A Comparative Study on Binary Artificial Bee Colony Optimization Methods for Feature Selection

Ozger Z. B., BOLAT B., DİRİ B.

International Symposium on Innovations in Intelligent Systems and Applications (INISTA), Sinaia, Romanya, 2 - 05 Ağustos 2016 identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/inista.2016.7571836
  • Basıldığı Şehir: Sinaia
  • Basıldığı Ülke: Romanya
  • Yıldız Teknik Üniversitesi Adresli: Evet

Özet

Feature selection is a major pre-processing technique which aims to pick out distinctive features from whole dataset. In this way it is intended to reduce computational cost of the classification process. Artificial Bee Colony (ABC) algorithm is an evolutionary based swarm intelligence optimization method. In this study, some of the variants of binary ABC algorithms are implemented to the feature selection problem using 10 UCI datasets. The results show that ABC algorithm is useful for this area.

Abstract:

Feature selection is a major pre-processins¸ technique which aims to pick out distinctive features from whole dataset. In this way it is intended to reduce computational cost o the classification process. Artificial Bee Colony (ABC) algorithm is an evolutionary based swarm intelligence optimization method In this study, some of the variants of binary ABC algorithms are implemented to the feature selection problem using 10 UC datasets. The results show that ABC algorithm is useful for the area.