Evaluation of classical and novel ensemble methods for handwritten word recognition

Handwritten text recognition is one of the most difficult problems in the field of pattern recognition. Recently, a number of classifier creation and combination methods, known as ensemble methods, have been proposed in the field of machine learning. They have shown improved recognition performance over single classifiers. In this paper a number of ensemble methods for handwritten word recognition are described, experimentally evaluated, and compared to each other. Those methods include classical, general purpose ensemble methods as well as novel ensemble methods specifically developed by the authors for handwritten word recognition. The aim of the paper is to investigate the potential of ensemble methods for improving the performance of handwriting recognition systems. The base recognition systems used in this paper are hidden Markov model classifiers.