A New Cluster Validity Criterion for Fuzzy C-Regression Models Clustering and Its Application to Fuzzy Model Identification

Autor:	Chung-Chun Kung, Jui-Yiao Su
Rok vydání:	2008
Předmět:	ComputingMethodologies_PATTERNRECOGNITION Fuzzy clustering Fuzzy classification Linear form Cluster (physics) Affine transformation Data mining computer.software_genre Cluster analysis computer Fuzzy logic Defuzzification Mathematics
Zdroj:	IFAC Proceedings Volumes. 41:7445-7450
ISSN:	1474-6670
DOI:	10.3182/20080706-5-kr-1001.01258
Popis:	In this paper, a new cluster validity criterion for fuzzy c-regression models (FCRM) clustering algorithm with affine linear functional cluster representatives is proposed. The proposed cluster validity criterion calculates the overall compactness and separateness of the FCRM partition and then determines the appropriate number of clusters. Besides, its application to fuzzy model identification is discussed. A TS fuzzy model identification algorithm is proposed to extract compact number of IF-THEN rules from data. Two simulation examples are provided to demonstrate the potential of the proposed cluster validity criterion and the accuracy of the constructed T-S fuzzy model.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::46d5d2dacc08bbde15cdf1562e132f4c https://doi.org/10.3182/20080706-5-kr-1001.01258 Zobrazit plný text záznamu