The final speaker in this Social Media and Society session is Moses Boudourides, who presents a study of the affective publics on Twitter surrounding the European refugee crisis and the Paris terrorist attacks. The project tracked some twenty keywords and hashtags relating to the refugee crisis, capturing a substantial volume of tweets that were further processed using Python.
The volume of activity was affected by a range of external events, from the Paris terror attacks through the New Year's Eve sexual assaults reported in Cologne to the Eidomeni camp crisis and the Brussels bombings – but not all of these generated similar spikes in Twitter activity. Some of them also spawned their own, event-specific hashtags that separated from the more generic, ongoing discussion of the refugee crisis.
Further network analysis of these data is also possible: networks of hashtags, retweets, @replies, and @mentions all provide useful additional insights into the data. Python libraries providing sentiment analysis can also help to explore correlations between sentiment assessment and the likelihood of retweeting.