Using Twitter for big data analytics to analyze disasters

Posted Aug 18, 2017 by Tim Sandle
There are 500 million tweets sent every single day. Can some of this data be put to good use, such as aiding communities in responding to a disaster? Researchers think so.
The application of Twitter data and its subsequent big data analysis to aid with disaster response has been raised by computer technologists from the Penn State College of Information Sciences and Technology. Understanding how a community is responding in the wake of disaster, such as a natural event like an earthquake or following a terrorist incident, is important for business planning and for governments to coordinate appropriate responses. While organizations can measure a community's ability to respond to a disaster or to assess the impacts in the aftermath, assessing such things has never really been possible in real time due a to a dearth of data.
However, the high volume of tweets sent via Twitter allow real-time information to be examined, such as through the use of intelligent computers to scan for key words or to group information via appropriate hashtags or at symbols (@digitaljournal). There are a lot of tweets sent: Every second, on average, around 6,000 tweets are tweeted on Twitter, which corresponds to over 350,000 tweets sent per minute, 500 million tweets per day and around 200 billion tweets per year.
The researchers have developed a case study titled "Embracing human noise as a resilience indicator" published in the journal Sustainable and Resilient Infrastructure. The study indicates the ability of social media to alert first responders. For this, the technologists compared tweets sent out during Hurricane Sandy (the deadliest and most destructive hurricane of the 2012 Atlantic hurricane season). They also looked at the corresponding power outage information relating to New York, New Jersey, and Pennsylvania. It was found that by comparing the information provided from the power grid and juxtaposing this with human Twitter chatter (some 10 million relevant posts), a verifiable methods for event detection was created. The relevant tweets selected contained such words as: "power," "outage," "electri," and "utility."
Based on the conclusion that Twitter can report events more quickly, lead researcher Nick Lalone states in a research note: "The goal of this research is to demonstrate that if the data stream changes, you can see what just happened. It would result in a real-time monitoring system." Examples include, Lalone adds: "We can count how many beds are in a hospital, or how many people can be injured," added Lalone. "But we know people are producing all of this noise and we know it can be useful. But we've never been able to correlate or demonstrate it in real-time."
Lalone expects that his model, of harnessing the power of social media to support communities, will be used by businesses and government on a larger scale to aid disaster recovery.