Unfortunately, any time there is a crisis, there are those who would try to take advantage of the system in an illegal fashion. And during the COVID-19 pandemic, it was no different for the PPP loan program, where the US government attempted to help millions of businesses to stay open during one of the most significant disruptions in decades.
According to a late 2021 academic paper from researchers at UT Austin, up to 15% of PPP loans showed at least one indicator of potential fraud, representing about 1.8 million of the overall 11.8 million loans. One of the authors, Samuel Kruger, indicated to the New York Times that “it’s very difficult to put [anecdotes] together and get at the scale of what’s going on”, which is precisely why Redhorse chose to tackle this issue- to help provide a facts-based and data-driven solution for the problem.
The goal for Redhorse was to take existing domain expertise and IP, combining it with graph-database technologies and graph data science techniques, including powerful and cutting-edge machine learning on unstructured text, to leap-frog the inadequate technology options available today.
Vince Bridgeman, VP of National Security Services for Redhorse, put it this way: “The challenge for detecting the depth of PPP loan fraud simply surpasses the technical capabilities of fraud platforms available today. By integrating our own expertise and IP, together with graph database, with a Knowledge Graph and Graph Data Science – the ability for analysts and even the system itself to uncover fraud– in time to make a difference– is massively improved. When we combine that with the highly usable Domo BI/Analytics user experience, now we have an integrated platform that can make a true impact on recovering our tax dollars from illegal and abusive contexts.”
Having worked with the Graphable team on a variety of cutting-edge initiatives both internally at Redhorse and even on classified defense initiatives, Redhorse knew this was the right partnership. Redhorse Director of Data Science Jodi Deprizio remarked, “Having successfully partnered with Graphable over time and in a variety of contexts, it was clear that combining Redhorse’s domain expertise and custom IP with Graphable’s deep graph database, knowledge graph and Domo technical expertise was a great fit for the Accountable App initiative.”
The Redhorse and Graphable data science teams conducted a series of innovation sessions
to brainstorm the kinds of capabilities required, and more importantly the questions that fraud analysts would need to answer, in what ways and in what timeframes.
In this iterative and cooperative process, the architecture emerged, including the graph model, the user capabilities and interface as well as the analytics output required to support the outcomes and actions.
This process highlighted the unusually strong partnership, the depth of domain, graph, data science and AppDev expertise on both teams, as well as the power of integrating expertise with existing technologies to drive new capabilities to create a net new platform and app that can make a real impact for all of us.
Download Full Case Study
Graphable delivers insightful graph database (e.g. Neo4j consulting) / machine learning (ml) / natural language processing (nlp) projects as well as graph and Domo consulting for BI/analytics, with measurable impact. We are known for operating ethically, communicating well, and delivering on-time. With hundreds of successful projects across most industries, we thrive in the most challenging data integration and data science contexts, driving analytics success.
Want to find out more about our Hume consulting on the Hume knowledge graph / insights platform? As the Americas principal reseller, we are happy to connect and tell you more. Book a demo by contacting us here.
Check out our article, What is a Graph Database? for more info.