PROBABILISTIC CALCULATION OF MISSING DATA VALUES OF THE SOCIAL NETWORK USERS: PROBABILISTIC ESTIMATE OF THE VALUES OF THE MISSING DATA OF THE SOCIAL NETWORK USERS

Abstract

Social networks are a unique source of data about the personal life and interests of real people. This opens up unprecedented opportunities for solving research and business problems (many of which could not be effectively solved earlier due to lack of data). In addition, this causes increased interest in the collection and analysis of social data from companies and research centers. However, many of them are hidden or not always correct. Therefore, before proceeding with the analysis of the data, it is necessary to carry out their adjustment, normalization, and propose a probabilistic estimate of the values of the missing data. This article explores the problem of predicting incomplete data using the method of group accounting of arguments (MGUA). To solve this problem, a model of the relationship between the studied data taken from the profiles of the social network VKontakte was found. The resulting model determines the relationship of a user's subscription to one interesting page of the network and user subscriptions to a group of other interesting pages, allows to predict the user's interest in a certain topic underlying the selected interesting page, depending on which interesting pages are already subscribed to at a certain moment