Probabilistic fuzzy clustering algorithm for fuzzy rules decomposition

Salgado, Paulo; Igrejas, Getúlio

http://hdl.handle.net/10198/2774

Use this identifier to reference this record.

Name:	Description:	Size:	Format:
RECPAD2007-1 - Probabilistic fuzzy clustering.pdf		41.48 KB	Adobe PDF	Download

Send Feedback

Authors

Salgado, Paulo

Igrejas, Getúlio

Abstract(s)

The Fuzzy C-Means (FCM) clustering algorithm is the best known and the most used method for fuzzy clustering and is generally applied to well defined sets of data. In this work a generalized Probabilistic Fuzzy C-Means (PFCM) algorithm is proposed and applied to fuzzy sets clustering. The methodology presented leads to a fuzzy partition of the fuzzy rules, one for each cluster, which corresponds to a new set of fuzzy sub-systems. When applied to the clustering of a flat fuzzy system the result is a set of decomposed sub-systems that will be conveniently linked into a Parallel Collaborative Structure.

Keywords

Fuzzy clustering

URI

http://hdl.handle.net/10198/2774

Citation

Salgado, Paulo; Igrejas, Getúlio (2007). Probabilistic fuzzy clustering algorithm for fuzzy rules decomposition. In RECPAD - 13º Conferência Portuguesa de Reconhecimento de Padrões. Lisboa