Revolutionizing the Way AML Teams Look at PEPs: An Entity Resolution Case Study

Maria Singson, VP & GM Data Science, Mastech Infotrellis
Deepti Soni, Senior Data Scientist, Mastech Infotrellis

The last decade has brought significant advancement to digital communications, especially in social media. For money laundering, messages in cryptic and secret languages are sent across social media platforms tens of thousand times a day. Other digital footprints are now also abounding, with every like, follow and comment potentially having some meaning related to illicit funding and financial malfeasance. In this paper/talk, we discuss new ways for compliance teams to analyze and update their Politically Exposed Persons (PEP) lists, leveraging knowledge graphs and machine learning. The foundation of our knowledge graph leveraged our PEP ontology cards that present a different approach to contextualizing an individual or commercial entity so that their identification includes their digital footprint. We present examples of how we have helped financial institutions and procurement teams better identify potential financial crimes in their portfolios by enhancing their identity resolution methodologies with graph-powered machine learning and PEP list vigilance.

Download the session slides here.

linkedin-icon Connect with Maria on LinkedIn 

linkedin-icon Connect with Deepti on LinkedIn