Randi A Paynter

US Patent:

8473415, Jun 25, 2013

Filed:

May 3, 2011

Appl. No.:

13/100042

Inventors:

Kevin Paul Siegel - Mountain View CA, US
Randi Annette Paynter - San Mateo CA, US
Robert L. Grossman - River Forest IL, US
Christopher Brown - Oakland CA, US
Charles Raymond Byce - Mill Valley CA, US
Thomas Dwyer - San Mateo CA, US
Aoyu Chen - San Jose CA, US

International Classification:

G06Q 40/00

US Classification:

705 44, 705 38

Abstract:

A system, apparatus, and method for reducing fraud in payment transactions by identifying a potential point of compromise (POC) that may lead to subsequent fraudulent transactions, or confirming that a suspected point of compromise is in fact responsible for a security breach leading to fraudulent payment transactions. The invention is also directed to a system, apparatus, and method for identifying data or transaction characteristics that may be used to most effectively identify a point of compromise in a payment transaction system.

US Patent:

8428915, Apr 23, 2013

Filed:

Dec 23, 2008

Appl. No.:

12/317705

Inventors:

Joseph C. Nipko - Burlingame CA, US
Randi A. Paynter - San Mateo CA, US
Robert L. Phillips - Palo Alto CA, US
Robin L. Raffard - Berkeley CA, US

Assignee:

Nomis Solutions, Inc. - San Bruno CA

International Classification:

G06F 17/10
G06F 7/60

US Classification:

703 2, 703 6, 706 12

Abstract:

Data for a transaction is modeled by receiving a source set of data. The source set of data comprises data representing a plurality of transactions stored in a source transaction database. An estimation model for modeling data for a transaction is received. A mapping between the source set of data and a model parameter database is received. The model parameter database comprises a plurality of model parameters for the estimation model. The parameters extracted from the model parameter database and the source set of data in a Bayesian framework are combined using a parameter estimation engine to obtain an updated set of model parameters. The updated set of model parameters is stored in the model parameter database.