Tweets (user posts on Twitter's microblogging service) usually reveal information about the current life-state of their author. We track this information, and then investigate ways of mapping it to an accurate system for identifying and tracking epidemics of infectious diseases and in particular Influenza like Illness (ILI) in several UK regions. Flu Detector is a demonstration of this work, a tool which nowcasts flu rates based on tweets on a daily basis.
Follow us on Twitter (we will tweet when our inferences indicate we should!)

People Involved

Vasileios Lampos, PhD Student - Contact: Bill.Lampos <at> bristol.ac.uk
Nello Cristianini, Professor of Artificial Intelligence
Tijl De Bie, Lecturer in Artificial Intelligence

Related Publications

V. Lampos and N. Cristianini. Nowcasting Events from the Social Web with Statistical Learning. In: ACM Transactions on Intelligent Systems and Technology (TIST), Vol. 3(4), ACM, September 2011.
V. Lampos and N. Cristianini. Tracking the flu pandemic by monitoring the Social Web. In: CIP 2010, pp. 411-416, IEEE Press, June 2010.
V. Lampos, T. De Bie and N. Cristianini. Flu Detector - Tracking Epidemics on Twitter. In: ECML PKDD 2010, pp. 599-602, Springer, September 2010.

Media Coverage / Related Articles / Press Releases

March 20, 2012 - Costing the Earth: Outbreak BBC Radio 4
March 9, 2012 - Diagnosing flu symptoms with social media Natural Hazards Observer, Volume XXXVI, Number 4, pp. 7-9
December 23, 2011 - 2011: The year in review Latest News of EPSRC (Engineering and Physical Sciences Research Council)
November 2, 2011 - International media coverage for research into whether social media could be used to detect disease outbreaks Press Release by the University of Bristol
November 1, 2011 - Research suggests that social media could track flu outbreaks by The Engineer
November 1, 2011 - Could social media be used to detect disease outbreaks? Press Release by the University of Bristol
November 17, 2010 - Engines of the future: The cyber crystal ball by New Scientist
July 14, 2010 - How Twitter Could Better Predict Disease Outbreaks by MIT Technology Review
July 1, 2010 - Predicting Flu from the content of Twitter in News of Computer Science Department, University of Bristol

Acknowledgements

We are grateful to NOKIA Research Centre for the various levels of support on our work. Vasileios Lampos would also like to thank the Department of Computer Science (University of Bristol) for supporting him partially during his PhD studies. Finally, we would like to thank HPA and Twitter since they form the main information sources for this project.

Contact Information

Feel free to send an email to Bill.Lampos <at> bristol.ac.uk for any questions or possible ideas for further development of this work.


Affiliations: