DocumentCode :
3029508
Title :
Fixed point relational fuzzy clustering
Author :
Brouwer, Roelof K.
Author_Institution :
Dept. of Comput. Sci., Thompson Rivers Univ., Kamloops, BC
fYear :
2009
fDate :
10-12 Feb. 2009
Firstpage :
115
Lastpage :
120
Abstract :
The proposed relational fuzzy clustering method called FRFP (fuzzy relational fixed point) is not based on minimizing an objective function, as in traditional methods, but rather on determining a fixed point of a function of the desired membership matrix with the proximity matrix as parameter. The proposed method is compared to other relational clustering methods including NERFCM, Rouben´s method and Windhams AP method. A clustering quality index is calculated for doing the comparison. using various proximity matrices as input. Simulations show the method to be very effective and less computationally expensive than other fuzzy relational data clustering methods.
Keywords :
fuzzy set theory; pattern clustering; clustering quality index; fixed point relational fuzzy clustering; fuzzy relational fixed point; membership matrix; proximity matrix; Clustering algorithms; Clustering methods; Computational modeling; Fuzzy sets; Knowledge acquisition; Partitioning algorithms; Prototypes; Rivers; Robots; Robustness; Fuzzy clustering; cluster quality; clustering quality; multidimensional scaling; optimization; prototype-less fuzzy clustering; relational fuzzy clustering; visual representation;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Autonomous Robots and Agents, 2009. ICARA 2009. 4th International Conference on
Conference_Location :
Wellington
Print_ISBN :
978-1-4244-2712-3
Electronic_ISBN :
978-1-4244-2713-0
Type :
conf
DOI :
10.1109/ICARA.2000.4803942
Filename :
4803942
Link To Document :
بازگشت