Graph Theory and its Application to Fraud, Waste and Abuse

Digital and Data Services Overview

The application of graph theory is a powerful methodology to uncover hidden relationships in your data for rapid identification for operational improvement. Graph theory is essentially the study of graphs rooted in decades long mathematical theory. In graph theory, pairwise relationships among nodes. Think of nodes as a variable with a value property (physician, patient) and the relation as defined name or value properties (prescribed to, transaction). Graph theory is highly customizable to define nodes and relations to dig deeper into your data for hidden patterns.

Industry and Functional Applicability

The adoption of graph theory can be utilized by any industry: healthcare fraud/waste/abuse, entertainment social media and network analysis, transportation route finding, supply chain logistics improvement, financial fraud and wealth management.

ROI Optimization

There are an increasing number of tools and open source software with graph theory applications. They key is understanding how to customize these solutions with your data to define proper connections in your data to quickly identify trends or hidden relationships. A well defined business problem that matches to your data is an important first step. Then, it is important to understand how to enrich your data to be read by these graph theory applications for customized visualizations that are true positive relationships. Lastly, tying the insights back to the business context for pull through and implementation is what truly makes graph theory a powerful methodology.

The DownloAAd is a Blog Series featuring the latest tips, trends and thoughts from the experts leading AArete’s Digital & Data Services practice. To learn more about AArete’s Digital & Data Services capabilities, click here. To learn more about the minds behind the blog, read more and connect with the author below.

About the Author

About the Author

Mike Kim is Vice President Data & Analytics and Head of AArete’s Center of Data Excellence (CODE). Mike’s over 15 years of analytics experience includes technical delivery of machine learning/predictive models, implementation of advanced database designs, custom application development, and thought leadership for various analytic strategies. Representing healthcare, entertainment, professional sports, and numerous other industries, Mike co-leads AArete’s Digital & Data Services solutions and international expansion efforts. He holds a PhD in Health Policy and Management from Johns Hopkins University, a Masters of Public Health degree and a Bachelor of Arts degree in Biology from Brown University. Mike is a published author in Chain Drug Review, American Pharmacists Association, Mass Market Retailers, Retail TouchPoints, Advertising Age, Retail IT Insights, Managed Healthcare Executive, Property Casualty 360 and Financial Executives.