A Safe Semi-supervised Classification Algorithm Based on Disagreement

In order to improve the performance of semi-supervised classifier, a safe disagreement-based semi-supervised classification algorithm named Safe Co-SSC was proposed. The limited labeled samples were divided into three equal training sets and used to train three classifiers by a supervised learning algorithm. A large number of unlabeled samples were used to increase the differences between the classifiers and the weighted voting strategy was used to achieve pseudo-labeled for unlabeled samples. Passing through secondary verification, the ones making classifier error minimum were added into the labeled samples set. Finally, the experiment was carried out on the UCI data set, the results showed that the proposed algorithm had higher classification rate and sample labeling rate.