Computer science and information processes

Research on main methods of automatic processing, grouping and annotation of information

D. V. Tikhonov

Financial University under the Government of the Russian Federation (Yaroslavl branch), 150003, Russia, Yaroslavl, 12a Cooperative street

Abstract: The article examines main methods of automatic processing, grouping and annotation of information. It is shown that methods of automatic analysis of Data Mining are based on the use of certain statistical patterns (classification, regression), search for keywords, but do not use algorithms for linguistic processing of texts. Thus, automatic analysis of text information carried out by modern means of analytical processing is not able to process texts content. To compare two simple sentences by content, the resolution method was used. Studies have shown the unification algorithm does not take into account © Тихонов Д. В., 2025 INFORMATICS AND INFORMATION PROCESSES 112 News of the Kabardino-Balkarian Scientific Center of RAS Vol. 27 No. 1 2025 content of sentences. Because the solution to the problem of comparative analysis of text information by content, a new algorithm for working with logical and linguistic models is proposed. The scientific novelty of the results obtained lies in the method of quickly extracting a set of local descriptors describing all parts of the image, which allows us to significantly speed up the annotation process and form a more complete global visual descriptor of the image.

Keywords: methods, automatic processing, grouping, annotation, information, Data Mining, resolution method

UDC: 004.912

MSC: 68U15

Received: 25.12.2024
Revised: 28.01.2025
Accepted: 03.02.2025

DOI: 10.35330/1991-6639-2025-27-1-111-119

Bibliographic databases:

•