Теоретические основы прогнозирования пользовательских настроений в социальных сетях

Abstract

Analyzed is the problem of predicting the mood of users in social networks. The properties of social network information objects are investigated, and a method for collecting and processing social data for forecasting is presented. The goal of this study is to prove or disprove the possibility of predicting user sentiment in social networks. A discrete model of the social network has been compiled. The theoretical foundations of forecasting are formulated based on the statistical characteristics of the discrete model, as well as algorithms for searching and systematizing information objects. A method is proposed and described that allows you to predict the reaction of users to a message published in a social network. A hypothesis about the possibility of predicting mood in social networks is formulated and proved. The possibility of forecasting user sentiment in social networks is proved. A new method that allows you to predict the reaction of users to a published news item is presented based on the principles of machine learning. The results of the work can be applied in the field of small and medium-sized businesses when directly searching for a target audience to increase the profitability of the enterprise. The development of sentiment analysis methods will allow planning measures to contain or speed up the spread of messages. Also, the application value in developing a forecasting model may lie in the development of news sources or identifying trends. The scientific significance of the work lies in the presentation of a method that allows predicting the reaction of social network users to the published news. The results of this work can be used to improve the performance of social networks and to solve the problem of “cold” start (what to show to new users).