Computing Veracity – the Fourth Challenge of Big Data


Pheme work on computing veracity wins Best Paper prize

Work on measuring how true a document is from the words it uses has won the best paper prize at a major workshop, W-NUT. This is the workshop on noisy, user-generated text, hosted in 2016 at the conference for Computational Linguistics (COLING) in Osaka, Japan. In this research, AI methods for determining how certain one could be about the […]

New PHEME tool tracks coverage of online media coverage of mental health issues

King’s College London, in collaboration with MODUL University Vienna and webLyzard technology, have recently launched a new online tool, which tracks how media coverage of selected mental health concerns changes over time. The tool provides service users, mental health professionals and the general public with the opportunity to rapidly identify how mental health concerns are being discussed about in […]

Exploring the association between legal highs in social media and electronic health records

Collaboration between King’s College London, the South London and Maudsley NHS Trust and the University of Sheffield has produced a recently published paper in European Psychiatry, titled ‘Novel psychoactive substances: an investigation of temporal trends in social media and electronic health records’. The paper explores the relationship between the emergence of legal highs in social media and their appearance […]

Ontotext Work on PHEME Has Reached the Top 4 in the EC Competition for Innovation Radar Prize 2016

The work of Ontotext on automatic rumor detection and veracity prediction has caught the attention of the European Commission and has been nominated among the top 16 innovators in Europe. After a public vote, Ontotext is in the top 4 of the Industry & Enabling Tech category. As a result, on the 26th of September, […]

Combining content and temporal information for rumour stance classification

Work resulting from collaboration between the University of Sheffield, the University of Melbourne and the University of Warwick on classification of stance expressed in tweets towards rumours has been recently accepted for publication and presentation at the Annual Meeting of the Association for Computational Linguistics (ACL 2016). This work makes progress on the state of the […]

Pheme partners talk social media rumours in podcast

Journalism partner SWI recently sat down with some of the researchers involved in the project to ask a simple question: what is Pheme and how will it work? Ok, that’s two questions, with distinct answers that reflect the potential impact of Pheme in its two use cases: digital journalism and health care. Anna Kolliakou […]

Using fuzzy logic for an improved feature selection from web pages

Web pages are generally rich in structure, which are written using HTML markup that defines the layout that is visualised to visitors. However, this HTML markup has rarely been exploited for selecting meaningful features that characterise web pages. In a new paper, accepted for publication in the IEEE Transactions on Fuzzy Systems journal, we have introduced […]

RumourEval: open evaluation of rumour systems

We’ve had a task accepted at the long-running SemEval workshop, on rumour evaluation. This is a shared task, meaning that anyone can enter their rumour detection systems. Tasks at previous SemEvals have really helped advance the state of the art. For our task, which’ll be held next year, we’ll use Pheme data and methods to supply a big pile […]

Social media analysis of activity around Brexit

The most vocal twitterers advocate for the UK to leave the EU in the upcoming referendum vote. Laura Tolosi Pheme partner Ontotext has conducted an analysis of recent tweets and provides findings: More Twitter Users Want to Split with EU and Support #Brexit  

PHEME rumour dataset: support, certainty and evidentiality

Along with our recent publication in PLoS ONE, we have released a dataset of social media rumours. This dataset can be found on figshare. The dataset includes rumour tweets, collected and annotated within the journalism use case of the project. These rumours are associated with 9 different breaking news. It contains Twitter conversations which are […]

Previous Posts Next posts