CyberEmotions

CyberEmotions
Collective Emotions in Cyberspace
Keywords	emotion, internet, data set, software, physics, social networks, sentiment analysis
Funding Agency	European Union
Project Type	Collaborative Project (Large-Scale Integrating Project)
Reference	231323
Objective	Understanding the role of collective emotions in creating, forming and breaking-up eCommunities
Participants	Janusz Holyst (coordinator), Warsaw University of Technology (Poland), École Polytechnique Fédérale de Lausanne (Switzerland), University of Wolverhampton (United Kingdom), Austrian Research Institute for Artificial Intelligence (Austria), ETH Zurich (Switzerland), Jožef Stefan Institute (Slovenia), Jacobs University Bremen (Germany), Berlin Institute of Technology (Germany), Gemius SA (Poland)
Budget	Total: €4,640,000 Funding: €3,600,000
Duration	1 February 2009 - 31 January 2013
Web Site	http://www.cyberemotions.eu/

CyberEmotions (Collective Emotions in Cyberspace) is a large-scale integrating project funded by the European Commission under the Seventh Framework Programme (FP7) in FET ICT domain theme 3: ‘Science of complex systems for socially intelligent ICT’.^[1] It started in February 2009 for a period of four years, and gathers approximately 40 scientists from Austria, Germany, Poland, Slovenia, Switzerland, and United Kingdom. Its main objective concerns understanding the role of collective emotions in creating, forming and breaking-up eCommunities.^[2][3][4]

So far, research performed within the project has helped uncovering some of the mechanisms driving eCommunities, such as negative emotions.^[5][6] It has led to the creation of several sentiment analysis computer programs, such as SentiStrength.^[7] Additionally, data gathered from online communities –such as the BBC message boards– have been made available to any interested researcher.^[8]

Objectives^[9]

• To understand the role of collective emotions in creating, forming and breaking-up ICT mediated communities as a spontaneous emergent behaviour occurring in complex techno-social networks.
• To understand the relationship between emotions of individuals as revealed by subjective experience, behaviour, physiological responses, and expressions with online emotional behaviours of ICT mediated dyads and groups in an integrative multi-level approach.
• To create decentralized adaptive tools which allow the amplification of positive or the suppression of negative collective emotions in e-societies and take into account the heterogeneity of interacting humans.
• To prepare the theoretical background for the development of the next generation emotionally-intelligent ICT services using models of self-organized active agents and sociophysics methods.

Consortium

The CyberEmotions consortium is composed of approximately 40 scientists in the area of physics, engineering, computing, artificial intelligence, virtual reality, and psychology.^[10] The additional business partner, Gemius SA, is an online research agency dealing with Internet market research.

• Centre of Excellence for Complex Systems Research, Warsaw University of Technology (Poland)
• Virtual Reality Lab, École Polytechnique Fédérale de Lausanne (Switzerland)
• Statistical Cybermetrics Research Group, University of Wolverhampton (United Kingdom)
• Austrian Research Institute for Artificial Intelligence (Austria)
• Chair of Systems Design, ETH Zurich (Switzerland)
• Department of Theoretical Physics, Jožef Stefan Institute (Slovenia)
• Emotion, Cognition, Social Context, Jacobs University Bremen (Germany)
• IKM Research Group, Berlin Institute of Technology (Germany)
• Gemius SA (Poland)

Organization

The project is divided into three layers: "data collection", "theory", and "ICT output".^[11] The first layer consists of research teams collecting data at a macro and micro level. For instance, eCommunities such as MySpace are being crawled and text exchanges between members automatically analyzed for positive or negative emotion. At a micro level, data is gathered on human psychophysiological responses to computer-mediated communication (CMC).^[12] The theory layer deals with building mathematical models of online emotional behavior, thanks to the data collected. Finally, the ICT output layer produces computer programs which integrate the models and data, in order to study CMC. Artificial dialog systems are for instance developed, which take into account emotional responses of the user, and respond accordingly (i.e., adapting their artificial affective state).^[13][14]

Research Results

During the first two years of the project, CyberEmotions led to the discovery of underlying emotional processes in eCommunities.^[15] For instance, it was shown that popular events (such as the Oscars) typically generate a majority of negatively valenced tweets,^[16] or that female users of MySpace send and receive proportionately more positive comments than males.^[17] The project fostered the development of several sentiment mining software programmes capable of retrieving the emotional valence and arousal in text produced by web users.^[7][18] Several physics and network models were developed that help understand the dynamic behind variations in valence and arousal of messages posted in eCommunities.^[19][20][21] Finally, software was designed that is able to respond in an emotionally-consistent way to messages produced by users varying in emotional content.^[22][23]

