Skip to Main Content
Bootstrap technique has been successfully used in many signal processing systems for data classification. Some of such systems are based on ensemble-based algorithms. These algorithms use multiple classifiers, generally to improve classification performance: each classifier provides an alternative decision whose combination may provide a superior solution than the one provided by any single classifier. In this paper, UChooBoost, a new supervised learning ensemble-based algorithm for extended data, based on bootstrap technique, is proposed. UChoo classifier is used as weak learner. UChoo classifier gives extended results expression. These results are combined by using new weighted majority voting founded on extended result expression.