Open AccessApplication of neural network language models based on distributive semantics for ontological modeling of the domain
Author(s) -
Maxim G. Shishaev,
V. V. Dikovitsky,
Vadim Pimeshkov
Publication year - 2022
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2182/1/012033
Subject(s) - simple knowledge organization system , computer science , ontology , artificial intelligence , natural language processing , distributive property , semantics (computer science) , artificial neural network , domain (mathematical analysis) , natural language understanding , natural language , semantic web , semantic computing , programming language , semantic analytics , mathematical analysis , philosophy , mathematics , epistemology , pure mathematics
The article discusses the technology of automated formation of SKOS-ontologies for semantic modeling of the subject area, based on natural language texts analysis. The technology is based on neural network and distributive (vector) language models. A brief description of the content and formulation of the problem of extracting concepts and relations from natural language texts is given, the results of constructing a neural network classifier of SKOS relations using the Glove vector model, as well as an example of using the technology to construct a fragment of an applied SKOS ontology are given.