Controlling the diversity in classifier ensembles through a measure of agreement

Pattern Recognition

Volumn 38, Issue 11, 2005, Pages 2195-2199

  Zouari, Héla ^a   Heutte, Laurent ^a   Lecourtier, Yves ^a  

^a UNIVERSITÉ DE ROUEN   (France)

Author keywords

Classifier agreement; Classifier ensemble; Classifier simulation; Dependency; Diversity measure; Kappa measure; Output generation algorithm

Indexed keywords

ALGORITHMS; COMPUTER SIMULATION; CONTRACTS; KNOWLEDGE BASED SYSTEMS;

CLASSIFIER AGREEMENTS; CLASSIFIER ENSEMBLE; CLASSIFIER SIMULATION; DEPENDENCY; DIVERSITY MEASURE; KAPPA MEASURE; OUTPUT GENERATION ALGORITHMS;

PATTERN RECOGNITION;

EID: 24044547645     PISSN: 00313203     EISSN: None     Source Type: Journal    
DOI: 10.1016/j.patcog.2005.02.012     Document Type: Article

References (5)

1. J. Fleiss Statistical Methods for Rates and Proportions 1981 Wiley New York
2. L.I. Kuncheva, and R.K. Kountchev Generating classifier outputs of fixed accuracy and diversity Pattern Recognition Lett. 23 2002 593 600
3. L.I. Kuncheva, and C.J. Whitaker Measures of diversity in classifier ensembles and their relationship with the ensemble accuracy Mach. Learning 51 2003 181 207
4. V.D. Lecce, G. Dimauro, A. Guerrierro, S. Impedovo, G. Pirlo, A. Salzo, Classifier combination: the role of a priori knowledge, Seventh International Workshop on Frontiers in Handwriting Recognition, Amsterdam, The Netherlands, 2000, 143-152.
5. H. Zouari, L. Heutte, Y. Lecourtier, M.A. Alimi, Simulating classifier outputs for evaluating parallel combination method, Fourth International Workshop on Multiple Classifier Systems, Guildford, UK, Lecture Notes in Computer Science, vol. 2709, 2003, pp. 296-305.