The perceptron-based hierarchical structure of classifiers constructed by the adaptive methodстатья
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:24
Последняя страница:28
Аннотация:A method to construct a hierarchical structure of neural network classifiers for large databases is proposed. Each of the nodes in the hierarchical tree is a three-layer perceptron classifying the patterns to be analyzed into a small number of classes (2-10). At the first stage of the classification, the membership of a pattern in a group is roughly determined. Each of the subsequent nodes of the hierarchical tree is created to process its group and classifies the pattern in more detail. The algorithm designed to group patterns automatically on the basis of certain features with the help of a perceptron is at the same time an algorithm of constructing the appropriate classifier node. The performance of the method was tested on model data.