Content uploaded by Shafiza Mohd Shariff
Author content
All content in this area was uploaded by Shafiza Mohd Shariff on Jan 19, 2016
Content may be subject to copyright.
User Perception of Information Credibility
of
News on Twitter
(Shaza Mohd Shari, Xiuzhen Zhang, and Mark Sanderson)
Methods
Problem
Statement
Tweet
Features
Topic Keyword Link in Tweet
Display Name User Belief
Credibility Keyword Hashtag
Retweet User Men*on
Association
Rules
Hypothesis
Testing Conclusions
•Through analysing user
credibility judgements
and comments, 8
features have been
identi!ed, where
display name, link in
tweet and user belief in
the tweet topic are
most important.
•By feature and
credibility association
analysis, there are
strong associations
between features and
tweet credibility.
•Credibility judgments
are also dependent
with tweet’s news type
and single tweet
features user’s use to
perceive the credibility
of news tweets, proven
by hypothesis testing.
•However, the tweet
features commented by
the users for credibility
judgment is not news
type dependent.
•Dataset:
o400 tweets (20 news topics
with 20 tweets each)
oDivided into 3 news types -
breaking news, politic news
and disaster news
oNews events occurred
between 1 June and 15
October 2013.
•Credibility judgment on crowd
source platform - 98 evaluators.
•Evaluators’ comments on their
judgments analysed.