Please use this identifier to cite or link to this item:

http://hdl.handle.net/10609/109819
Title: On the design of an ECOC-compliant genetic algorithm
Author: Bautista, Miguel Ángel
Escalera Guerrero, Sergio
Baró Solé, Xavier  
Pujol Vila, Oriol
Others: Internet Interdisciplinary Institute
Keywords: ECOC
Genetic Algorithms
Multi-class classification
Issue Date: 1-Feb-2014
Publisher: Pattern Recognition
Citation: Bautista, M.A., Escalera Guerrero, S., Baró Solé, X. & Pujol Vila, O. (2014). On the design of an ECOC-compliant genetic algorithm. Pattern Recognition, 47(2), 865-884. doi: 10.1016/j.patcog.2013.06.019
Also see: https://www.sciencedirect.com/science/article/abs/pii/S0031320313002719
Abstract: Genetic Algorithms (GA) have been previously applied to Error-Correcting Output Codes (ECOC) in state-of-the-art works in order to find a suitable coding matrix. Nevertheless, none of the presented techniques directly take into account the properties of the ECOC matrix. As a result the considered search space is unnecessarily large. In this paper, a novel Genetic strategy to optimize the ECOC coding step is presented. This novel strategy redefines the usual crossover and mutation operators in order to take into account the theoretical properties of the ECOC framework. Thus, it reduces the search space and lets the algorithm to converge faster. In addition, a novel operator that is able to enlarge the code in a smart way is introduced. The novel methodology is tested on several UCI datasets and four challenging computer vision problems. Furthermore, the analysis of the results done in terms of performance, code length and number of Support Vectors shows that the optimization process is able to find very efficient codes, in terms of the trade-off between classification performance and the number of classifiers. Finally, classification performance per dichotomizer results shows that the novel proposal is able to obtain similar or even better results while defining a more compact number of dichotomies and SVs compared to state-of-the-art approaches.
Language: English
URI: http://hdl.handle.net/10609/109819
ISSN: 0031-3203MIAR
Appears in Collections:Articles
Articles

Share:
Export:
Files in This Item:
File SizeFormat 
Baro_PR_ondesign.pdf1.79 MBAdobe PDFView/Open

Items in repository are protected by copyright, with all rights reserved, unless otherwise indicated.