A New Fuzzy Classifier Based on Simulated Annealing and Subtractive Clustering

TORUN Y., Tohumoglu G.

IEEE 17th Signal Processing and Communications Applications Conference, Antalya, Türkiye, 9 - 11 Nisan 2009, ss.694-697 identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Basıldığı Şehir: Antalya
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.694-697
  • Dokuz Eylül Üniversitesi Adresli: Hayır

Özet

In this study, subtractive clustering method is used in order to obtain membership functions and rule base of fuzzy classifier system. Neighborhood radii of subtractive clustering method which directly effects the system dynamic and number of rules in the fuzzy classifier, is obtained by new simulated annealing optimization algorithm. The proposed classifier is firstly applied on pap-smear lest which is used in diagnosis of cervical cancer. The classifier is also tested with some well known biomedical and other classification problems in the literature. The results obtained by the offered hybrid classifier have better performance than the other methods.