Editorial content organized by topic
Sponsored content from industry partners
PRODUCT/CONTRACT ANNOUNCEMENTS
Latest offerings by category 
Articles submitted by industry partners

 
17 patents awarded in predictive analytics, credit scoring, and fraud detection E-mail


FICO was awarded 17 new patents by the U.S. Patent and Trademark Office. These patents were awarded to members of the FICO Labs team for inventions underlying FICO offerings in predictive analytics, credit scoring, and fraud detection.

“Advances in predicting customer behavior quickly become competitive assets for businesses in the era of big data,” says Dr. Andrew Jennings, chief analytics officer at FICO and head of FICO Labs.

Seven of the new patents are for inventions in predictive analytics, such as: better assessing risk using predictive models that characterize cross-interactions among both categorical and noncategorical input variables; adding new data transaction profile variables to pre-existing models; characterizing risk associated with consolidated debt on a secured instrument; more comprehensive data mining; analyzing insurance claims to detect revenue leakage and over-billing; and creating a new real-time system for event probability prediction.

Two of the 17 new patents are for inventions related to the FICO Score, the standard measure of consumer credit risk in the United States. One invention models the impact of future actions on subsequent creditworthiness. This invention was used to develop the FICO Credit Capacity Index, which helps lenders determine a consumer’s ability to take on additional debt. The second invention in this area identifies how making various hypothetical changes to a consumer’s credit behavior will affect their FICO Scores. This feature has been integrated into the www.myFICO.com site to help consumers better understand their credit health.

Eight of the patents cover enhancements used by FICO’s fraud solutions, including: detecting new forms of fraud, and doing so with less dependency on historical data; updating existing fraud models; identifying fraudulent online transactions over public networks; ensuring security of payment cards; and using transaction data to identify deviations from frequent spending behaviors that could signal fraudulent activity.
 
 
[This article was posted on June 5, 2012, on the website of ABA Banking Journal, www.ababj.com.]       
Trackback(0)
Comments (0)add comment

Write comment
quote
bold
italicize
underline
strike
url
image
quote
quote
smile
wink
laugh
grin
angry
sad
shocked
cool
tongue
kiss
cry
smaller | bigger

security image
Write the displayed characters


busy