Comparison of multi-objective algorithms applied to feature selection

TÜRKŞEN, ÖZLEM; Vieira, Susana; Madeira, José; APAYDIN, AYŞEN; Sousa, João

doi:10.1007/978-3-642-30278-7_28

Comparison of multi-objective algorithms applied to feature selection

Atıf İçin Kopyala

TÜRKŞEN Ö., Vieira S. M., Madeira J. F. A., APAYDIN A., Sousa J. M. C.

Studies in Fuzziness and Soft Computing, cilt.285, ss.359-375, 2013 (SCI-Expanded)

Yayın Türü: Makale / Tam Makale
Cilt numarası: 285
Basım Tarihi: 2013
Doi Numarası: 10.1007/978-3-642-30278-7_28
Dergi Adı: Studies in Fuzziness and Soft Computing
Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, zbMATH
Sayfa Sayıları: ss.359-375
Ankara Üniversitesi Adresli: Evet

Özet

The feature selection problem can be formulated as a multiobjective optimization (MOO) problem, as it involves the minimization of the feature subset cardinality and the misclassification error. In this chapter, a comparison of MOO algorithms applied to feature selection is presented. The used MOO methods are: Nondominated Sorting Genetic Algorithm II (NSGA-II), Archived Multi Objective Simulated Annealing (AMOSA), and Direct Multi Search (DMS). To test the feature subset solutions, Takagi-Sugeno fuzzy models are used as classifiers. To solve the feature selection problem, AMOSA was adapted to deal with discrete optimization. The multiobjective methods are applied to four benchmark datasets used in the literature and the obtained results are compared and discussed. © Springer-Verlag 2013.