Is machine learning the answer to the evolving fraud threat?

Is machine learning the answer to the evolving fraud threat?

Tuesday, January 30, 2018 Totally Gaming
No alternative text provided

With 3.5% of all online gambling payments now estimated to be fraudulent and bot attacks on the rise, it is becoming clear that traditional rule-based fraud prevention is no longer fit for purpose in the gaming sector.

In its latest white paper, Playing a Smarter Game: How Machine Learning is Changing Online Gambling, global payment solution provider Acapture finds that machine learning could provide a compelling, and lucrative, answer. 

At ICE 2018, Aldrin Mangalabal, Acapture’s fraud expert, will be hosting a series of 1on1 demonstrations showing how their AceProtect solution uses machine learning to fight fraud and improve conversion rates for online gambling operators. 

And, as news from the past year shows, it is an appointment that operators cannot afford to miss. 

The online gambling threat

The size of the sector’s fraud threat is already well established. In fact, Europay reports that more than 20% of online financial fraud cases occur on gambling sites. What is perhaps even more alarming, however, is how quickly the threat is evolving.

Declan Hill, journalist and author of international bestseller The Fix: Soccer and Organized Crime, says that those looking to manipulate sports events for the purposes of gambling are now ‘using tactics similar to a fraudulent stock promoter’. Deploying bots, they flood social media with false team news or statistics in order to influence betting patterns in their favor. 

The potential damage done by malignant bots doesn’t stop there. Bot attacks are growing in regularity, allowing cyber criminals to breach the security of operators’ platforms and steal consumer data or create false accounts using data stolen elsewhere. 

And, as these threats become more sophisticated and unpredictable, it is increasingly clear that traditional rule-based fraud prevention is not up to the task of beating them back.

No alternative text provided

Rule-based Vs. Machine learning 

The majority of online fraud prevention solutions still work off a rule-based system. This means the operator encodes a collection of rigid rules regarding how transactions are handled on their site. If any of these rules are broken, the payment is flagged as fraudulent and may be rejected. 

The issue with this system is two-fold. Firstly, its lack of nuance means honest players often find their payment rejected as, unbeknownst to them, at some stage of the payment process something happened that stepped out of the rule-based solution’s directions. Secondly, these solutions do not have the flexibility to adapt to new threats as they emerge without significant human input, making it difficult to stay ahead of the fraudsters’ innovations. 

That’s why machine learning, with its ability to learn from new data and change its behavior on that basis, is such a radical and exciting proposition for fraud prevention. Rather than simply executing a fraud strategy based on a set of predetermined guidelines, machine learning fraud learns and updates in real time. By evaluating new data and signals alongside previous data patterns it can effectively adapt to new threats as they emerge, make better judgements on whether a consumer is legitimate and limit the risk of potentially damaging chargebacks. 

Machine learning fraud solutions take into account the location of the player, their browsing behavior on the site, their transaction details, device information, historical order information and thousands of other signals. This allows them to make real time judgements, tackling all the fraud types common in online gambling including bonus abuse, promo abuse and account hijacking. Crucially, however, they also shield honest players, providing a smoother, undisrupted experience of the platform. 

Are you interested to know about how machine intelligence can lower the fraud risk for your business? Then be sure to book your 1on1 demo with Aldrin here and find out everything you need to know.

No alternative text provided

No alternative text provided

RISQ's Julian Borg-Barthet - The evolving jackpot landscape

No alternative text provided

QTech Games CEO Jonas Alm: The challenges and complexities of the Asian market

No alternative text provided

Number Crunching - John Asher, Cherry AB and GIVEMESPORT

No alternative text provided

Number Crunching - STS, DrückGlück and Etoro

Gaming Products & Services Directory

The essential directory for the gaming industry