An experimental comparison of ensemble of classifiers for biometric data

Multi-matcher systems based on ensembles of classifiers are presented. We performed an empirical comparison of ensemble methods. The obtained results are very encouraging, our results improved the average predictive accuracy obtained using the individual learners. We show that the method “random subspace” outperforms the other ensemble methods tested in this paper. In this work, we study the fusion among the scores obtained by on-line signatures verification methods and the fusion among the scores of the systems submitted to Fingerprint Verification Competition 2004 (FVC2004).