Social Media for Emergency Response
In the SUPER project, funded by the European Commission’s 7th framework programme, the University of Glasgow have been tracking and analysing social media data for emergency response, including real-time:
targeted sentiment analysis – identifying what are users saying on social media about entities important to the event, e.g. political leaders
topic tracking – identifying what aspects of the crisis/emergency people are discussing over time
Summarisation – generating a timeline summarising how a given event is evolving
rumour detection – identifying discussion threads of misinformation surrounding events.
community detection – grouping active users into the categories, allowing a common operational picture (COP) to be tracked
The SUPER platform was recently trialled at the Civil Protection Service of the Campania Region (Italy) in a simulated flooding scenario. Another deployment of the system is due to be shortly held in Bucharest (Romania) in a simulated security crisis scenario.
Related Research & External Links
- Saul Vargas, Richard McCreadie, Craig Macdonald, Iadh Ounis: Comparing Overall and Targeted Sentiments in Social Media during Crises. In Proceedings of ICWSM 2016
- Richard McCreadie, Craig Macdonald, Iadh Ounis. EAIMS: Emergency Analysis Identification and Management System. In Proceedings of SIGIR 2016
- Richard McCreadie, Craig Macdonald, Iadh Ounis: Crowdsourced Rumour Identification During Emergencies. In Proceedings of RDSM2015 at WWW 2015
- Richard McCreadie, Craig Macdonald and Iadh Ounis. Incremental Update Summarization: Adaptive Sentence Selection based on Prevalence and Novelty. In Proceedings of CIKM 2014