Title :
Performance and diversity evaluation in hybrid and non-hybrid structures of ensembles
Author :
Canuto, Anne M P ; Oliveira, Ld.M. ; Xavier, João C., Jr. ; Santos, Ad.M. ; Abreu, Marjory C C
Author_Institution :
Dept. of Informatics & Appl. Math., Rio Grande do Norte Fed. Univ., Natal, Brazil
Abstract :
This paper presents a wide evaluation of performance and diversity in hybrid and non-hybrid structures of ensembles. In applying some diversity measures at the chosen ensemble, it is intended to analyse the effect of varying diversity in ensembles and how the variation of diversity can affect the performance of several combination methods (selection-based and combination-based methods). Finally, it is also intended to understand the reasons that some combination methods are more affected by variation in diversity.
Keywords :
pattern classification; combination methods; diversity evaluation; ensembles; pattern classification; performance evaluation; Competitive intelligence; Computational intelligence; Diversity methods; Diversity reception; Informatics; Mathematics; Pattern recognition; Performance analysis; Topology; Training data;
Conference_Titel :
Hybrid Intelligent Systems, 2005. HIS '05. Fifth International Conference on
Print_ISBN :
0-7695-2457-5
DOI :
10.1109/ICHIS.2005.87