This article discusses the solution of the problem of monitoring contact information in a social network and determining the personality psychotype based on it. Most of the material proposed by the developers works with English-language texts to determine the emotional coloring, their classification and, to a lesser extent, clustering. The approach proposed in the article is based on the use of the K-means method to solve the clustering problem of Russian-language texts and subsequent comparison with the results of working with the same data of an already trained model. During the work, was developed an effective method of parsing for collecting data. A proprietary data set has been collected and its processing and detailed analysis have been carried out. The result of the study allows us to make a primary assumption about what kind of psychotype a particular person has. It is quite difficult to determine the accuracy of the results of the task, since there is no predefined correct answer in machine learning without a teacher, and determining the psychotype is a difficult task from the point of view of psychology. All program code is written in the Python programming language.
Keywords: clustering, psychotype, social network, sanguine, phlegmatic, choleric, melancholic