Artificial Intelligence 3 MIN Read

Machine Learning Uncovering New Insights from Endless Data

October 12th, 2021

OUR CAPABILITIES

Learn more about GDIT’s analytics work and our approach.

The way federal agencies make sense of data is changing drastically. And it stands to reason, right? Agencies are creating more data than ever before, and there are more tools at their disposal than ever before to make sense of it. Deriving value from data can transform the way agencies deliver services and, ultimately, positively impacting the people they serve.

Today, artificial intelligence, machine learning, natural language processing, computer vision, intelligent automation and cognitive agents are being used to support missions in entirely new ways. The following examples represent is a new way of using analytics to solve hard problems.

  • Detecting fraudulent Medicare payments at CMS: Machine learning is identifying potentially fraudulent Medicare payments at the Centers for Medicare and Medicaid Services. In a single day, CMS receives millions of claims representing billions in payments. Our team leveraged machine learning alongside regression analysis, link analysis, outlier detection and behavioral analysis to identify risky providers, which has given CMS the ability to analyze large volumes of data that went previously unanalyzed.
  • Classifying medical images for the VA: Working in partnership with the Department of Veterans Affairs, a machine learning-based classifier was developed to help physicians spot malignant skin lesions. We used computer vision and built deep learning image classification models trained on more than 10,000 images of skin lesions. Once field tested, primary care providers may be able to use the tool to more rapidly and accurately diagnose these conditions, many of which may have gone undetected as care was delayed during the pandemic.

Mission Insights discovers value from data

Without question, analytics are evolving – and will continue to evolve – the way that agencies deliver services. They’re also evolving the way that agency support partners must show up for them. At GDIT, an intentional methodology was developed to get value from data called Mission Insights, and we’re seeing spectacular results.

Mission Insights is a cloud-native analytics platform that allows us to ingest multiple types of data, enrich it with a host of AI/ML services and present it to users in an interface that allows them to search it, study it, build models or create alerts about items that are of interest to them.

From there, with Mission Insights, users have access to a suite of services from AWS. This includes entity extraction, text indexing, sentiment analysis, language translation, image and video entity extraction, predictive analytics and more.

On the backend, the innovations in Mission Insights also leverage AWS services, such as continuous integration-continuous deployment (CI/CD), infrastructure as code, event-driven architectures, data pipelines, zero trust security and serverless state programming library support.

In partnership with customers, we’re using Mission Insights to discover high-value use cases and develop strategic analytics roadmaps. Recent roadmap exercises are already shaping analytics initiatives at CMS, FAA, the U.S. Environmental Protection Agency, the Department of Homeland Security, the U.S. Patent and Trademark Office and the Office of Personnel Management, to name a few.