DataRobot Case Studies Valley Bank Reduces Anti-Money Laundering False Positive Alerts by 22%

Edit This Case Study Record

	Valley Bank Reduces Anti-Money Laundering False Positive Alerts by 22% DataRobot

Valley Bank Reduces Anti-Money Laundering False Positive Alerts by 22%

DataRobot

Technology Category	Analytics & Modeling - Predictive Analytics Platform as a Service (PaaS) - Data Management Platforms
Applicable Industries	Finance & Insurance
Applicable Functions	Business Operation
Use Cases	Fraud Detection Predictive Quality Analytics
Services	Data Science Services Cloud Planning, Design & Implementation Services
Challenge	Valley Bank, a regional bank with approximately $50 billion in assets, was facing a challenge in its Anti-Money Laundering (AML) department. The bank was dealing with an overwhelming volume of false positives in its effort to uncover money laundering activities across millions of transactions. The bank's AML team was seeking to reduce the manual work involved in predictive modeling. The process of creating models manually was time-consuming, taking weeks to complete. The bank was looking for a solution that could automate its fraud detection process and manage the volume of false positives in a realistic way. Read More
About Customer	Valley National Bank is the principal subsidiary of Valley National Bancorp. It is a regional bank with approximately $50 billion in assets. The bank operates many convenient branch locations and commercial banking offices across New Jersey, New York, Florida, Alabama, California, and Illinois. Valley is committed to providing the most convenient service, the latest innovations, and an experienced and knowledgeable team dedicated to meeting customer needs. The bank is also committed to helping communities grow and prosper, which is at the heart of Valley’s corporate citizenship philosophy. Read More
Solution	Valley Bank turned to DataRobot AI Cloud to automate its fraud detection process. The AI Cloud optimizes the entire AI lifecycle, helping to manage the volume of false positives without the need for data scientists on staff. The bank's model risk management team was involved from the start to validate the results. The bank fed predictions into its AML case management system and, alongside DataRobot data scientists, built and validated more than 100 models with a robust backtesting strategy. The bank also generated 175 features via DataRobot's automated Feature Discovery. During the trial, the bank reduced false positives by more than 30 percent. The bank's model risk management team was able to recreate those models successfully in the platform. Read More Log in to view content
Contents

Technology Category

Analytics & Modeling - Predictive Analytics

Platform as a Service (PaaS) - Data Management Platforms

Applicable Industries

Finance & Insurance

Applicable Functions

Business Operation

Use Cases

Fraud Detection

Predictive Quality Analytics

Services

Data Science Services

Cloud Planning, Design & Implementation Services

Challenge

Valley Bank, a regional bank with approximately $50 billion in assets, was facing a challenge in its Anti-Money Laundering (AML) department. The bank was dealing with an overwhelming volume of false positives in its effort to uncover money laundering activities across millions of transactions. The bank's AML team was seeking to reduce the manual work involved in predictive modeling. The process of creating models manually was time-consuming, taking weeks to complete. The bank was looking for a solution that could automate its fraud detection process and manage the volume of false positives in a realistic way.

About Customer

Valley National Bank is the principal subsidiary of Valley National Bancorp. It is a regional bank with approximately $50 billion in assets. The bank operates many convenient branch locations and commercial banking offices across New Jersey, New York, Florida, Alabama, California, and Illinois. Valley is committed to providing the most convenient service, the latest innovations, and an experienced and knowledgeable team dedicated to meeting customer needs. The bank is also committed to helping communities grow and prosper, which is at the heart of Valley’s corporate citizenship philosophy.

Solution

Valley Bank turned to DataRobot AI Cloud to automate its fraud detection process. The AI Cloud optimizes the entire AI lifecycle, helping to manage the volume of false positives without the need for data scientists on staff. The bank's model risk management team was involved from the start to validate the results. The bank fed predictions into its AML case management system and, alongside DataRobot data scientists, built and validated more than 100 models with a robust backtesting strategy. The bank also generated 175 features via DataRobot's automated Feature Discovery. During the trial, the bank reduced false positives by more than 30 percent. The bank's model risk management team was able to recreate those models successfully in the platform.

Impact #1	Valley Bank was able to create new models or retrain models in a day, compared to weeks.
Impact #2	The bank was able to operate productively even when their only data scientist left the role, demonstrating the platform's ease of use.
Impact #3	The bank was able to save approximately 22 percent in total alert volume every month.

Impact #1

Valley Bank was able to create new models or retrain models in a day, compared to weeks.

Impact #2

The bank was able to operate productively even when their only data scientist left the role, demonstrating the platform's ease of use.

Impact #3

The bank was able to save approximately 22 percent in total alert volume every month.

Benefit #1	Reduced false positives by more than 30 percent during the trial.
Benefit #2	Saved approximately 22 percent in total alert volume every month.
Benefit #3	Increased the number of alerts escalating to case by three percentage points.

Benefit #1

Reduced false positives by more than 30 percent during the trial.

Benefit #2

Saved approximately 22 percent in total alert volume every month.

Benefit #3

Increased the number of alerts escalating to case by three percentage points.

Download PDF Version

Overview

Valley Bank Reduces Anti-Money Laundering False Positive Alerts by 22%

Operational Impact

Quantitative Benefit