Integrated Discursive Approach to the Communicative Activity of Internet Users

The growing interest in the study of psychological characteristics of media content is due to the spread of the internet in modern life. In this regard, the study of communicative activity of users in the space of the internet as a new discursive phenomenon of digital society is of particular relevance. This task was accomplished by developing a methodology for the integrated study of discourse based on psychological, psycho-linguistic and mathematical methods. The aim of the article is to present the results of applying this approach to the analysis of the intentional component of communicative activity, to determine its possibilities, limitations and prospects. To identify the intentions of the users communicative activity, the method of intent analysis was used to describe and verify discursive markers which are indicators of intention realization, psycho-linguistic methods of lexicon-grammatical analysis. The author's experimental method of understanding of the utterances used to assess the potential of the discursive markers, the syntactic analyzer, genetic algorithm as based on DEAP, the neural network model BERT, which takes into account the parameters of emotional tone of discourse. 44 categories of intentions (4897 realizations) are revealed, among which negative intentions were the most frequent. Neutral analytic intentions were the most diverse in composition. The lexical, syntactic, graphic and punctuation markers of intentions of communicative activity were described, and their specificity for Internet discourse was established. The ability of discourse markers to make intention more explicit is confirmed empirically. At the same time, the variability / inequality of the potential of markers was revealed, which along with the cognitive proximity of certain categories can complicate their unambiguous identification. Prospects for taking into account the context of utterances, which significantly improves the recognition of intentions, with the use of artificial intelligence models and methods were identified. According to the results of the study, the integrated discursive approach helps to overcome the problems related to the implicitness of the psychological content of the utterances and the contextual variability of its verbalization, creating a basis for technologies to assess the dynamics of interaction on the internet, the impact and its social consequences.