An Accelerated SMO-type Online Learning Algorithm

摘要

In order to accelerate the learning speed for online learning algorithm,a fast support vector machine online learning algorithm is presented in this paper.In the proposed algorithm,the learning condition is relaxed and a novel learning strategy is presented while Sequential Minimal Optimization (SMO) training method which has been improved by Keerthi.is embedded In order to verify the performance of the proposed algorithm,it has been applied to seven UCI datasets and a benchmark problem.Experimental results show that the novel algorithm is very faster than Online Support Vector Classifier (OSVC),SimpleSVM algorithms without losing generalized performance.