Collective Emotions Online
This chapter analyzes patterns in messages posted to several Internet discussion forums from the perspective of the sentiment expressed in them and the collective character of observed emotions. A large set of records describing comments expressed in diverse cyber communities—blogs, forums, IRC channels, and the Digg community—was collected, and sentiment classifiers were used to estimate the emotional valence (positive, negative, or neutral) of each message. A comparison with simple models showed that the data included clusters of comments with the same emotional valence that were much longer than similar clusters created by a random process. This shows that there are emotional interactions between participants so that future posts tend to have the same valence as previous posts. Threads starting from a larger number of negative comments also last longer so negative emotions can be treated as a kind of discussion fuel; when the fuel (negativity) is used up in the discussion, it may finish. Moreover, the amount of user activity in a particular thread correlates positively with the presence of negative emotions expressed by the individual user in the thread. In summary, the analyses describe individual and collective patterns of emotional activities of Web forum users and suggest that negativity is needed to fuel important discussions.
This work was supported by a European Union grant by the 7th Framework Programme, Theme 3: Science of complex systems for socially intelligent ICT. It is part of the CyberEmotions (Collective Emotions in Cyberspace) project (contract 231323). J.A.H, A.Ch. and J.S. acknowledge support from Polish Ministry of Science Grant 1029/7.PR UE/2009/7.
- Barabási A-L (2005) The origin of bursts and heavy tails in human dynamics. Nature 207:435–433Google Scholar
- Frijda NH (1986) The emotions. Cambridge University Press, Cambridge, MAGoogle Scholar
- Macdonald C, Ounis I (2006) The TREC Blogs06 collection: creating and analyzing a blog test collection (Technical Report TR-2006-224). Department of Computer Science, University of Glasgow, GlasgowGoogle Scholar
- Paltoglou G, Thelwall M, Buckely K (2010) Online textual communication annotated with grades of emotion strength. In: Proceedings of the third international workshop on EMOTION (satellite of LREC): corpora for research on emotion and affect, Valletta, Malta, pp 25–31Google Scholar
- Riloff E, Patwardhan S, Wiebe J (2006) Feature subsumption for opinion analysis. In: Proceedings of the conference on empirical methods in natural language processing, Morristown, NJ, USA, pp 440–448Google Scholar
- Sabucedo JM, Durán M, Alzate M, Barreto I (2011) Emotions, ideology and collective political action. Univ Psychol 10:27–34Google Scholar
- Sienkiewicz J, Skowron M, Paltoglou G, Hołyst JA (2013) Entropy-growth-based model of emotionally charged dialogues. Adv Complex Syst 16:1350026Google Scholar
- Taylor V (1995) Watching for vibes: bringing emotions into the study of feminist organizations. In: Ferree MM, Martin PY (eds) Feminist organizations: harvest of the new women’s movement. Temple University Press, Philadelphia, pp 223–233Google Scholar
- Tumasjan A, Sprenger TO, Sandner PG, Welpe IM (2010) In: Proceedings of the fourth international AAAI conference on weblogs and social media. AAAI Press, Menlo Park, CA, pp 178–185Google Scholar
- Walther J, Parks M (2002) In: Knapp M, Daly J, Miller G (eds) The handbook of interpersonal communication. Sage, Thousand Oaks, CA, pp 529–563Google Scholar
- Weroński P, Sienkiewicz J, Paltoglou G, Buckley K, Thelwall M, Hołyst JA (2012) Emotional analysis of blogs and forums data. Acta Phys Pol A 121:B128–B132Google Scholar