Development of adaptive neural network classifiers for analysis of time seriesстатья
Статья опубликована в журнале из списка RSCI Web of Science
Статья опубликована в журнале из перечня ВАК
Статья опубликована в журнале из списка Web of Science и/или Scopus
Дата последнего поиска статьи во внешних источниках: 29 мая 2015 г.
Местоположение издательства:Road Town, United Kingdom
Первая страница:33
Последняя страница:34
Аннотация:In this paper, basic principles for the use of a hierarchical system of neural network classifiers for analysis of dynamics of complex objects and finding out its structure in an objective way are presented. Earlier, the authors developed an algorithm for adaptive construction of these classifiers to analyze a class of objects represented as static patterns. An algorithm was applied to vowels, textures, and isolated words. The further development of the algorithm includes its generalization to the case of more complex objects represented, in particular, in the form of time series of various natures, testing on model problems and standard databases, as well as application to a number of practical problems.