СОЗДАНИЕ АВТОМАТИЗИРОВАННОГО МЕХАНИЗМА МОНИТОРИНГА ВЛИЯНИЯ ПОЛИТИЧЕСКИХ ИНТЕРНЕТ-МЕМОВ НА РУССКОЯЗЫЧНЫХ ПОЛЬЗОВАТЕЛЕЙ ИНТЕРНЕТА

Abstract

The study of Internet memes is undergoing significant transformations: from traditional approaches (semiotic, structural-functional) to innovative (network, object-oriented). In the Internet environment, memes have become the most important information carrier, which makes it possible to maintain multivariate communication. The elementary and at the same time capable of reformatting the structure of the meme, its high virulence led to the emergence of new communication strategies and the representation of current events. Memes act as the optimal material through the analysis of which all the specificity of virtual communications is revealed. New communications require the involvement of new methods of studying them with the use of automated and intelligent analysis of data received on a flow basis. Internet memes are a set of unstructured data, the study of which can lead to the identification of hidden variables and the determination of patterns of distribution of memes within the Internet and social networks in particular. This is especially true of political memes, the appearance of which can be regarded as a reactive or predictive indicator that allows you to monitor the dynamics of the political situation or predict it. The article provides information on the creation of the «Memometrics» software package, which allows automated search and collection of memes with their subsequent analysis. Special attention should be paid to the use of neural networks for offline data processing on the basis of self-learning. A base of memes, combined with a formal listing of reactions to a meme, can be a preparatory stage in the process of training a neural network.