Survey on Approaches, Problems and Applications of Ensemble of Classifiers

Abstract multiple classifiers are learned from same dataset and this multiple trained classifiers are used to predict the unlabeled data, this approach is known as Ensemble of classifiers. Ensemble of classifiers outperforms than using single classifier. Well known approaches of ensembles are boosting, random subspace, bagging, and random forest. Although its success some approaches has some limitations this paper describes distinct designs for ensemble of classifiers, different works to improve the ensemble of classifiers and applications of the ensemble of classifier.