Title :
A subclass-based mixture model for pattern recognition
Author :
Kudo, Mineichi ; Tenmoto, Hiroshi ; Sumiyoshi, Satoru ; Shimbo, Masaru
Author_Institution :
Div. of Syst. & Inf. Eng., Hokkaido Univ., Sapporo, Japan
Abstract :
A classifier based on a mixture model is proposed. The expectation maximisation algorithm for construction of a mixture density is sensitive to the initial densities. It is also difficult to determine the optimal number of component densities. In this study, we construct a mixture density on the basis of a hyper-rectangles found in the subclass method, in which the number of components is determined automatically. Experimental results show the effectiveness of this approach
Keywords :
maximum likelihood estimation; pattern classification; classifier; expectation maximisation algorithm; hyper-rectangles; mixture density construction; pattern recognition; subclass-based mixture model; Clustering algorithms; Covariance matrix; Kernel; Maximum likelihood estimation; Neural networks; Pattern recognition; Piecewise linear approximation; Piecewise linear techniques; Systems engineering and theory;
Conference_Titel :
Pattern Recognition, 1998. Proceedings. Fourteenth International Conference on
Conference_Location :
Brisbane, Qld.
Print_ISBN :
0-8186-8512-3
DOI :
10.1109/ICPR.1998.711288