Patent Showcase: Find-More-Like-This Technology & Inventor Interview

“The D&B Commercial Graph^™, verified business identity, and proprietary insights provide a unique foundation for analyzing companies in the context of their broader business ecosystems. These ecosystem-level insights matter because understanding how a company is connected to its counterparties — and how those counterparties are connected to their own counterparties — can reveal a more nuanced view of credit exposure, supply chain vulnerability, commercial opportunity, and many other critical insights. 

The challenge is that ecosystem-level insights emerge only when hundreds or even thousands of constantly changing counterparty relationships are understood not as isolated data points, but as a whole — something even a large team of human analysts could not realistically accomplish manually. 

This is where the invention provides a powerful solution framework. Using advanced graph theory techniques, it locates anomalous ecosystem patterns and transforms overwhelming, otherwise unmanageable relationship structures into mathematical objects that can be quantified, compared, and searched across the D&B Commercial Graph.  

Detected anomalies can then be fingerprinted, capturing the mathematical essence of what makes them unusual, and stored as anomaly fingerprint libraries that can be used to identify similar anomalies in otherwise unrelated business ecosystems. 

This is yet another example of how we are combining the D&B Commercial Graph with advanced analytical technics to generate trusted, valuable insights for our customers.” 

-- Ilya Meyzin, Senior Vice President, Data Science at Dun & Bradstreet 

“D&B Commercial Graph contains an extraordinary depth of business relationships and signals. This invention helps translate that depth into a prioritized set of companies, clusters, or ecosystem patterns that merit attention. The capability helps surface where the most meaningful signals are likely to be — including similarities or anomalies that may not be apparent from company-level attributes alone. That is the practical value: clients can move more quickly from broad market or risk questions to a focused set of companies and patterns to investigate, monitor, or act on.” 

-- Yan Vtorov, Vice President, Data Science at Dun & Bradstreet 

“We created a set of purpose-built mathematical formulations to represent business ecosystems as graphs and make their structures measurable. This was not simply applying standard graph analytics out of the box. We developed new ways to quantify commercial relationship patterns — how entities connect, how those connections behave, where anomalies emerge, and when two ecosystems are meaningfully similar. That math is what makes the fingerprint possible: it turns complex business relationships into graph-based representations that can be compared, searched, and reused across the D&B Commercial Graph.” 

-- Seyed Mohammad Nikouei, Director, Data Science at Dun & Bradstreet 

“Part of the process was about removing some of the noise. Sometimes you’d notice a business was different from the norm, so to speak – you’d notice something special about it. This enabled us to recognize some of the same traits in other parties in other relationships. Then we were able to compare and determine which resembled each other.” 

-- Dave Spingarn, Director, Data Science at Dun & Bradstreet