This article is a seminal piece and a foundational resource in the field of social media analytics and open source intelligence by some of the field’s leading authors.
McCreadie, Macdonald and Ounis discuss the use of social media platforms, such as Twitter, to track significant events. Emergency response agencies are increasingly looking to social media as a source of real-time information about such events. However, false information and rumours are often spread during these events, which can influence public opinion and limit the usefulness of social media for emergency management. In this paper, the authors present an initial study into rumour identification during emergencies using crowdsourcing. Through an analysis of three tweet datasets relating to emergency events from 2014, they propose a taxonomy of tweets relating to rumours. They then perform a crowdsourced labeling experiment to determine whether crowd assessors can identify rumour-related tweets and where such labeling can fail. Their results show that overall, agreement over the tweet labels produced were high, indicating that crowd-based rumour labeling is possible. However, not all tweets are of equal difficulty to assess, with many being cryptic or even impossible to understand.
This research will be of interest to crowdsource initiatives that rely on Twitter, Facebook, or other social media outlets. As the researchers have detailed, such an attempt requires vigilance and an ability to detect rumours and their impact. Much work remains to be done in this area, as the authors contend that tweets containing controversial information were subject to higher levels of disagreement by assessors. Meanwhile, tweets containing misinformation were not properly identified.