The bank said the FCC intends to use Silent Eight as part of its On Screening Optimization initiative. The initiative applies machine learning and natural language techniques to improve the name screening process where customer and related party names are matched against various watch lists using fuzzy logic.
The solution will review each risk event using both private and public domain data, to provide true or false match recommendations. These recommendations will be supported by a written narrative explaining in plain English the decisions or recommendation. The machine will continually learn over time, constantly updating its algorithms to further improve the quality of its recommendations.
According to Enterprise Innovation, the Screening Optimization engine aims to replicate analyst actions during assessment, by automating the research and enriching the cases. The tool should also help improve case quality through a comprehensive review of all the attributes factored into each decision.
Every day we send out a free e-mail with the most important headlines of the last 24 hours.
Subscribe now