Information Propagation Cultures on Sina Weibo and Twitter
1. Information Propagation Cultures on
Sina Weibo and Twitter
Qi Gao1, Fabian Abel1, Geert-Jan Houben1, Yong Yu2
1 Web Information Systems, Delft University of Technology
2 APEX Lab, Shanghai Jiao Tong University
1
fraction of posts that
contain a hashtag
Q1: How quickly do users Q2: What are the syntactical 0.1
propagate information on Sina characteristics of messages that
Weibo and Twitter? people propagate on Sina Weibo 0.01
and Twitter?
time distance (in hours)
Twitter users repost messages Weibo 0.001 Weibo-all posts
1000 Weibo-reposts
faster than Sina Weibo Users. Twitter Users on Twitter seem to be more Twitter-all posts
100 triggered by hashtags and URLs 0 Twitter-reposts
10 when propagating information 0% 20% 40% 60% 80% 100%
users
Explanation: Twitter users may than on Sina Weibo.
1
1
have the impression that they
fraction of posts that
0.1
play an important role in the
contain a URL
Explanation: Twitter users seem 0.1
information propagation process, 0 to be more eager to let their
i.e. they act as if they are in the 0% 20% 40% 60%
users
80% 100%
posts appear in the public 0.01
power of spreading information discussion – possibly a higher Weibo-all posts
(power distance). demand to profile themselves 0.001 Weibo-reposts
(individualism). Twitter-all posts
0
Twitter-reposts
0% 20% 40% 60% 80% 100%
users
Temporal Analysis Syntactic Analysis
Cultural Models
Microblogging Platforms
Culture-aware User Modeling and
Analysis
data acquisition multilingual NER
Cultural Analytics
metadata extraction semantic enrichment
topic/interest modeling user profile construction
Semantic Web
Sentiment Analysis Semantic Analysis
Q3: What are the sentiment Q4: To what extent do the user
characteristics of messages that interests vary between Sina Weibo
people propagate on Sina Weibo and Twitter?
and Twitter?
fraction of positve reposts
100% Weibo
Twitter The topics and concepts users
Chinese microbloggers have a 80% discuss on Sina Weibo differ from
stronger tendency to propagate 60% more positive posts
Twitter. User interests are broader
positive messages. on Sina Weibo than on Twitter.
#distinct reposted users
more negative posts
40%
Weibo
ratio of #reposts to
1000 Twitter
20% Explanation: Chinese
Explanation: The positive nature
microblogging behavior follows a 100
of the propagated information on 0%
rather collectivistic culture -
Sina Weibo might point at the 0% 20% 40% 60%
users
80% 100%
actual content of a message 10
long term orientation that is
seems to be more important.
attributed to the Chinese culture. 1
0% 20% 40% 60% 80% 100%
users
[1] Qi Gao, Fabian Abel, Geert-Jan Houben, Yong Yu. Information Propagation Cultures on Sina Weibo and Twitter.
In Proceedings of ACM Web Science Conference 2012. Evanston, USA. June, 2012
[2] Qi Gao, Fabian Abel, Geert-Jan Houben, Yong Yu. A Comparative Study of Users’ Microblogging Behavior on Sina Weibo and Twitter.
In Proceedings of 20th International Conference on User Modeling, Adaptation, and Personalization. Montréal, Canada. July, 2012
Contact:
Qi Gao, Fabian Abel, Geert-Jan Houben, Yong Yu
http://wis.ewi.tudelft.nl - q.gao@tudelft.nl