Return to page

USE CASE

Application / Account Opening Fraud

Objective

  • More accurately score applicants for fraud to drive down manual reviews and increase auto-acceptance rates

  • Reduce false positive rates to ensure only applications that require manual reviews are sent to fraud analysts 

  • Improve the customer experience during digital onboarding

  • Extend fraud prevention coverage earlier in the account lifecycle

Outcome

  • Understand customer behavior and KYC to better recognize illegitimate customers and create identity risk scores

  • Determine fraud risk before a customer’s application is accepted or declined

  • Use application (document or digital based) data and monitor all payment channels to dtect complex fraud emanating from stole and synthetic ID, as well as mule activity, with high accuracy

Business Value

  • Keep fraudsters out while increasing the number of customers banks can onboard, and thus revenue

H2O's AI and Data Approaches

  • Document classification to identify ghost invoices from companies that do not exist whose identity a fraudster assumes; identify legitimate invoices from forged ones

  • Logistic regression to determine if the result of reviewing customer data is fraud or not fraud

  • Decision Trees

  • Random Forest

  • Neural Networks 

Resources

Get Started with H2O.ai for Free

Start Your 90-day Free Trial Today

Get free, hands-on experience with the H2O AI Cloud. The H2O AI Cloud is an end-to-end platform that enables organizations to rapidly build world-class AI models and applications in the cloud or on premises.

Request an H2O AI Cloud Demo

We’re here to help you get started with H2O AI Cloud. Our demos will walk you through the capabilities of the platform and AI applications. We will help you determine how H2O AI Cloud can solve your organization’s specific challenges.