Analysing How People Orient to and Spread Rumours in Social Media by Looking at Conversational Threads

By Arkaitz Zubiaga, Geraldine Wong Sak Hoi, Maria Liakata, Rob Procter, P. Tolmie. PLoS ONE, 2015

708 citations recorded by the source corpus.

As breaking news unfolds people increasingly rely on social media to stay abreast of the latest updates. The use of social media in such situations comes with the caveat that new information being released piecemeal may encourage rumours, many of which remain unverified long after their point of release. Little is known, however, about the dynamics of the life cycle of a social media rumour. In this paper we present a methodology that has enabled us to collect, identify and annotate a dataset of 330 rumour threads (4,842 tweets) associated with 9 newsworthy events. We analyse this dataset to understand how users spread, support, or deny rumours that are later proven true or false, by distinguishing two levels of status in a rumour life cycle i.e., before and after its veracity status is resolved. The identification of rumours associated with each event, as well as the tweet that resolved each rumour as true or false, was performed by journalist members of the research team who tracked the events in real time. Our study shows that rumours that are ultimately proven true tend to be resolved faster than those that turn out to be false. Whilst one can readily see users denying rumours once they have been debunked, users appear to be less capable of distinguishing true from false rumours when their veracity remains in question. In fact, we show that the prevalent tendency for users is to support every unverified rumour. We also analyse the role of different types of users, finding that highly reputable users such as news organisations endeavour to post well-grounded statements, which appear to be certain and accompanied by evidence. Nevertheless, these often prove to be unverified pieces of information that give rise to false rumours. Our study reinforces the need for developing robust machine learning techniques that can provide assistance in real time for assessing the veracity of rumours. The findings of our study provide useful insights for achieving this aim.

Subfields: Artificial Intelligence, Sociology and Political Science, Statistical and Nonlinear Physics

View Paper: https://www.semanticscholar.org/paper/e5bcf6cacfe36722d51c0ee7684abb84e3a0fe70

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Sociology and Political Science

In Statistical and Nonlinear Physics

Citations in the corpus

Just Say No: Analyzing the Stance of Neural Dialogue Generation in Offensive Contexts (2021, background): “…and 82.4% agreement for offensiveness in SBIC (Sap et al., 2020) 9Comparable to stance label pairwise agreement of 62.3% for rumor-stance dataset (Zubiaga et al., 2016) ever, we find that aggregating worker annotations produces gold labels of sufficiently high quality for training and…”
Predicting Factuality of Reporting and Bias of News Media Sources (2018, methodology): “…rumor detection have been primarily addressed using information extracted from social media, i.e., based on how users comment on the target claim (Canini et al., 2011; Castillo et al., 2011; Ma et al., 2015, 2016; Zubiaga et al., 2016; Ma et al., 2017; Dungs et al., 2018; Kochkina et al., 2018).”
Detection and Resolution of Rumours in Social Media (2017, methodology): “This is an approach that was used ﬁrst by [Zubiaga et al. 2016] and subsequently by [Giasemidis et al. 2016].”
SemEval-2017 Task 8: RumourEval: Determining rumour veracity and support for rumours (2017, methodology): “The data has already been annotated for veracity and SDQC following a published annotation scheme ( Zubiaga et al., 2016b), as part of the P HEME project (Derczynski and Bontcheva, 2014), in which the task organisers are partners.”
Fighting an Infodemic: COVID-19 Fake News Dataset (2020, methodology): “(Zubiaga et al., 2016) presented a methodology to identify and annotate a dataset having 330 rumour threads (4,842 tweets) associated with 9 newsworthy events.”
A Survey on Natural Language Processing for Fake News Detection (2018, background): “PHEME contains 330 twitter threads (a series of connected Tweets from one person) of nine newsworthy events, labeled as true or false according to thread structures and follow-follower relationships.”
COVID-19–Related Infodemic and Its Impact on Public Health: A Global Social Media Analysis (2020, background): “Rumor is defined as unverified information that canbe foundas true, fabricated, or entirely false after verification.(14) Stigma is a socially constructed process through which a person with stigma can experience discrimination and devaluation in society.”
DTCA: Decision Tree-based Co-Attention Networks for Explainable Claim Verification (2020, methodology): “Figure 4 depicts the results based on a speciﬁc sample in PHEME, where at the comment level, red arrows represent the captured evidence and grey arrows denote the unused comments; at the word level, darker shades indicate higher weights given to the corresponding words, representing higher attention”
Detect Rumors in Microblog Posts Using Propagation Structure via Kernel Learning (2017, background): “…inspired by prior analyses on rumors spreading, e.g., user information can be a strong clue in the initial broadcast, content features are important throughout entire propagation periods, and structural and temporal patterns help for longitudinal diffusion (Zubiaga et al., 2016; Kwon et al., 2017).”
SemEval-2019 Task 8: Fact Checking in Community Question Answering Forums (2019, methodology): “, there has been work on checking the factuality/credibility of a claim, of a news article, or of an information source (Ba et al., 2016; Zubiaga et al., 2016; Ma et al., 2016; Castillo et al., 2011; Baly et al., 2018).”
A survey on fake news and rumour detection techniques (2019, methodology): “The most promising efforts on data gathering have been made by [116], concerning social media posts and discussions during breaking news events, and”
An exploratory study of COVID-19 misinformation on Twitter (2020, positive): “This method was first employed by [37].”
Rumor Detection on Twitter with Tree-structured Recursive Neural Networks (2018, methodology): “…supervised classiﬁ-cation problem, which learns a classiﬁer f from labeled claims, that is f : C i → Y i , where Y i takes one of the four ﬁner-grained classes: non-rumor , false rumor , true rumor , and unveriﬁed rumor that are introduced in the literature (Ma et al., 2017; Zubiaga et al., 2016b).”
Combating Fake News (2019, background): “Analyzing How People Orient to and Spread Rumours in Social Media by Looking at Conversational Threads [133] (https://figshare.”
On Adversarial Examples for Character-Level Neural Machine Translation (2018, background): “Companies such as Google are using text classifiers to detect abusive language(1) , and concerns are increasing over deception (Zubiaga et al., 2016) and safety (Chancellor et al.”
False Information on Web and Social Media: A Survey (2018, background): “Since people tend to spread unverified claims [91, 121], making false information believable may not be as important as persistently spreading it.”