Dynamical model of conversation of users of online news media

Abstract

The paper studies discussions in the online news media. The dynamic processof increasing the number of comments on an article published on a news site is considered. The proposed model distinguishes between comments on the text of the article itself (root comments), and comments on comments (discussion). To test the model, two empirical cases are considered, which are sets of user comments on two political articles published in The Guardian. Empirical time series have been constructed for the number of comments of both types; these series were juxtaposed with the results of the calculation based on the model. For each of the two cases, it was found that the model dynamics for each of the variables basically correspond to the empirical ones. Consequently, the present study demonstrates the qualitative concordance between the proposed model and empirical observations. Moreover, our findings reveal that the dynamics of discourse are influenced not solely by the inherent interest evoked by the article, but also by the degree to which user engagement is propelled by the comments themselves. In other words, readers do not just comment on the article but also form a discussion.