Open AccessMarking a collection of texts with the keywords – automation aspects
Author(s) -
D. O. Zhaxybayev
Publication year - 2022
Publication title -
vestnik nacionalʹnoj inženernoj akademii respubliki kazahstan
Language(s) - English
Resource type - Journals
eISSN - 2709-4707
pISSN - 2709-4693
DOI - 10.47533/2020.1606-146x.137
Subject(s) - computer science , markup language , consistency (knowledge bases) , search engine indexing , information retrieval , automation , identification (biology) , natural language processing , keyword extraction , automatic indexing , test (biology) , artificial intelligence , world wide web , xml , mechanical engineering , paleontology , botany , biology , engineering
This article presents and discusses the results of automatic indexing of keywords in 27 functional collections of Russian texts in three functional styles: scholarly, journalistic and fiction. The approach to the processing of markup results is presented and the data on the consistency of experts are given. Depending on the nature of the project’s problems, the design tasks provide for an automated system of document and keyword identification. The aim of this study is to identify the problems of a modified automatic text scoring system (CAPT) with keywords and to analyse in detail the results of the scoring test in order to create the conditions for the next discourse. These functions constitute the content of one of the research stages aimed at creating an effective algorithm for CS extraction for Russian language.