1. Home
  2. Fraud Prevention
  3. Anti-Money Laundering

Fraud prevention report: use holistic approach & machine learning + beware DarkNet

The focus on fraud prevention has accelerated as the number of new cases has increased, e.g.  Yahoo's recent admission that 500 million of their accounts had been hacked:

Over the last month the useful tips on fraud prevention have included:

  • the merchant’s account on file-based e-commerce model in file-based e-commerce model in which they store all sorts of personal information requires a whole new holistic approach according to Andrew Nauman at CyberSource. He recommends a holistic approach to fraud management at each stage of the buying process, while helping to control costs and minimise the impact on the customer experience. (This is key: minimising impact on the customer.) He recommends tuning each layer individually and optimising the system as a whole, e.g. deploying selective asset authentication instead of a manual review, adjusting the rules in automated screening to send fewer orders for review. He believes that fraud prevention optimisation is a process, not an end state. Collecting information, tracking trends and testing new rules are the key to effective ongoing fraud prevention.
  • Europol’s 2016-internet-organized-crime-threat-assessment report shows that one of the biggest threats is from the so-called ‘Darknet’ in which the techniques that have normally been associated with cybercrime are now being used elsewhere. They are particular concerned about the use of ransomeware to extract huge sums of money from corporate's. (This report is essential reading.)
  • PwC’s Global State of Information Security Survey 2016 highlighted the prospects of increasing state sponsored hacks.
  • CyberSource are promoting machine learning which, they believe, outperforms traditional methods of fraud detection - manual checking just takes too long. See The Balancing Act report for tips on how to use machine learning to choose the right models to apply to the historical dataset and optimise the levels of recall and precision.
  • Cardinal Commerce published an important note on the seven secrets of online fraud, see. 

Develop your own fraud prevention strategy

Fraud prevention is a combination of processes and systems. Each company has to develop their own methodology for minimising fraud an excellent starting point is place is what these organisations recommend:

Like this item? Get our Weekly Update newsletter. Subscribe today

This item appears in the following sections:
Fraud Prevention
Anti-Money Laundering
ID Systems & Services in Fraud Prevention
Minimizing Fraud Procedures
Minimizing Payment Fraud

Also see


No comment yet, why not be the first?

Add a comment