An approach to the automatic linguistic ontology construction to determine the interests of social networks users

Abstract

Social networks provide researchers with the opportunity to obtain an array of text data for further analysis within a certain subject area. Each subject area has its own specific professional vocabulary and writing style. When defining the subject area of text material there is a big problem with building dictionaries, thesauri, and ontologies. In this article a linguistic ontology is considered under ontology and which is aimed to determine the subject area of text material. An algorithm for the automatic construction of an ontology based on the Wikidata knowledge graph is presented. The task is to map a set of objects of the Wikidata knowledge graph to a set of entities of a linguistic ontology. The article pro-poses an algorithm for determining the degree of belonging of the text material to the subject area. Experiments on assessing the time of building an ontology and the applicability of the obtained linguistic ontologies to the problem of determining the degree of belonging of text materials in the subject area have shown: the running time of the algorithm and the number of terms in the formed ontology are directly proportional to the number of analyzed properties and Wikidata objects; the formed linguistic ontology is applicable to the problem of determining the degree of belonging of a text to a subject area