Open AccessAPPLICATION OF SELF-ORGANIZING KOHONEN MAPS FOR ANALYSIS OF THE PATENT BASE
Author(s) -
G. Pastuhova
Publication year - 2021
Publication title -
nacionalʹnaâ associaciâ učënyh/nacionalʹnaâ associaciâ učenyh
Language(s) - English
Resource type - Journals
eISSN - 2782-2869
pISSN - 2413-5291
DOI - 10.31618/nas.2413-5291.2021.2.65.392
Subject(s) - self organizing map , typology , cluster analysis , basis (linear algebra) , computer science , artificial intelligence , set (abstract data type) , base (topology) , adaptation (eye) , data mining , machine learning , mathematics , geography , psychology , geometry , archaeology , programming language , mathematical analysis , neuroscience
One of the clustering technologies is considered - self-organizing Kohonen networks, bottlenecks for data analysis with similar algorithms are analyzed. The general problems of the adaptation of mathematical models and the applicability of the clustering algorithms themselves are touched upon.The classification problem is one of the most ancient problems, the essence of which is to divide the set of objects under study into homogeneous groups in a certain sense. The basis for the classification is dictated by the nature of what we are classifying, although sometimes it is necessary to take as the basis such metrics for which there are objective ways to measure them.You also need to clearly distinguish between classification and typology, the latter is much broader. Typology is understood as a method of scientific knowledge, based on the dismemberment of objects and their grouping using a generalized, idealized model or type.