A Class-Incremental Learning Method for Multi-Class Support Vector Machines in Text Classification

Abstract

To solve multi-class problems of support vector machines (SVM) more efficiently, a novel framework, which we call class-incremental learning (CIL), is proposed in this paper. CIL consists of two phases: incremental feature selection and incremental training, for updating the knowledge of old SVM classifiers in text classification when new classes are added to the system. CIL reuses the old models of the classifier and learns only one binary sub-classifier with an additional phase of feature selection when a new class comes. In the testing phase, current classifier is applied to the vectors' projections on the sub-spaces concerned. CIL can serve as a flexible approach for all binary classification algorithms in text classification. Our experiment shows that the CIL-based SVM was not only substantially faster in training time than the popular batch SVM learning methods such as 1-against-rest, 1-against-1 and divide-by-2 but also almost competed to the best performances in effectiveness of them.