Research of Neural Network Methods of Text Information Classification

Automatic categorization of texts according to their content is an urgent task in the conditions of constant increase in the volume of text information. In this case, the assessment of subjectivity or objectivity of the text in relation to the subject is called the task of analyzing the tonality of the text and is fundamental in the analysis of opinions. The paper analyzes both theoretical and practical aspects of text classification methods, as well as proposes new approaches to text classification using recurrent neural networks, allowing to analyze the tonality of reviews.