News
Watch and learn (more) about Pheme
Pheme’s use case partner SWI swissinfo.ch has created a short video to explain its role in the project and how the Pheme dashboard currently being developed will help journalists to verify user-generated content on social media. swissinfo.ch spent several months tracking news stories online, analysing tweets for rumours, and checking if claims turned out to […]
Ontological Modelling of Rumors
In this paper at the RUMOUR 2015 workshop, we present on-going work pursued in the context of the Pheme project. There, the detection of rumors in social media is playing a central role in two use cases. In order to be able to store and to query for information on specific types of rumors that can […]
The language of mental health problems in social media
Online social media, including sites such as Reddit, has become an important platform to share and discuss personal experiences. On Reddit, there are forums dedicated to specific topics, so-called subreddits, where users can interact with each other. Many of these subreddits are about mental health-related issues, such as anxiety, depression, suicide watch, and bipolarity. This […]
PHEME RTE dataset
For the special purpose of Natural Language Processing-based information verification, we have built a new Recognizing Textual Entailment (RTE) resource from Twitter data. The PHEME RTE dataset is compiled based on naturally occurring contradiction in manually labeled claims in tweets related to crisis events, and to our knowledge is the first resource for 3-way judgement […]
Summer Training Course: Mining social media content with GATE
The 9th GATE training course will be held from 6-10 June 2016 at the University of Sheffield, UK. Early registration deadline is 1 May 2016. GATE is the General Architecture for Text Engineering, a powerful tool for text processing and analytics, for both programmers and non-programmers. This event will follow a similar format to that of […]
On Twitter, it takes two hours to resolve a true rumor and 14 hours to debunk a false one
Rumors that are ultimately proven true tend to be resolved faster than those that turn out to be false. So are the results of a a recent paper published in PLoS One, from Pheme participants at the University of Warwick and Swissinfo. False stories take much longer on average to be debunked because counter-evidence takes longer […]
PHEME D3.2 to be presented at NewsIR’16
A short paper on D3.2 Algorithms for Implicit Information Diffusion Networks Across Media will be presented at the First International Workshop on Recent Trends in News Information Retrieval (NewsIR’16) that takes place in Padua, Italy on 20th March 2016 in conjunction with ECIR 2016. Visualising the Propagation of News on the Web Svitlana Vakulenko, Max […]
Workshop on Extraction and Processing of Rich Semantics from Medical Texts
PHEME will support and help organise a workshop on rich semantics for medical text. This helps automate extraction in one of our use-cases, the medical domain. See more: http://www.dfki.de/richmedsem/
New PLOS ONE paper looking at conversations around rumours in social media
A new paper in collaboration between the University of Warwick and swissinfo.ch has now been published in the PLOS ONE journal, titled ‘Analysing How People Orient to and Spread Rumours in Social Media by Looking at Conversational Threads‘. This paper looks at conversations that are initiated by rumourous statements on Twitter during breaking news, as […]
Social Media in the Newsroom workshop at ICWSM-16
PHEME is co-organising a workshop on ‘Social Media in the Newsroom’, co-located with ICWSM-16, and will take place on May 17, 2016 in Cologne, Germany. The workshop aims to focus on the intersection of social media and journalism, as a subset of Computational and Data Journalism. It will particularly focus on the development of novel algorithms, methods, […]