Open AccessMATHEMATICAL MODEL FOR AUTOMATIC CREATION THE SEMANTIC THESAURUS FOR THE SCIENTIFIC TEXT
Author(s) -
Oleg Volkovskiy,
Egor Kovylin
Publication year - 2019
Publication title -
sistemnì tehnologìï
Language(s) - English
Resource type - Journals
eISSN - 2707-7977
pISSN - 1562-9945
DOI - 10.34185/1562-9945-6-125-2019-08
Subject(s) - computer science , thesaurus , relevance (law) , information retrieval , natural language processing , vocabulary , field (mathematics) , semantic computing , semantic similarity , artificial intelligence , linguistics , semantic web , mathematics , philosophy , political science , pure mathematics , law
The paper deals with the issues related to the use of the algorithm of constructing the semantic model of the document for the creation the thesaurus of the scientific text terms in the natural language. The purpose of the paper is to develop an approach to create matching between elements of scientific text that do not have a direct syntactic link, but are semantically related to the one field. The relevance of the research is that the system does not use linguistic or vocabulary knowledge during its work, which makes it a universal tool for forming semantic correspondence between terms in a scientific text. Obtained results show that the semantic labels of a document have the highest number of intersections with semantic contours when they contain the largest number of semantically significant stems in their composition, which allows to make assumptions about a direct semantic connection between terms corresponding to such stems.