An incremental learning algorithm based on support vector domain classifier

Incremental learning technique is usually used to solve large-scale problem. We firstly gave a modified support vector machine (SVM) classification method -- support vector domain classifier (SVDC), then an incremental learning algorithm based on SVDC was proposed. The basic idea of this incremental algorithm is to obtain the initial target concepts using SVDC during the training procedure and then update these target concepts by an updating model. Different from the existed incremental learning approaches, in our algorithm, the model updating procedure equals to solve a quadratic programming (QP) problem, and the updated model still owns the property of spars solution. Compared with other existed incremental learning algorithms, the inverse procedure of our algorithm (i.e. decreasing learning) is easy to conduct without extra computation. Experiment results show our algorithm is effective and feasible.