Bevza M. Semantically-relevant search in text databases using neural networks

The relevance of the research topic is due to the fact that the current state of development of the information society is characterized by the digitalization of all spheres of life and, accordingly, the exponential growth of data presented in digital form. Characteristically, effective machine analysis today requires converting data into a format acceptable to software algorithms at the level of human-machine interaction, which increases the workload of operators and leads to errors due to the presence of the human factor. The basic task is to analyze textual data presented in natural language, which is expanded by analyzing related multimedia materials, working with speculative queries, and predicting the audience's reaction to blocks of textual data presented through the rating system, the presence of shares, and sentiment analysis of comments from service users. The organization of a system for machine analysis of large amounts of text data should combine the solution of the problem of reducing the query processing time with increasing the accuracy of data processing, which is proposed to be solved on the basis of semantically relevant search algorithms. The advantage of the procedure of semantically relevant search and determination of the contextual meaning of the input query is the possibility of building a system that combines the partial use of appropriate algorithms together with methods characterized by a minimum load on the computing resource. This makes it possible to obtain a correlation between the accuracy of the analysis and the processing time of the input data according to the task.