A sufficient set of statistical parameters for the classification of Russian-language texts

A sufficient set of statistical parameters for the classification of Russian-language texts / V. V. Filimonov, A. A. Zhivodyorov, A. M. Amieva, E. D. Pykhova - DOI: 10.1063/1.5055095 // V International Young Researchers' Conference. Physics, Technologies, Innovation (PTI.2018) Proceedings of the V International Young Researchers Conference, Ekaterinburg, Russian Federation, 14-18 May 2018. AIP Conference Proceedings. - 2018. - V. 2015. - 020022-1—020022-6.

Документ доступен в ЦНБ УрО РАН:

Нет

Год:

2018

Связанные персоналии:

Нет

Вид издания:

сборник трудов конференции

Коллекции:

Издания ЦНБ УрО РАН >> ЦНБ Статьи

h1

Публичные страницы

Аннотация

The study aims to build a methodology for computer classification of Russian-language texts. In this paper we present the results of estimating various statistical parameters in terms of the degree to which they influence the correctness of the recognition of genre content of texts. The methods of discriminant and factor analysis are applied in the study. A set of parameters is constructed that simultaneously meets the requirements of informativeness and compactness. The work is carried out at the Department of Polygraphy and Web Design IRIT-RtF UrFU.