INTELLECTUAL DATA ANALYSIS BASED ON A RIVAL SIMILARITY FUNCTION
N. G. Zagoruiko
S. L. Sobolev Institute of Mathematics, Siberian Branch, Russian Academy of Sciences, Novosibirsk, Russia, E-mail: zag@math.nsk.ru
Pages: 211-217
Abstract
A uniform approach to constructing intellectual data analysis, or data mining is proposed. The approach is based on using a rival similarity function (FRiS function) representing the way of human estimation of similarity and dissimilarity. A brief description of data mining methods based on this approach is presented. Examples of solving model and real problems by these methods are illustrated.
|
