Elements of artificial intelligence in solving problems of text analysis

Due to the ever-increasing volume of textual information on the Internet and the need to navigate it, the automation of the text analysis process has become urgent. The analysis of the subject area has shown a great demand for the identification of textual information coloring and the application of works on this problem in practice. This paper deals with the development of a neural network model for analyzing commentary tone. Recurrent neural network models with long short-term memory modules (LSTM) are used for the purpose. We have developed an information system that determines the tone of comments on posts in the communities of the social network “VKontakte”. As a result of training of the artificial neural network, the model showed good accuracy in determining the tone of the text. The information system was implemented in the marketing department of the Nizhnevartovsk Construction College Budget Institution.