Big data for better crisis response

More people, including both emergency responders and disaster victims, are choosing social media as a way to respond to crises…and for good reason! Given its immediacy, reach, ability to receive instant feedback and host multimedia content, urgent information can be collaborated in real time.

screen-shot-2016-10-18-at-9-35-57-pm

Avengers assemble! Source: https://goo.gl/DBVXLQ

Contrary to popular belief, the value of social media is not only limited to during and after a crisis. Focus is now shifting towards forecasting (yes, just like a weather forecast!screen-shot-2016-10-18-at-9-43-32-pm) to better improve preparedness and cushion social and economic damage.

With the proliferation of unorganised data sets on social media and the net, Virgina Tech has seized the opportunity to use big data to predict trends and create knowledge to forecast the future of emergency response. The ‘Early model-based event recognition using surrogates’, or better known as EMBERS, provides analysis of open-source data to recognise significant societal events around the world. Since its conception, it has correctly predicted Paraguay’s presidential riots in 2012 and Chile and Argentina’s Hantavirus epidemic in 2013.

big-data-protests.jpg

EMBERS accurately predicted that violence and police brutality would spill over to multiple cities in the wake of student protests in Venezuela, 2014. Source: http://www.vtmag.vt.edu/winter15/big-data-at-Virginia-Tech.html

With a reported mean lead time of 7.54 days prior to the crisis event, the impact of big data carries undeniable practical implications. According to Naren Ramakrishnan from Virginia Tech’s Discovery Analytics Centre (DAC), governments are now better equipped to address citizens’ top priorities and understand what hot-ticket items are in civil unrest.

These applications of faster response time are not lost on businesses who are hard-pressed to avoid the high costs and risks of crisis. Researchers from Virginia Tech have developed an informatics system that allows automobile manufacturers to discover vehicle defects more efficiently through data sets on social media before the crisis breaks. As technology and social media matures, it won’t be long before more high-risk industries recognise the good business sense behind predictive analytics too.

 

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s