On the combination of kernels for support vector classifiers

The problem of combining different sources of information arises in several situations, for instance, the classification of data with asymmetric similarity matrices or the construction of an optimal classifier from a collection of kernels. Often, each source of information can be expressed as a kernel (similarity) matrix and, therefore, a collection of kernels is available. In this paper we propose a new class of methods in order to produce, for classification purposes, an unique and optimal kernel. Then, the constructed kernel is used to train a Support Vector Machine (SVM). The key ideas within the kernel construction are two: the quantification, relative to the classification labels, of the difference of information among the kernels; and the extension of the concept of linear combination of kernels to the concept of functional (matrix) combination of kernels. The proposed methods have been successfully evaluated and compared with other powerful classifiers and kernel combination techniques on a variety of artificial and real classification problems.