United StatesFinance and insurance2022 – ongoingValley Bank
Machine learning for anti-money-laundering alert prioritization
Valley Bank reduced monthly AML alert volume by approximately 22% with DataRobot
Problem
Valley Bank's compliance team spent weeks building predictive models manually. Large volumes of unproductive alerts consumed investigators' time.
Approach
The bank worked with DataRobot to build and validate models connected to its existing AML case management system. After testing, it put a model into production to defer alerts unlikely to become investigations. It retained a sample of those alerts for ongoing quality checks.
Results
Takeaway
Reducing the investigation queue depended on continued testing of alerts excluded by the model.
Categories
- Industry
- Finance and insurance
- Business function
- Finance, risk and legal
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Off-the-shelf tool or platform
- Evidence
- Vendor case study
Sources
Source summaries
DataRobot described production deployment after model validation. It reported lower monthly alert volume and a higher proportion of alerts becoming cases.
Full details are in the original sources.