A day after two bombs exploded near the finish line of the Boston Marathon, killing three and injuring more than 170 people, an Instagram user named “amasonicvibes” posted a photo from the sideline on race day. The photo shows crowds of spectators gathered along Boylston Street, watching the stream of runners on the course. Flags from various nations add color to scene. The photo is stamped by date, latitude and longitude, with the description “Minutes before the attacks, #Life can #change so fast. #Grateful to be ok…”
Any online user can sift through Google or social media sites with search term keywords in mind, whether they’re scanning consumer views about a big box retailer or seeking signs of activity at the site of a tragedy. What’s notable about this photo is that it’s one of scores of images and messages to surface in a location-based search of the Copley Square neighborhood in Boston on April 16 using a service called Geofeedia.
Started two years ago, Geofeedia is a Web-based analytics offering that distills geolocation data from smartphones and social media networks to create a location-based structure around the unstructured data—images, text, and video— of user activities on Twitter, Instagram, Flickr, YouTube and Picasa.
The service is designed with business users in mind: Think of retailer or consumer goods brand managers scouting consumer sentiment about their wares in the places where and when shoppers are active. But Phil Harris, Geofeedia’s CEO said that since its launch, the Naples, Fla.-based company has seen growing interest in the service from public safety officials and the media.
Harris said he could not comment specifically on work Geofeedia is doing with investigators in Boston, but said that the bombing incident represents “a classic example” of rising interest in the service. “A byproduct of our work is that public safety officials want the structured data, to know what is happening at a certain place and a certain time. That is a perfect application of our dataset, and we’re helping lots of law enforcement agencies.”
That application of public safety and operations management is what drew the interest of Esri, the geographic information systems and mapping application maker, too. Earlier this month, Esri announced it would add Geofeedia social media analytics tools to its ArcGIS platform for mapping and geospatial analysis.
“Geofeedia has a very eloquent capability to manage social media in a way that you can begin to make it actionable in an organized way, which is not easy to do,” said Russ Johnson, the director of public safety at Esri. “They also have the ability to segment that [data] geographically, based upon a large or small area of interest, and they have methodology and some tools; that enable you to look at the source of each one of the social media inputs.”
This kind of capability is potentially valuable to investigators in an incident such as the Boston Marathon bombing because it can add data “that could provide intelligence and chronology for this type of an attack,” Johnson added. Who said what, where and when, with images added—all of these are potential clues that were not readily accessible without the Geofeedia presentation, he said.
Geofeedia has built an infrastructure to support petabytes of social media data and is working with a number of vertical industry partners on different use cases, Harris said. Consumer marketers are a key focus, while the hospitality industry represents another opportunity to engage customers where they tweet. He said the company will work on making deeper archives available so users can delve back in time.
The service is dataset agnostic, he said, and can change its lineup to add new networks that emerge as popular choices among users.
Michael Goldberg is editor of Data Informed. Email him at firstname.lastname@example.org.