Replies to This Discussion

Permalink Reply by Kumud Joseph Kujur on January 25, 2010 at 3:08am

Hi Tomas,

Baisically it depends upon the type of frauds. I've used Association & link analysis for some of the financial frauds
It indeed showed some good results

Regards,
Kumud

Permalink Reply by Tomas Keller (formerly Ohlson) on January 25, 2010 at 1:04pm

Thanks for the feedback Kumud,
association and link analysis sounds like an interesting approach in fraud detection, not applicable in my case but anyway.

Thanks,

Tomas

Permalink Reply by Larry on June 17, 2010 at 7:52am

Do you have any good references of Association and Link Analysis? Perhaps links that explain.

Permalink Reply by John Crout on June 16, 2010 at 1:20pm

What does this refer to? I understand what is meant by "fraud detection" and by "unsupervised" but don't know if it is the fraud itself, or if it is detection of the fraud that is "unsupervised". In either usage, how does "unsupervised" fit?

Permalink Reply by Vincent Granville on June 16, 2010 at 4:01pm

I think it means that the algorithm used for detection is unsupervised clustering. In the context of spam or click fraud, unsupervised is typically used; in the context of credit card fraud, supervised is used. Supervised means that we have a training set at our disposal, with known fraudulent and non-fraudulent transactions.

Permalink Reply by Tomas Keller (formerly Ohlson) on June 18, 2010 at 12:57am

That's correct Vincent.
The client wants me to develop a model that can identify suspicious behaviour that can be put on a watch list.
@John, I don't have any confirmed fraud cases since the client wants to monitor a new type of fraud. So I can't use a supervised method to find suspicious activity, instead some clustering method (unsupervised) can be used to identify transactions that can be related to the new fraud approach. That's why I used the term 'Unsupervised fraud detection' , what I ment was 'unsupervised method to be used in fraud detection'.

Regards,
Tomas

Permalink Reply by DataLLigence on June 16, 2010 at 9:31pm

Hi Thomas,

I had worked on a small project where i identified/detected fraudulent transactions (fuel cards). The client was using SQL based rules to detect frauds. I used plain clustering/profiling methods and was able to detect almost twice the number of fraudulent transactions (compared to the number the client was able to catch with their existing rules).

Regards,
Datalligence

Permalink Reply by Tomas Keller (formerly Ohlson) on June 18, 2010 at 12:39am

Hi

that's the approach I'm using and I find it quite successfull.

My client is using some 'gut feeling generated' rules so the clustering method is of course much better at identifying odd/fraudulent behaviour.

Tomas

Permalink Reply by Larry on June 17, 2010 at 12:06pm

I've been interested in this as well. Based on my limited research it seems Peer Group Analysis is a popular method.

Here is a paper on it. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.24.5743&am...

Also here is a short Wikipedia blurb on it as well (under Unsupervised Methods)
http://en.wikipedia.org/wiki/Data_Analysis_Techniques_for_Fraud_Det...

Permalink Reply by Tomas Keller (formerly Ohlson) on June 18, 2010 at 12:31am

Hi Larry,

thanks a lot for the reference!


Tomas