An evaluation of ensemble methods in handwritten word recognition based on feature selection

Handwritten text recognition is one of the most difficult problems in the field of pattern recognition. The combination of multiple classifiers has been proven to be able to increase the recognition rate in difficult problems when compared to single classifiers. In this paper, several novel methods for the creation of classifier ensembles are compared where the individual classifiers use different feature subsets. The methods are evaluated in the context of handwritten word recognition, using a hidden Markov model recognizer as basic classifier.