Title :
Pattern classifiers with adaptive distances
Author :
de M Silva Filho, T. ; de Souza, Renata M. C. R.
Author_Institution :
Centro de Inf., Univ. Fed. de Pernambuco, Recife, Brazil
fDate :
July 31 2011-Aug. 5 2011
Abstract :
This paper presents learning vector quantization classifiers with adaptive distances. The classifiers furnish discriminant class regions from the input data set that are represented by prototypes. In order to compare prototypes and patterns, the classifiers use adaptive distances that change at each iteration and are different from one class to another or from one prototype to another. Experiments with real and synthetic data sets demonstrate the usefulness of these classifiers.
Keywords :
learning (artificial intelligence); pattern classification; vector quantisation; adaptive distances; discriminant class regions; input data set; learning vector quantization classifiers; pattern classifiers; Classification algorithms; Equations; Error analysis; Euclidean distance; Prototypes; Shape; Training;
Conference_Titel :
Neural Networks (IJCNN), The 2011 International Joint Conference on
Conference_Location :
San Jose, CA
Print_ISBN :
978-1-4244-9635-8
DOI :
10.1109/IJCNN.2011.6033403