Subscribe to Tendercast, Data Science Statistics, Data Preparation, Machine Learning with Python courses to receive complimentary 4-hour 1:1 use case support

Financial Crimes Enforcement Network (FinCEN) guidance regarding customer due diligence and the recent beneficial ownership information reporting requirement enable the opportunity to extend machine learning analytics to the know your customer (KYC) program.

Continued augmentation and refinement of KYC due diligence programs with machine learning tools to report risk procedures, profiles and ongoing monitoring is required.  Adding the beneficial ownership information as a data attribute and creating a feature using feature importance statistical modeling is one of the first steps. 

The value of machine learning in KYC by quantifying the risk score using machine learning tools such as feature importance, gradient boosting, and random forest.  Categorizing these risk scores by customer and integrated into the KYC program will enable ongoing monitoring by using machine learning to update the algorithm regularly.  The ongoing data feed of internal data (i.e. transactions, products, relationships, etc.) and external data (i.e. sanctions, negative media screening, cross-institution / geography banking) into the KYC program enables a refinement of risk scores which can then identify transactions of concern to explore and monitor.