See also

• Affective computing
• Cyberspace
• Emotion
• Group emotion
• Sentiment analysis

References

1. ^ CORDIS. "FP7: FET Proactive Intiative: Science of Complex Systems for Socially Intelligent ICT (COSI-ICT)", European Commission. Retrieved on 2011-06-01.
2. ^ KOL. "Prof. Hołyst - coordinator of European research into emotions in cyberspace", Polish Ministry of Science and Higher Education, 2010-06-08. Retrieved on 2011-07-15.
3. ^ Petautschnig, F. "Emotionen im Internet", Ö1, aired on 2011-04-15.
4. ^ Cho, A. "Ourselves and Our Interactions: The Ultimate Physics Problem?", Science. 2009.
5. ^ Condliffe, J. "Flaming drives online social networks", NewScientist, 2010-12-07. Retrieved on 2010-12-13.
6. ^ Sosnowska, J. "Polscy naukowcy odkrywają prawdę o sieci: im dłuższa dyskusja, tym większa agresja", Gazeta Wyborcza, 2010-12-03. Retrieved on 2011-07-22.
7. ^ ^{a b} Thelwall, M., Buckley, K., Paltoglou, G. Cai, D., & Kappas, A. "Sentiment strength detection in short informal text", Journal of the American Society for Information Science and Technology. 2010.
8. ^ Kentucky FC. "The 70 Online Databases that Define Our Planet", The Physics arXiv Blog, 2010-12-03. Retrieved on 2011-01-20.
9. ^ CyberEmotions. "CyberEmotions objectives", CyberEmotions consortium. Retrieved on 2011-06-01.
10. ^ CyberEmotions. "CyberEmotions partners", CyberEmotions consortium. Retrieved on 2011-06-01.
11. ^ Holyst, J. A. et al. "CyberEmotions: Collective Emotions in Cyber-Space", Procedia Computer Science. 2011.
12. ^ Beck, K. "Science beyond fiction: Jacobs University participates in EU CyberEmotions project", Informationsdienst Wissenschaft. 2008.
13. ^ Clinicum. "Gefühle im virtuellen Raum verstehen", Clinicum. 2009.
14. ^ Skowron, M., Pirker, H., Rank, S., Paltoglou, G., Ahn, J., & Gobron, S. "No Peanuts! Affective Cues for the Virtual Bartender", Proceedings of the Twenty-Fourth International FLAIRS Conference. 2011.
15. ^ Holyst, J. A. et al. "CyberEmotions: Collective Emotions in Cyber-Space", The European Future Technologies Conference and Exhibition. 2011.
16. ^ Thelwall, M., Buckley, K., & Paltoglou, G. "Sentiment in Twitter events", Journal of the American Society for Information Science and Technology. 2010.
17. ^ Thelwall, M., Wilkinson, D., & Uppal, S. "Data mining emotion in social network communication: Gender differences in MySpace", Journal of the American Society for Information Science and Technology. 2010.
18. ^ Paltoglou G., & Thelwall M. "Twitter, MySpace, Digg: Unsupervised sentiment analysis in social media", ACM Transactions on Intelligent Systems and Technology. 2010.
19. ^ Chmiel, A., Sobkowicz, P., Sienkiewicz, J., Paltoglou, G., Buckley, K., Thelwall, M., & Holyst, J. A. "Negative emotions boost user activity at BBC forum", Physica A: Statistical Mechanics and its Applications. 2011.
20. ^ Mitrovic, M., & Tadic, B. "Bloggers behavior and emergent communities in Blog space", The European Physical Journal B - Condensed Matter and Complex Systems. 2010.
21. ^ Schweitzer, F., & Garcia, D. "An agent-based model of collective emotions in online communities", The European Physical Journal B - Condensed Matter and Complex Systems. 2010.
22. ^ Skowron, M. "Affect Listeners. Acquisition of Affective States by means of Conversational Systems", Lecture Notes in Computer Science. 2010.
23. ^ Gobron, S., Ahn, J., Paltoglou, G., Thelwall, M., & Thalmann, D. "From sentence to emotion: a real-time three-dimensional graphics metaphor of emotions extracted from text", The Visual Computer. 2010.

External links

• CyberEmotions official website
• CyberEmotions project description, on the European Commission website
• SentiStrength, sentiment mining software developed under the CyberEmotions framework.
• Commetrix, dynamic network visualization and analysis software applied for visualizing sentiment dissemination in networks for the CyberEmotions framework.