Non-Euclidean Genetic FCM Clustering Algorithm
Sergio López García, Luis Magdalena, Juan R. Velasco · Studies in fuzziness and soft computing · 2002
The standard FCM clustering algorithm is a powerful mathematical tool widely used in many practical problems. Nevertheless, it is dependent on initial conditions and either the number of clusters and the distance definition must be predefined. In [15,11] the authors presented the Genetic FCM clustering, that improves the first and second drawbacks, but not the third one. This article shows how the definition of the distance can be included in the genetic structure. Several results applied to the Iris data set are also shown. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.