This study undertakes a systematic analysis of media discourse on migration in Sweden from 2012 to 2019. Using a novel data set consisting of mainstream newspapers, Twitter and forum data, the study answers two questions: What do Swedish media actually talk about when they talk about “migration”? And how do they talk about it? Using a combination of computational text analysis tools, I analyze a shift in the media discourse seen as one of the outcomes of the European refugee crisis in 2015 and try to understand the role of social media in this process. The results of the study indicate that messages on social media generally had negative tonality and suggest that some of the media frames can be attributed to a migration-hostile discourse. At the same time, the analysis of framing and sentiment dynamics provides little evidence for the discourse shift and any long-term effects of the European refugee crisis on the Swedish media discourse. Rather, one can hypothesize that the role of the crisis should be viewed in a broader political and historical context.

This paper presents a study on the dynamics of sentiment polarisation in the active online discussion communities formed around a controversial topic—immigration. Using a collection of tweets in the Swedish language from 2012 to 2019, we track the development of the communities and their sentiment polarisation trajectories over time and in the context of an exogenous shock represented by the European refugee crisis in 2015. To achieve the goal of the study, we apply methods of network and sentiment analysis to map users’ interactions in the network communities and quantify users’ sentiment polarities. The results of the analysis give little evidence for users’ polarisation in the network and its communities, as well as suggest that the crisis had a limited effect on the polarisation dynamics on this social media platform. Yet, we notice a shift towards more negative tonality of users’ sentiments after the crisis and discuss possible explanations for the above-mentioned observations.

This work aims to study the social construction of migrant categories and immi-gration discourse on Swedish-speaking Facebook pages in the last decade. It combinesthe insights from computational linguistics and the distributional semantics approachwith those from classical sociological theories to explore a corpus of more than 1MFacebook posts. This allows one to compare the meanings of labels denoting variouscategories of migrants and identify the interpretative repertoires used by Facebookusers to discuss the immigration topic. The study concludes that, despite the expres-sions of tolerance and support for refugees and immigrants, the Facebook audience isnevertheless active in the objecti cation and discursive discrimination of those iden-ti ed as belonging to either of those discursive categories. The study results are thenrelated to the technological design of new media and the overall social and politicalclimate surrounding the Swedish immigration agenda.

In this paper, we present our methodology for supervised stance classification of sparse and imbalanced social media data. We test our framework on a manually labeled dataset of 5700 messages about immigration in the Swedish language posted on the Flashback forum, a controversial online discussion platform. Our proposed approach currently achieves a macro- averaged F1-score of 0.72 for test data on a two-class problem compared against 0.27 for a baseline four-class model. Since effective classification of imbalanced and sparse textual data in under-resourced languages presents certain methodological challenges, our study contributes to a discussion on the best pathways to achieve highest model performance given the character of the data and unavailability of large training datasets for this task. Moreover, this work exemplifies the application of ML methodology to social media data, which can be particularly relevant for social scientists working in this area and interested in leveraging the possibilities of machine learning in their research field. This methodology and the obtained results provide a foundation for further in-depth analyses of social media texts in the Swedish language following a data-driven approach.