Title :
Maximum mutual information vector quantization
Author :
Wilcox, Lynn D. ; Niles, Les T.
Author_Institution :
Xerox PARC, Palo Alto, CA, USA
Abstract :
A method is proposed for designing a maximum mutual information (MMI) vector quantizer, for applications in which quantization is used to extract a set of discrete features for use in classification
Keywords :
vector quantisation; feature classification; feature extraction; maximum mutual information; vector quantization; Data mining; Distortion measurement; Entropy; Euclidean distance; Feature extraction; Mutual information; Training data; Vector quantization; Yield estimation;
Conference_Titel :
Information Theory, 1995. Proceedings., 1995 IEEE International Symposium on
Conference_Location :
Whistler, BC
Print_ISBN :
0-7803-2453-6
DOI :
10.1109/ISIT.1995.550421
