Title :
A clustering model on similarities for dynamic changes
Author_Institution :
Inst. of Policy & Planning Sci., Tsukuba Univ.
Abstract :
Presents a clustering model which can classify 3-way data consisting of similarities among objects over times by using the adaptable metric for representing the similarities of objects on different times. A well-known problem with the classification for 3-way data is how to represent the differences through times. As the reasonable estimation of the differences, the estimated similarities on the different times using 11-norm is shown using the exact data as similarities of objects
Keywords :
fuzzy set theory; matrix algebra; pattern classification; 11-norm; 3-way data; clustering model; dynamic changes; similarities; Additives; Clustering methods; Electronic mail; Least squares methods; Pareto optimization;
Conference_Titel :
Systems, Man, and Cybernetics, 1998. 1998 IEEE International Conference on
Conference_Location :
San Diego, CA
Print_ISBN :
0-7803-4778-1
DOI :
10.1109/ICSMC.1998.725078
