Subscribe to DSC Newsletter

Credit Card Fraud

Each circle is a fraudulent transaction in one particular fraud case, over several months. Circle radius represents dollar amount. Color represents recency, from blue (old) to red (new). The fraud spread from the East to the West coast, as you can tell by the colours.

Views: 2353

Comment by Mark Richards on May 10, 2008 at 4:26pm
Eye Candy!
(Closer inspection allows me to see a very well formed & informative graph. I particularly like the color/time element!)
Comment by Anuj Taneja on June 19, 2008 at 11:05pm
Interesting representation. Are these the fraud transactions on a bunch of credit cards compromised at a merchant/issuer?
Comment by Vincent Granville on June 19, 2008 at 11:08pm
It involved several credit cards, several issuers and several merchants. Soome credit cards had just one fraud occurence (single ping).
Comment by Jagdish Belwal on August 17, 2008 at 11:45am
What I would be interested in is "The representation is great but what is the objective of this graph - can it predict the next big one or does it help the investigation agencies in usefully analysing the trends...."
Comment by chris on August 17, 2008 at 1:27pm
Colorful, but it appears one needs to know something about which colors between blue and red in order to understand "recency" - presumably chronology.
You would have a more informative picture if you put a "legend" for the color along the bottom or side axis
Comment by Eric on September 21, 2008 at 4:13am
!!Excellent work!! - it really speaks to the magnitude and complexity of an organized fraud.
Comment by hafiz kashif munir on September 4, 2009 at 11:50am
hi, im student & doing BCS final year project on "Distributed Data Mining in credit card fraud detection",
in which i have transaction records in which i want to layout fraudulent transaction by using Algo's approach, plz guide me
Comment by hafiz kashif munir on September 4, 2009 at 11:51am
im using Weka & C# with SQL server
Comment by David Klemitz on March 30, 2015 at 6:41pm

Besides scalability and efficiency, the fraud-detection task exhibits technical problems that include skewed distributions of training data and nonuniform cost per error, both of which have not been widely studied in the knowledge-discovery and data mining community.

In this article, we survey and evaluate a number of techniques that address these three main issues concurrently. Our proposed methods of combining multiple learned fraud detectors under a "cost model" are general and demonstrably useful; our empirical results demonstrate that we can significantly reduce loss due to fraud through distributed data mining of fraud models.


You need to be a member of AnalyticBridge to add comments!

Join AnalyticBridge

Follow Us

On Data Science Central

On DataViz

On Hadoop

© 2018 is a subsidiary and dedicated channel of Data Science Central LLC   Powered by

Badges  |  Report an Issue  |  Terms of Service