Analysis of Different Binarization Techniques within Whale Optimization Algorithm

Golcuk I., ÖZSOYDAN F. B., DURMAZ E. D.

Innovations in Intelligent Systems and Applications Conference (ASYU), İzmir, Türkiye, 31 Ekim - 02 Kasım 2019, ss.72-76, (Tam Metin Bildiri) identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası:
  • Doi Numarası: 10.1109/asyu48272.2019.8946368
  • Basıldığı Şehir: İzmir
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.72-76
  • Anahtar Kelimeler: binary optimization, binarization techniques, whale optimization algorithm
  • Dokuz Eylül Üniversitesi Adresli: Evet

Özet

In this study, Whale Optimization Algorithm (WOA), which is a recently proposed swarm intelligence-based algorithm, has been used to solve binary optimization problems. As the WOA algorithm has been developed for optimizing realvalued functions, binary versions of the WOA algorithm have been adopted for the binary optimization problems. The performance of modulation-, normalization-, s-shaped transfer function-, and angle modulation-based binarization approaches are compared. The proposed algorithms tested on the well-known benchmark problems, namely one-max, plateau, deceptive, and royal road. Our computational experiments show that angle modulation and normalization-based binarization approaches give the best results, and binary WOA is promising and has potential to handle difficult binary optimization problems.