Denmark’s Danske Bank Launches State-of-the-Art AI-Driven Fraud Detection Platform

Danske Bank, a leading financial services provider in the Nordics, has collaborated with Think Big Analytics, a Teradata company, to develop an advanced artificial intelligence (AI)-driven fraud detection platform. The innovative solution is expected to meet 100 percent return on investment (ROI) in its first year of production.

Combating Growing Banking Fraud

Application fraud is a critical issue for banks, and criminals are becoming increasingly sophisticated. To combat this growing threat, Danske Bank has developed an AI-powered engine that uses machine learning to analyze tens of thousands of latent features and score millions of online banking transactions in real-time.

Key Features

• Analyzes tens of thousands of latent features
• Scores millions of online banking transactions in real-time
• Provides actionable insights into true and false fraudulent activity

Reducing False Positives and Increasing Efficiency

Danske Bank’s original fraud detection system relied heavily on handcrafted rules that had been proactively applied by the business over time. However, this approach led to record numbers of false positives - at times reaching 99.5 percent of all transactions - resulting in significant costs and time associated with investigation.

The AI-powered engine has significantly reduced the cost of investigating false positives, increasing overall efficiency and poised for substantial savings.

Joint Effort between Danske Bank and Think Big Analytics

The Teradata Think Big Analytics team began working with Danske Bank in autumn 2016 to augment their advanced analytics team with specialist knowledge on utilizing data to bring greater benefits to the wider business. The joint team built a framework within the bank’s existing infrastructure and created advanced machine learning models to detect fraud within millions of transactions per year.

Benefits

• Reduced false positives by 50 percent
• Increased detection rate by around 60 percent
• Improved transparency through an interpretation layer

Future Plans

Danske Bank’s anti-fraud program is the first to put machine learning techniques into production while simultaneously developing deep learning models to test the techniques. The bank will share details about its AI-powered engine for fighting financial fraud at the annual Teradata PARTNERS Conference in Anaheim, California, on Monday, October 23.

“This type of solution is one we’ll begin to see throughout organizations in the financial services industry,” said Mads Ingwar, Client Services Director at Think Big Analytics. “All banks need a scalable, advanced analytics platform, as well as a roadmap and strategy for digitalization to bring data science into the organization.”