From Idea to App in Washington's Tech Land

Jeff Godin had a good job as a police dispatcher in Chester County, Penn., but quit to work full-time on other projects, including building an app for police and emergency professionals.

By
Mon, February 10, 2014

Computerworld — WASHINGTON -- Jeff Godin had a good job as a police dispatcher in Chester County, Penn., but quit to work full-time on other projects, including building an app for police and emergency professionals.

In the six months since his launch of his Badge Buddy app, he's sold about 1,000 for $5.99 apiece.

Ok, so Godin's not getting rich by selling an app - yet. But we're jumping a little ahead in telling his story.

Godin had an idea and acted on it. The app was good enough to be picked as a finalist for the American Council for Technology and Industry Advisory Council's (ACT-IAC) Igniting Innovation 2014 awards. The group provides a collaborative forum for the government and IT industry.

Similar to Godin, Jane Blake, a lead associate in bio surveillance at Booz Allen Hamilton, along with Catherine Ordun, an associate at this firm, had an idea for tracking disease using social media.

Their idea, Open Source Health Intelligence System (OSHINT), was selected in an internal innovation contest that Booz Allen conducts for its employees. From there, it was picked as a finalist for the Ignite Award.

Last week, Godin's Badge Buddy and Booz Allen's OSHINT were in Ronald Reagan Building here, among the 30 finalists to display their various technologies. The company size didn't matter in the selection process. What mattered to judges was the idea.

"People like to complain on social media when they don't feel well," said Blake, explaining why it's worth examining social media as a tool for alerting health professionals to a disease outbreak.

Blake can compare the data that they get in real time with Centers for Disease Control data. CDC data is collected from various sources, such as reports by doctors, but it could take as long as two weeks before an outbreak picture emerges.

The early social media detection system built by Blake and Ordun uses a natural language processing algorithm and numerical statistical analysis to identify outbreaks of illnesses.

To show its effectiveness, the researchers acquired social media information collected by an aggregator and then compared what they pulled from it with CDC reporting.

For real-time data, Blake and Ordun are using Twitter, which is making a small part of its feed available for such uses.

"If you can get closer to the truth in real time you will be doing very well," said Blake.

The benefit of using social media for outbreak analysis is to provide an early warning, and for targeting resources that much faster, said Blake.

Continue Reading

Originally published on www.computerworld.com. Click here to read the original story.
Our Commenting Policies