatly without significant loss of accuracy.○○

Key words: support vector machines, polynomial kernel, bias term