Title :
N-dimensional views in fuzzy data analysis
Author :
Umayahara, K. ; Nakamori, Yoshiteru
Author_Institution :
Adv. Res. Alliance, Tsukuba Univ., Ibaraki, Japan
Abstract :
This paper considers the problem of detecting local substructures of a system in a high dimensional data space by applying the fuzzy clustering technique. First, a new objective function to improve existing approaches is proposed, and then an efficient algorithm for detecting clusters with different dimensionalities is presented. Finally, a new type of fuzzy modeling using elliptic membership functions is discussed
Keywords :
data analysis; fuzzy set theory; pattern recognition; cluster detection; elliptic membership functions; fuzzy clustering; fuzzy data analysis; fuzzy modeling; high dimensional data space; local substructure detection; objective function; Clustering algorithms; Data analysis; Ear; Eigenvalues and eigenfunctions; Linearity; Scattering; Shape; Stress;
Conference_Titel :
Intelligent Processing Systems, 1997. ICIPS '97. 1997 IEEE International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7803-4253-4
DOI :
10.1109/ICIPS.1997.672738
