I see four pathways to improving how credit card fraud is prevented in e-commerce:
- raising the average level of expertise;
- improving the design of information systems;
- designing more accurate algorithms; and
- improving the quantity and quality of data.
|More Expertise||The first path is to raise the average level of expertise in managing payment fraud. This can be done by having analysts pass the Certified Fraud Examiners (CFE) exam  and by offering other forms of training, e.g., webinars, white papers, online articles, and conferences, targeting e-commerce analysts and decisions makers.|
|Better UI/UX||The second pathway is to improve the design of information systems used to verify payments. Here, the aim is to not only automate most of the recurring tasks e.g., preparing audit reports or updating lists, but also reduce the number of applications analysts use when verifying payments (typically, analysts switch between reservation system(s), Google, social media accounts, email software, and various other websites).|
|More accurate algorithms||The third pathway is to design more accurate algorithms to detect fraud. Here, the two objectives are to update parameters for rules and scores as fast as possible when new cases of fraud are detected and to create algorithms that can consider the cost of each type of error , i.e., failing to detect fraud or suggesting that a payment is fraud when it is a payment from a real client.|
|More data||Finally, the fourth path is to improve the quantity and quality of data used by fraud detection systems. The poor design of the current information systems limit how data is used, particularly to audit risks and predict the likelihood of fraud. In addition, many e-merchants lack data, which makes them more vulnerable to risk, e.g., when entering new markets or when accepting more types of payment methods.|
[..] How can credit card fraud protection be improved? Quora.
 Association of Certified Fraud Examiners
 If you are trying to detect credit card fraud, do you care more about Type I or Type II errors?
 Photo by Benjamin F Clay (Own work) CC BY-SA 3.0