ASSESSMENT OF THE NETWORK COMMUNITY SUBJECTIVITY BY INDICATORS OF CONTENT AUTOMATIC RELATIONAL-SITUATIONAL ANALYSIS

Abstract

Aim. Presentation of regression models of the subjectivity of network communities based on automatically determined indicators of the content relational situational analysis (RSA). Methodology. To develop these models 64 network communities of various thematic focus from the open segment of social networks (Facebook, VKontakte, Odnoklassniki, Pikabu, Telegramm, etc.) were analyzed. The networks communities texts were subjected to psycholinguistic analysis using a previously developed list of discourse markers, and the results allowed to identify indicators of subjectivity. Automatic relational situational analysis of texts was performed using an RSA machine developed at the Institute for System Analysis of the Russian Academy of Sciences. Results. Comprehensive regression models of satisfactory quality were constructed for all indicators of subjectivity. Research implications. The use of the obtained regression models will allow to monitor various sectors of the Runet in an automated mode and to assess he subjectivity of the content.

Keywords

• internet
• network community subjectivity
• discourse markers
• text mining
• digital trails
• relational-situational analysis of text