Pruning a classifier based on a self-organizing map using Boolean function formalization

Author

Lobo, Victor Jose ; Swiniarski, Roman ; Moura-Pires, Fernando

Author_Institution

Escola Naval, Portugeses Naval Acad., Almada, Portugal

Volume

3

fYear

1998

fDate

4-9 May 1998

Firstpage

1910

Abstract

An algorithm is presented to minimize the number of neurons needed for a classifier based on Kohonens self-organizing maps (SOM), or on any other “code-book type” (or “prototype based”) classifier such as Kohonens linear vector quantization (LVQ), K-means or nearest neighbor. The neuron minimization problem is formalized as a problem of simplification of Boolean functions, and a geometric interpretation of this simplification is provided. A step by step example with an illustrative classification problem is given

Keywords

Boolean functions; covariance matrices; geometry; learning (artificial intelligence); pattern classification; self-organising feature maps; vector quantisation; Boolean function formalization; K-means classifier; Kohonens linear vector quantization classifier; Kohonens self-organizing maps; code-book type classifier; nearest neighbor classifier; neuron minimization problem; Boolean functions; Covariance matrix; Drives; Gaussian distribution; Nearest neighbor searches; Neural networks; Neurons; Organizing; Prototypes; Vector quantization;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks Proceedings, 1998. IEEE World Congress on Computational Intelligence. The 1998 IEEE International Joint Conference on

Conference_Location

Anchorage, AK

ISSN

1098-7576

Print_ISBN

0-7803-4859-1

Type

conf

DOI

10.1109/IJCNN.1998.687150

Filename

687150