IJPAM: Volume 50, No. 2 (2009)

ARTIFICIAL LEARNING CLASSIFIERS WITH
THE POTENTIAL FUNCTIONS. AN APPROACH
OF THE GENERAL CASE

Alina Constantinescu
Department of Mathematics
University Valahia - Târgoviste
2, G. Enescu Str., Târgoviste, 130056, ROMANIA
e-mail: alinaconsta@yahoo.com


Abstract.One of the most important applications of the neural networks is in statistical classification theory. In the specialized literature are procedures which use the artificial process of intelligent machines learning [#!5!#]. The Potential Functions Method for two classes is one of them. This paper presents an extension of Potential Functions Method. Our procedure is suitable for the general case when the data set has many classes. We start from the classical method for $n=2$ classes and then we provide a procedure for $n\ge3$.

Received: August 14, 2008

AMS Subject Classification: --???--

Key Words and Phrases: neural network, classification, potential functions

Source: International Journal of Pure and Applied Mathematics
ISSN: 1311-8080
Year: 2009
Volume: 50
Issue: 2