Generally speaking, we are getting tens of thousands of new customers on our network a day, tens of thousands of baddies of different sorts, not paying their bills or whatever, a day. So, the fact that we are able to distill this down to less than a thousand things for our fraud team to investigate, it is a big crunching of this down.

In a physical sense, as a physicist, you have got all of this data. Where do you look? So, these tools are to guide our physicists to show, this is where you have got to look.

I am just going to close things by saying this is ongoing work, a lot going on, and thank you for your patience and your time. Thanks.

MS. KELLER-MC NULTY: While we are getting John hooked up for the next session, if there are a couple of questions?

[Question off microphone.]

MR. PREGIBON: The question is, do I ever use spectral techniques for these graphs. I think possibly what you mean there is some of the Hudson Authority type computations?

[Comments off microphone.]

MR. PREGIBON: No, we have not. We should probably talk off line. There is some mathematics associated with analysis of graphs in the literature, with argon analysis. Those sorts of spectral methods are used to characterize and process findings of special nodes on the graph. For people who are familiar with the work of John Kleinberg from Cornell, his Hudson Authority work is of that ilk.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement