An Analysis on Ensemble Methods In Classification Tasks

Abstract: Ensemble approaches in classification is a very popular research area in recent years. An ensemble consists of a set of individually trained classifiers such as neural networks or decision trees whose predictions are combined for classifying new instances. It integrates multiple classifiers to build a classification model, and also used for improving the prediction performance. “Diversity” is one of the elements required for accurate prediction when using an ensemble. It is used in wide area of research such as statistics, pattern recognition, and machine learning. This paper presents an updated survey of ensemble methods in classification tasks, and introducing a new taxonomy for characterizing them. The new taxonomy presented here, is based on five dimensions: inducer, combiner, diversity, size, and members dependency.

Keywords: Ensemble-methods, Classification, Boosting, Bagging, Random Subspaces, Random Subspaces, Rotation Forest, and Extended space Forest.