Incorporating Prior Knowledge in Support Vector Machines: Retrospect and Prospect

In the last few years, several works in the literature have addressed the problem of incorporating knowledge into support vector machines. The importance of this problem derives from the fact that, once incorporated, the knowledge can act as numerous training instances by which the machine performance would be considerably enhanced. In this paper, we propose a taxonomy for characterizing knowledge incorporation, briefly survey major knowledge incorporation methods described in the literatures, and provide a prospect for knowledge incorporation. Hopefully, this work will stimulate other studies aimed at a more comprehensive analysis of knowledge incorporation into support vector machines.