Collective Emotions in Cyberspace

The CyberEmotions consortium began in February 2009 and ran for a period of four years. The project focused on the role of collective emotions in creating, forming and breaking-up e-communities.

The project involved nine partners in six different countries in Europe, including experts in the psychology of emotions, complexity, web data collection, artificial intelligence and virtual reality.

YouTube video summary; Final project brochure.

In its final review on 10 July 2013 by independent experts appointed by the EU, CyberEmotions was rated as EXCELLENT. The Consortium was encouraged to submit a follow-up project related to the CyberEmotions research. A presentation of selected results is available here.

CyberEmotions 2013 conference

The Final Conference of the EU FP7 CyberEmotions Project was held in January 2013 and discussed the collective dynamics of socio-techno networks and affective interactions in such systems. The project will formally be completed in July, 2013. Here are some talks from the conference.

1. Collective emotions in Cyberspace, short review of Cyberemotions Project results, Janusz Hołyst
2. The Psychology of (Cyber)Emotions, Arvid Kappas
3. The social sharing of emotion, Bernard Rimé
4. How Emotional Are Users' Needs? Emotion in Query Logs, Marina Santini       
5. Social web sentiment strength detection: methods and issues, Mike Thelwall
6. Dynamics of emotions in voice during real-life arguments, Magdalena Igras 
7. Application of semantic spaces to sentiment analysis for words, Marcin Tatjewski
8. Online Networks and the Diffusion of Protests, Yamir Moreno
9. The impact of cyberspace upon current society, Aaron Ben-Ze'ev
10. Online discussions modelled by an evolving Ising-like dynamics, Julian Sienkiewicz
11. A modelling framework for collective emotions in online communities, David Garcia 
12. Patterns of Online Chats with Emotional Bots: Data Analysis and Agent-Based Simulations, Vladimir Gligorijević
13. Transition due to preferential cluster growth of collective emotions in online communities, Anna Chmiel        
14. Psychological Aspects of Social Communities, Renaud Lambiotte
15. The Simmel effect and babies' names, Krzysztof Kułakowski
16. A new model of individual opinion dynamics based on information and emotions, Paweł Sobkowicz
17. Human behavior in online social networks, Andrzej Grabowski
18. Facial asymmetry and affective communication in 3D Online Virtual Society, Junghyun Ahn


The consortium has produced many publications as well as software for dialog systems and representing emotion in 3D virtual worlds. One of our programs, SentiStrength, is sold commercially, is used in several countries and has been translated into a variety of different langauges. Project partners are also always interested in commercial collaboration with those interested in exploiting our technology.

As a sample, here is a still of a virtual reality experiment with people interacting with an intelligent, sentiment-aware bartender. The 3D agents expressed emotions through body movements and facial expressions as they interacted with the humans.

Bartender and Lydia

And below is a video demonstration of network visualization software with sentiment filtering and dynamic network evolution. Download HD (55Mb) or SD (40Mb) QuickTime versions.



CYBEREMOTIONS is a research domain that studies observable and analyzable phenomena related to any means of communication provided by the Internet - such as text, sound, visual, or any combination of these - that are related to emotional processes in individuals or groups. Transcending the classical analysis of human emotions, researchers in this domain also deal with emergent properties of the interplay between individual human emotions, the technical infrastructure of the Internet, and communication processes linked to all spheres of life. Therefore, research in the area of CYBEREMOTIONS is closely linked with human emotional processes in general (biological, behavioral, and experiential), with Internet-mediated communication (partially erasing space and time boundaries, and with creating specific networks of interactions).

A particular emphasis lies on the automatic analysis of online messages using methodologies such as sentiment analysis to provide access to emotional cues in large samples. The precise relationships between individual and collective emotions on the Internet are not yet known and likely require complex systems and network approaches.

The EU funded CYBEREMOTIONS consortium was created to better understand collective emotional phenomena in cyberspace, with the help of knowledge and methods from natural, social, and engineering sciences.

What is the difference between emotions and sentiments?

Emotions refer to bodily and psychological states in individuals. The causes of emotions are often social and the communication of emotion is an important facet of much human communication. Messages, verbal or nonverbal, do not, strictly speaking, contain emotions, but they are influenced, consciously and outside of awareness, by the emotional state of the sender, they may be sent to elicit or modify emotions in others, and they may be interpreted by receivers as emotional or create emotions in receivers. A research domain has developed that is referred to as "sentiment analysis" - to avoid confusion the analysis of emotion in messages refers to sentiments. Sentiment in this sense, refers to emotion-related aspects of messages. The use of sentiments as being synonymous with feelings in everyday and in scientific language is considered outdated.

Please contact the project leader Professor Janusz Holyst jholyst [at] with any enquiries.

FET tre


The consortium passed its second annual review after presenting its preliminary findings to a panel of experts in Brussels - See brief summary, extended summary

and slides for more info.

See also published results.

Collective emotions in cyberspace
OFAIWarsaw University of TechnologyÉcole Polytechnique Fédérale De LausanneJacobs UniversityStatistical Cybermetrics Research Group
GemiusInstitut Jozef StefanETHikm research