I am looking for a book/case study etc on how to build a fraud detection model at the transaction level. Something applied rather than theoretical would be really helpful.
$\begingroup$
$\endgroup$
3
-
$\begingroup$ I can't actually remember if such a specific application is covered in Biometrics: Theory, Methods, and Applications, by Boulgouris et al. (Wiley, 2009), but it may be worth checking on Google Books. $\endgroup$– chlFeb 7, 2011 at 9:51
-
$\begingroup$ The articles referred to here may be relevant. finance30.com/forum/topics/template-for-benfords-first $\endgroup$– Graham WebsterFeb 9, 2011 at 10:50
-
$\begingroup$ Do you have a particular application domain in mind? I suspect the techniques might depend upon the application domain. There's a lot of work in the computer science literature which may be relevant, under the phrase "anomaly detection". $\endgroup$– D.W.Dec 4, 2011 at 21:35
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
1
Fraud detection is a rare class problem. Chapter Six of Charles Elkan's Notes for his Graduate Course in Data Mining and Predictive Analytics at UCSD walks you through the prediction of a rare class, and the pitfalls and proper ways to evaluate the success of such a model. The methods he specifically uses are Isotonic and Univariate Logistic Regression. The software he uses in the class is Rapidminer, but I prefer R. If you choose to use R, you can perform both of these functions using the isoreg and glm functions. Many people also like to use SVMs in fraud detection, but part of the model selection criterion should be the speed with which you need to validate the transactions. If, for example, it's the swipe of a credit card, SVMs are wholly unfeasible because it will take far too long to process. This is why, in production environments, variants on regression models are typically used for fraud detection.
-
1$\begingroup$ Those notes appear to be here now, in Chapter 7: cseweb.ucsd.edu/~elkan/255/dm.pdf $\endgroup$– ClayJul 8, 2013 at 20:22