UDC 005.519.8
ANALYSIS OF WORDS SEMANTIC CLOSURE WITH KOHONEN CARDS
A. V. Gigolaev, a student, group 443; This email address is being protected from spambots. You need JavaScript enabled to view it.
N. I. Tsukanova, PhD (technical sciences), Associate Professor, Department of Computing and Applied Mathematics, RSREU, Ryazan; This email address is being protected from spambots. You need JavaScript enabled to view it.
The questions of Kohonen maps application for text word organization into groups according to context are considered. It is suggested to use the matrix of words succession frequency as one model of document text. The aim of the paper is to investigate the applicability of Kohonen self-organizing maps for the clustering of words of the text in terms of meaning, on the basis of the characteristics of words connection among themselves, to develop a program taking into account the characteristics of the problem being solved. The paper considers the following issues: presentation of the text in the form of a fuzzy graph (fuzzy relation), ways to determine the strongest path between the vertices of the graph and, therefore, the strongest connection between words, the formation of «objects – signs» matrix for solving the clustering problem, the choice of the metric for evaluation of closeness of clustering objects. The results of the study obtained with the help of Kohonen mapping program are presented.
Key words: text, words, context, clustering, Kohonen map, metric, neural network, fuzzy sets and fuzzy relations, fuzzy graph.