Algorithmisation of the Means of Representing the “New Knowledge” Frame in Popular Science Discourse

The paper outlines research identifying key elements of the ‘new knowledge’ frame in popular science Telegram channels. Highlighting categories that can be used to describe this frame, authors propose principles for developing an algorithm to search for it. The article demonstrates that the terminals’ automatic recognition of the frame ‘new knowledge’ is accomplished by developing a multistage search algorithm that involves describing programmed units located within the different-level linguistic means, creating a qualitative sample, and drawing conclusions about the inclusion of the category in the algorithm’s requirements. The paper employs both quantitative and qualitative methods to identify trends in various types of media to make some basis for further development of a supervised ML system. An analysis of 288 units from popular science Telegram channels reveals that the location and frequency of framing techniques in texts indicate the quality and level of audience preparation. The results of the study have demonstrated that the term ‘new knowledge’ is explicated via the set of semantics components that form the following microfi elds: a) The course of study; b) The result of study; c) The subject of research work; d) Comparison between new knowledge and old knowledge; e) Novelty of knowledge; f) Denial of previous knowledge. The aim of the frame description is to develop an algorithm that can identify units within this frame in real speech materials of social media. This will aid future research on objection’s stylistics and enable the development of AI systems that can be trained to recognize objection styles. Keywords: framing, computational linguistics, popular science discourse, automatic text analysis, objection