Streamlining grants management with machine learning
- By Sara Friedman
- Jun 28, 2018
As a Center of Excellence at the Department of Health and Human Services, GrantSolutions.gov works with 10 federal agencies to improve the grants management process and is now looking to incorporate predictive analytics and machine learning into the processes.
“For agencies that are looking to leverage predictive analytics and machine learning, you have to be willing to share information, and you will get better results over time,” Julius Chang, program director of the System and Support Division at GrantSolutions.gov, said at a June 27 Government Executive event. “At GrantSolutions, we are doing a pilot on machine learning and data analytics that can yield predictive analytics capabilities that will improve the workflow and results through our systems.”
Machine learning can be used to determine which applications are likely to be accepted or rejected, Chang said. With a machine learning program that reviews applications when they come in, grants officers could spend more time on “good applications,” and the “likely rejected applications” could be handled by a lower level specialist.
The machine learning system can predict with 96 percent accuracy whether applications will be rejected because they are poorly written, Chang said. However, its ability to identify applications that will be approved is still under development.
Another initiative underway at GrantSolutions involves creating risk ratings for recipient organizations.
“We want to produce risk ratings for organizations that receive funding by looking at their history, the outcomes that they’ve been able to produce for their specific projects and the ability to hold them accountable for the funds that they’re expending,” Chang told GCN. ”We want to be able to say that an organization has had problems in the past, and they can improve their standing by fixing their processes.”
GrantSolutions is also testing a natural language processing program that reviews the progress and status reports that grant recipients submit for problems before grant officers start their evaluations.
“Sometimes these reports are several hundred pages long, and a grants officer is responsible for a thousand of them,” Chang said. If reports were summarized and problematic ones flagged, grants officers would know which ones “they need to review with a fine-tooth comb.”
The goal around these initiatives is making it easier for grants officers to do their jobs efficiently, said Chang.
With predictive analytics, grants officers "can be just as effective at administering those 1,000 grants as they were a few years back when they only had 200 in their portfolio,” he said. “It is about engaging with our customers on a case-by-case basis, and we are piloting solutions to improve processes.”
Sara Friedman is a reporter/producer for GCN, covering cloud, cybersecurity and a wide range of other public-sector IT topics.
Before joining GCN, Friedman was a reporter for Gambling Compliance, where she covered state issues related to casinos, lotteries and fantasy sports. She has also written for Communications Daily and Washington Internet Daily on state telecom and cloud computing. Friedman is a graduate of Ithaca College, where she studied journalism, politics and international communications.
Friedman can be contacted at firstname.lastname@example.org or follow her on Twitter @SaraEFriedman.
Click here for previous articles by Friedman.