An incremental learning algorithm for SVM based on voting principle

In this paper, we mainly adopt the voting combination method to implement the incremental learning for SVM. The incremental learning algorithm proposed by this paper has contained two parts in order to tackle different types of incremental learning cases, the first part is to deal with the on-line incremental learning and the second part is to deal with the batch incremental learning. In the final, we make the experiment to verify the validity and efficiency of such algorithm.