Authors:
Ema Kušen
1
and
Mark Strembeck
2
Affiliations:
1
Vienna University of Economics and Business, Austria
;
2
Vienna University of Economics and Business, Secure Business Austria (SBA) and Complexity Science Hub (CSH), Austria
Keyword(s):
Emotion Analysis, Riots, Social Network, User Behavior.
Related
Ontology
Subjects/Areas/Topics:
Cloud Computing
;
Collaboration and e-Services
;
Data Communication Networking
;
e-Business
;
Enterprise Information Systems
;
Information Systems
;
Methodologies and Technologies
;
Operational Research
;
Platforms and Applications
;
Social Networks
;
Telecommunications
Abstract:
In this paper, we present a study on the impact of emotions on information diffusion during a riot event. In
particular, we analyze a data-set consisting of more than 750 thousand social media messages related to the
2017 G20 summit that have been extracted from Facebook, Twitter, and YouTube. Because of the controversies
surrounding police operations during violent protests, our analysis especially focuses on emotions conveyed
in messages related to the local police. We found that a) negative emotions of high arousal (anger and fear)
dominate in messages mentioning the police on all three social networks, b) emotional content was forwarded
(retweeted) more often, regardless of the corresponding emotion valence, and c) in contrast to previous studies
we found that emotions have a considerably larger impact on the retweeting behavior than the number of
hashtags a message contains.