Blog article
See all stories »

The rise of machine learning in fraud detection

Fraudulent applications present a significant risk to a businesses’ success and reputation. As criminals become more sophisticated in targeting credit cards, loans, current accounts and other financial products, its critical lenders and other organisations ensure their fraud prevention systems are fit for purpose.

One of the ways businesses can meet this challenge is via Machine learning technology, which is transforming the way businesses approach identifying and preventing fraudulent applications.

The current fraud landscape

Experian’s latest fraud statistics show the scale of the task at hand. Overall financial fraud has risen by 14% in the first six months of 2019, compared to the same time a year previously, driven largely by a substantial 60% increase in card fraud.

Youngsters are also at risk. Fraud against younger people who are first-time buyers has increased by 35% in the first half of the year, while those aged 25-34 are 78% more likely to be the target of fraud, based on the size of the population.

Experian identifies a new fraud every 15 seconds. A staggering figure – and one each represents an individual left, at the very least, inconvenienced, or at the other extreme, distressed, worried and out of pocket.

Machine learning leading the fight

Machine learning is leading the fight against fraud, helping businesses and organisations be more efficient and accurate in their hunt for fraudulent application. It also helps to reduce the level of ‘false positives’, where a genuine is customer is flagged as potentially fraudulent and then investigated by the organisation’s fraud prevention team.

Under typical fraud prevention systems, whether an application is marked as potentially fraudulent depends on the framework and rules set by the organisation, its appetite for risk and disrupting a customer journey.

Machine learning technology goes beyond this rule-based approach. It looks at the results of an application, whether it was found to be fraudulent or not, and then uses this information to make better decisions in the future. The more information it has at its disposal, the higher quality decisions can be made.  

The results are impressive. Globally, genuine customers being flagged as fraudulent has reduced by more than 50%, while increasing fraud detection by 75% and reducing missed fraud by up to 80%. Across the world, each year Experian saves nearly £4 billion stopping fraud before it is committed.

The future of machine learning

The opportunities of machine learning are limitless.  But when it comes to fraud, the benefits for businesses and organisations are already there. The technology can is today helping lenders to accurately pinpoint fraudulent applications in a way that previously would have been impossible, helping protect the organisation’s reputation and giving genuine customers a seamless experience when applying for products at the same time.

4797

Comments: (0)

Micah Willbrand

Micah Willbrand

Product Director ID&F

Experian UK&I

Member since

03 May

Location

London

Blog posts

2

Comments

0

This post is from a series of posts in the group:

Artificial Intelligence and Financial Services

Artificial Intelligence and Financial Services


See all