Featured 5912385701_470d2b97ac_z

Published on January 2nd, 2015 | by EJC

0

Using Social Media To Anticipate Human Mobility And Resilience During Disasters

This article was written by Patrick Meier and originally published at iRevolution on 1 December, 2014. Republished with permission.

The analysis of cell phone data can already be used to predict mobility patterns after major natural disasters. Now, a new peer-reviewed scientific study suggests that travel patterns may also be predictable using tweets generated following large disasters. In “Quantifying Human Mobility Perturbation and Resilience in Hurricane Sandy,” co-authors Qi Wang and John Taylor analyze some 700,000 geo-tagged tweets posted by ~53,000 individuals as they moved around over the course of 12 days. Results of the analysis confirm that “Sandy did impact the mobility patterns of individuals in New York City,” but this “perturbation was surprisingly brief and the mobility patterns encouragingly resilient. This resilience occurred even in the large-scale absence of mobility infrastructure.”

screen-shot-2014-11-30-at-5-05-14-pm

In sum, this new study suggests that “Human mobility appears to possess an inherent resilience—even in perturbed states—such that movement deviations, in aggregate, follow predictable patterns in hurricanes. Therefore, it may be possible to use human mobility data collected in steady states to predict perturbation states during extreme events and, as a result, develop strategies to improve evacuation effectiveness & speed critical disaster response to minimize loss of life and human suffering.”

Authors Wang and Taylor are now turning their attention to “10 other storms and typhoons that they’ve collected data on.” They hope to further demonstrate that quantifying mobility patterns before and after disasters will eventually help cities “predict mobility in the face of a future disaster, and thereby protect and serve residents better.” They also want to “understand where the ‘upper limit’ of resilience lies. ‘After Haiyan,’—the deadliest-ever Philippine Typhoon that struck last November—’there was a total breakdown in mobility patterns,’ says Taylor.”

Of course, Twitter data comes with well-known limitations such as demographic bias, for example. This explains why said data must be interpreted carefully and why the results simply augment rather than replace the analysis of traditional data sources used for damage after needs assessments after disasters.

About the Author:

Patrick Meier (PhD) is an internationally recognized thought leader on the application of new technologies for crisis early warning, humanitarian response and resilience. Presently serves as Director of Social Innovation at the Qatar Foundation’s Computing Research Institute. Previously co-directed Harvard’s Program on Crisis Mapping & Early Warning and served as Director of Crisis Mapping at Ushahidi. Patrick holds a PhD from The Fletcher School, a Pre-Doctoral Fellowship from Stanford & MA from Columbia. He was born & raised in Africa. Follow on Twitter: @patrickmeier

Photo: Eric Fischer

Tags: , , , , , , , , , , , , , , , , , , , , , , , ,


About the Author



Comments are closed.

Back to Top ↑