A Novel Membership Function Definition for Fuzzy Classification

Authors : Nur UYLAŞ SATI

Pages : 404-411

Doi:10.53433/yyufbed.1239769

View : 38 | Download : 39

Publication Date : 2023-08-31

Article Type : Research Paper

Abstract :In this paper, a novel membership function is defined for fuzzy sets using a supervised learning approach. Firstly, the training dataset is separated using the previously defined polyhedral conic functions in a supervised learning approach. Then obtained polyhedral conic functions are used for defining a new membership function. After that, a new fuzzy classification algorithm is formed to classify fuzzy sets with a similar structure. The algorithm with all suggested methods is implemented on real-world datasets, and the performance values are compared with the state of art classification algorithms.
Keywords : Bulanık Sınıflandırma, Çok Yüzlü Konik Fonksiyonlar, Matematiksel Optimizasyon, Üyelik Fonksiyonu, Veri Madenciliği