Paul Starrett, Private Investigator, Starrett Consulting, Inc.

Paul Starrett

Private Investigator, Starrett Consulting, Inc.

Leveraging Data Science Tools to Build Robust Graph Databases for Use in Fraud Examinations

As data volume and complexity become more troublesome, data-science tools become ever more vital to the anti-fraud community. To compound matters, unstructured data is the largest and most challenging area of “big-data.” Specific tools found in data science come to the rescue.

In this session, you will learn the synergy between open-source tools in information retrieval (e.g. search-engine technology), natural-language processing and graph databases. We will discuss how information retrieval can be used to remove unwanted information and to key in on relevant topics and events.

From there, natural-language processing tools can be used to extract named entities (e.g. people, places, associations, etc.), keywords, phrases and concepts. Graph databases can be built from this data to find important relationships and patterns. These three areas, leveraged in various permutations, can further refine results.

In an investigation, we are most interested in summarizing data to find what is going on, to look for useful correlations and relationships, and to identify red flags that require further review. Learn how these indispensable data-science tools provide powerful and efficient results to the anti-fraud community.


Paul Starrett is a licensed attorney and private investigator specializing in high-profile investigations, compliance consulting and legal counseling especially where electronic data is central. He is founder and CEO of Starrett Consulting, Inc., a full-service investigations and consulting firm where they leverage open-source and commercial data-science applications to analyze structured and unstructured data.

His 25-year career includes roles as general counsel and chief risk officer, information-security software engineer, information-management consultant and corporate-security executive.

Paul holds a Master of Science degree in Predictive Analytics from Northwestern University and a Master of Laws (LL.M.) in Taxation from Golden Gate University. He is also a Certified Fraud Examiner (CFE) and EnCase Certified Computer Forensics Examiner (EnCE). Paul is also an avid Python programmer and co-organizer of PyBay, the largest Python-based conference in Silicon Valley.