First Published on 25th July, 2018
Ai Editorial: A risk-averse mindset is commonly associated with rule-based systems, which is built with hard rules or buying limits, such as geo-location rules that could block out all transactions from one region, writes Ai's Ritesh Gupta
When merchants rely on conventional or long-used methods of spotting fraud, it tends to be associated with evaluating the standard fields (name, address, email, IP location, fingerprint and what can be found on the order form) and what transactions have cleared through the set rules.
The issue here is that those standard fields and hard rules are not tough for fraudsters/ hackers to break into and get breached once they have understood the rules worked out. For instance, it is quite straightforward for fraudsters to focus on new fake emails, and once they comprehend that a time based rule is set, they will attempt to set their program to go past the system. Not only so, authentic buyers are likely to be blocked. For instance, a geo-location rule would block customers booking transactions from ‘riskier’ locations.
Machine learning systems are meant to be an improvement from rule-based systems, to reduce reliance on hard rules and to filter out fraud while passing more genuine users. However, machine learning systems only provide probability scores - or fraud scores - and would still require a team of manual reviewers to make sense of the score and thereafter a decision to pass or reject a transaction.
Unfortunately, the fraud team’s KPI is still to ensure fraud rates are low - perpetuating the risk-averse mindset as they would rather reject a transaction than to risk passing a fraudulent one. To overcome such “risk-averse” mindset, it would require the fraud team to understand that risk is very much similar to financial risk; it should be managed, not eliminated. Since 0% risk gives 0% returns, having little to no fraud would mean much revenue has been lost. For merchants to fully overcome having a “risk-averse” fraud management system, a financial algorithm could be combined with the machine learning system to make sense of the risk financially, allowing for more revenue based on a greater risk appetite.
Also by focusing on machine learning, carriers can eradicate all those needless rules that would have otherwise stopped authentic buyers from competing their respective transactions.
The blend of big data and machine learning paves way for more solid fraud prevention.
As we highlighted in our previous articles, to simplify big data and machine learning, big data is first used to garner details about the user’s behaviour on the website (how the mouse moves, what he likes or puts into his wishlist, etc), and this information is combined with machine learning, which uses pattern recognition to map the pattern of his behaviour to match it either with positive (genuine) or negative (fraudulent) behaviour, as well as predictive analytics that records the positive/negative behaviour and uses that on future transactions for potential signs of fraud.
Lastly, an optimized fraud risk algorithm should be used to make decisions on whether or not to accept a transaction based on calculated risks to best optimize sales while controlling fraud and chargeback rates.
Hear from airlines and other industry executives about travel fraud at the upcoming 7th Annual Airline & Travel Payments Summit (ATPS), co-hosted with UATP, (4- 6 September 2018 in Phuket, Thailand).
For more click here
Follow Ai on Twitter: @Ai_Connects_Us