(Guest post from Dr Anna Kolliakou, who gave a guest seminar in DDH a few weeks ago. Anna and Robert would be very interested in collaborating with anyone in DH who has interests in their project.)
Computing Veracity in Social Media
From a business and government point of view there is an increasing need to interpret and act upon information from large-volume media, such as Twitter, Facebook, and newswire. However, knowledge gathered from online sources and social media comes with a major caveat – it cannot always be trusted. Pheme will investigate models and algorithms for automatic extraction and verification of four kinds of rumours (uncertain information or speculation, disputed information or controversy, misinformation and disinformation) and their textual expressions.
Veracity intelligence is an inherently multi-disciplinary problem, which can only be addressed successfully by bringing together currently disjoint research on language technologies, web science, social network analysis, and information visualisation. Therefore, we are seeking to develop cross-disciplinary social semantic methods for veracity intelligence, drawing on the strengths of these four disciplines. The Department of Digital Humanities, an international leader for the application of technology in social sciences, was the appropriate platform for researchers from the SLAM Biomedical Research Centre at KCL, one of PHEME’s partners, to present their proposed work in veracity intelligence for mental healthcare with an aim to develop academic collaborations with academics interested in social media analysis, NLP and text mining. For more information…
Seminar: June 2, 2014: Robert Stewart and Anna Kolliakou
Social media poses three major computational challenges, dubbed by Gartner the 3Vs of big data: volume, velocity, and variety. PHEME will focus on a fourth crucial but hitherto largely unstudied, big data challenge: veracity. The relationship between clinicians and their patients has already been changed by the internet in three waves. First, the provision of pharmaceutical data, diagnostic information and advice from drug companies and health care providers created a new source for self-directed diagnosis. Secondly, co-creation sites like Wikipedia and patient support forums (e.g. PatientsLikeMe) have more recently added a discursive element to the didactic material of the first wave. Thirdly, the social media revolution has acted as an accelerant and magnifier to the second wave.
Prof Robert Stewart and Dr Anna Kolliakou, from the SLAM Biomedical Research Centre at King’s College London, have started the process of re-tooling medical information systems to compete with this new context. This will facilitate practical applications in the healthcare domain, to enable clinicians, public health professionals and health policy makers to analyse high-volume, high-variety, and high-velocity internet content for emerging medically-related patterns, rumours, and other health-related issues. This analysis may in turn be used (i) to develop educational materials for patients and the public, by addressing concerns and misconceptions and (ii) to link to analysis of the electronic health records.
In this seminar, they will be discussing the development of 4 main demonstration studies that aim to:
- Identify social media preferences and dislikes about certain medication and treatment options and how these present in clinical records
- Monitor the emergence of novel psychoactive substances in social media and identify if and how promptly they appear in clinical records
- Explore how mental health stigma arises in social media and presents in clinical records
- Ascertain the type of influence social media might have on young people at risk of self-harm or suicide