FIELD: computer equipment.
SUBSTANCE: invention relates to computer engineering for processing data arrays. Technical result is achieved due to generating a first data structure, on which a first SDA data structure is formed, comprising elements of said first data structure, formation of a database of linguistic features (DLF), on which linguistic features of text elements (TE) of a linguistic sentence are identified, from which a database is formed, which is a DLF of text elements of a linguistic sentence; generating a second data structure, on which a second SDA data structure is formed, comprising elements of said second data structure; generating a third data structure, on which a third SDA data structure is formed, comprising elements of said third data structure; and formation of the fourth data structure, on which the fourth SDA data structure is formed, containing elements of said fourth data structure.
EFFECT: technical result consists in improvement of accuracy of preliminary processing of text in natural language for its subsequent indexing and processing.
33 cl, 45 dwg
