Practical Fraud Prevention

Part I. Introduction to Fraud Analytics

The following five chapters cover the basic elements of fraud analytics, as well as the wider context that is shared across the various verticals of the ecosystem. The first two chapters break down the phenomenon of fraud into behaviors that characterize fraudsters and the typology of attacker personas. We believe it’s crucial for every researcher to be able to envision the fraudsters’ mindset. Fraudsters are definitely persona non grata in your system, but they do still have a persona. They are people driven by very human needs, and understanding where they come from can be a big step in your evolution as a researcher (and a human being).

The next two chapters focus on the more practical tactics that every fraud researcher keeps in their toolbelt (which is almost as cool as Batman’s utility belt)—for example, being able to spot anomalies and figure out whether they may be seasonal, or being able to assess the risk of fraud in the context of your profit margins and then deciding whether it’s worth it to develop an in-house solution or invest in working with a third-party provider. Without these fundamentals, it’s virtually impossible to go into the business of risk management.

Finally, Chapter 5 is devoted to opening the black box (and sometimes Pandora’s box) of fraud modeling. This chapter cannot be used to teach machine learning, but it can certainly start an important (and fun!) discussion with your data scientists.