|
This article is cited in 7 scientific papers (total in 7 papers)
Parametrization in applied problems of search of empirical reasons
A. A. Grushoa, N. A. Grushoa, M. I. Zabezhailob, D. V. Smirnovc, E. E. Timoninaa a Institute of Informatics Problems, Federal Research Center “Computer Sciences and Control” of the Russian
Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
b A. A. Dorodnicyn Computing Center, Federal Research Center “Computer Sciences and Control” of the Russian
Academy of Sciences, 40 Vavilov Str., Moscow 119133, Russian Federation
c Sberbank of Russia, 19 Vavilov Str., Moscow 117999, Russian Federation
Abstract:
The authors define description of a finite class of objects in the form of a set of characteristics (parameters) of these objects as parametrization of the considered class. Besides identification of objects, the problem of the causality that some objects have property $P$ exists. For the solution of this task, in the conditions of emergence of new objects, an initial set of characteristics cannot be enough. In this case, it is necessary to change parametrization. The paper is devoted to creation of methods of changes of the initial parametrization in the problem of specification of the empirical causality of emergence of the property $P$ at expansion of basic data. The constructed methods are shown on practical examples.
Keywords:
JSM-methods of artificial intelligence; parametrization of classes of objects; empirical reason;
authentication.
Received: 10.06.2018
Citation:
A. A. Grusho, N. A. Grusho, M. I. Zabezhailo, D. V. Smirnov, E. E. Timonina, “Parametrization in applied problems of search of empirical reasons”, Inform. Primen., 12:3 (2018), 62–66
Linking options:
https://www.mathnet.ru/eng/ia548 https://www.mathnet.ru/eng/ia/v12/i3/p62
|
Statistics & downloads: |
Abstract page: | 255 | Full-text PDF : | 67 | References: | 35 |
|