Formulation of Fuzzyc-Means Clustering Using Calculus of Variations and Twofold Membership Clusters
Sadaaki Miyamoto · Journal of Advanced Computational Intelligence and Intelligent Informatics · 2008
A membership matrix of fuzzyc-means clustering is associated with corresponding fuzzy classification rules as membership functions defined on the whole data space. We directly derive such functions in fuzzyc-means and possibilistic clustering using the calculus of variations, generalizing ordinary fuzzyc-means and deriving new twofold membership clustering using this framework.