A Study of Tweet Veracity to Separate Rumours from Counter-Rumours
Chua, Alton Yeow Kuan
Date of Issue2017
Proceedings of the 8th International Conference on Social Media & Society (SMSociety17)
Wee Kim Wee School of Communication and Information
Rumours are known to propagate easily through computer-mediated communication channels such as Twitter. Their outbreak is often followed by the spread of 'counter-rumours', which are messages that debunk rumours. The probability of a tweet to be a counter-rumour is referred to as 'tweet veracity' in this paper. Since both rumours and counter-rumours are expected to contain claims of truth, the two might not be easily distinguishable. If Internet users fail to separate rumours from counter-rumours, the latter will not serve its purpose. Hence, this paper investigates the extent to which tweet veracity could be predicted by content as well as contributors' profile. The investigation focuses on the death hoax case of Singapore's first Prime Minister Lee Kuan Yew on Twitter. A total of 1,000 tweets (500 rumours + 500 counter-rumours) are analyzed using binomial logistic regression. Results indicate that tweet veracity could be predicted by clarity, proper nouns, visual cues, references to credible sources, as well as contributors' duration of membership, and number of followers. The significance of these findings are highlighted.
© 2017 Association for Computing Machinery (ACM). This is the author created version of a work that has been peer reviewed and accepted for publication by Proceedings of the 8th International Conference on Social Media & Society, Association for Computing Machinery. It incorporates referee’s comments but changes resulting from the publishing process, such as copyediting, structural formatting, may not be reflected in this document. The published version is available at: [http://dx.doi.org/10.1145/3097286.3097290].