Ex Parte Zoldi et al

Patent Trial and Appeal Board

Jul 6, 2016

12040796 (P.T.A.B. Jul. 6, 2016)

UNITED STA TES p A TENT AND TRADEMARK OFFICE
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR
12/040,796 02/29/2008
76615 7590 07/08/2016
Mintz Levin/Fair Isaac
Mintz Levin Cohn Ferris Glovsky and Popeo, P.C.
One Financial Center
Boston, MA 02111
Scott M. Zoldi
UNITED STATES DEPARTMENT OF COMMERCE
United States Patent and Trademark Office
Address: COMMISSIONER FOR PATENTS
P.O. Box 1450
Alexandria, Virginia 22313-1450
www .uspto.gov
ATTORNEY DOCKET NO. CONFIRMATION NO.
35006-621F01US 1110
EXAMINER
BUCHANAN, CHRISTOPHER R
ART UNIT PAPER NUMBER
3627
NOTIFICATION DATE DELIVERY MODE
07/08/2016 ELECTRONIC
Please find below and/or attached an Office communication concerning this application or proceeding.
The time period for reply, if any, is set in the attached communication.
Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the
following e-mail address( es):
IPDocketingBOS@mintz.com
IPFileroombos@mintz.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte SCOTT M. ZOLDI, LARRY PERANICH, JEHANGIR ATHWAL,
UWE MA YER, and SAJAMA
Appeal2014-003757 1
Application 12/040,7962
Technology Center 3600
Before MICHAEL C. ASTORINO, NINA L. MEDLOCK, and
ROBERT J. SILVERMAN, Administrative Patent Judges.
MEDLOCK, Administrative Patent Judge.
DECISION ON APPEAL
STATEMENT OF THE CASE
Appellants appeal under 35 U.S.C. § 134(a) from the Examiner's final
rejection of claims 1-20. We have jurisdiction under 35 U.S.C. § 6(b).
We REVERSE.
1 Our decision references Appellants' Appeal Brief ("App. Br.," filed
September 24, 2013) and Reply Brief ("Reply Br.," filed February 10,
2014), and the Examiner's Answer ("Ans.," mailed December 18, 2013) and
Final Office Action ("Final Act.," mailed April 5, 2013).
2 Appellants identify Fair Isaac Corporation of Roseville, Minnesota as the
real party in interest. App. Br. 2.
Appeal2014-003757
Application 12/040,796
CLAIMED INVENTION
Appellants' claimed invention "relates generally to a computer-based,
real-time adaptive system for fraud detection implementing automatic
binning, feature selection, adaptive statistical models, and score blending"
(Spec. ii 1 ).
Claim 1, reproduced below, is illustrative of the subject matter on
appeal:
1. A computer-implemented method comprising:
receiving a new data record associated with a transaction;
generate, using a base model executed by a computer, a
first score being a first likelihood of the transaction being
associated with fraud;
generating, when the first likelihood is more than a
threshold and using an adaptive model executed by the computer,
a second score to represent a second likelihood of the transaction
being associated with fraud, the adaptive model receiving
feedback from one or more external data sources, the feedback
containing information about one or more previous data records
associated \'l1ith the base model generated by scoring parameters
from at least one of the one or more external data sources, the
feedback being used to update scoring parameters within the
adaptive model that are used to generate the second score.
REJECTION
Claims 1-20 are rejected under 35 U.S.C. § 103(a) as unpatentable
over Hammond (US 2009/0076870 Al, pub. Mar. 19, 2009) and Official
Notice.
2
Appeal2014-003757
Application 12/040,796
ANALYSIS
Independent claim 1 and dependent claims 2-10 and 20
We are persuaded by Appellants' argument that the Examiner erred in
rejecting claim 1under35 U.S.C. § 103(a) because Hammond does not
disclose or suggest "generat[ing], using a base model executed by a
computer, a first score being a first likelihood of the transaction being
associated with fraud" and "generating, when the first likelihood is more
than a threshold and using an adaptive model executed by the computer, a
second score to represent a second likelihood of the transaction being
associated with fraud, the adaptive model receiving feedback from one or
more external data sources," as recited in claim 1 (App. Br. 12-16).
By way of background, Hammond is directed to a system and method
for detecting suspected fraud when processing a merchandise return
transaction requested by a customer (Hammond, Abstract, ,-r 6). Hammond
discloses, with reference to Figure 5, that data 510 are collected during the
return transaction and processed with stored data 520 (e.g., information
about the customer, the clerk, the store) to create variables 530 that are
indicators of fraud (id. i-fi-1 82, 83). Variables 530 are based on the
customer's previous transactions with the merchant and may include the
number of returned transactions over a particular period of time, the
customer's pattern of travel, sales and return seasonality, etc. (id. i-fi-184, 87-
95). A fraud detection model 540 uses variables 530 to calculate a fraud
score 550 that indicates the likelihood that fraud is occurring in the return
transaction (id. ,-r 85). And Hammond discloses that score 550 is used by the
return decision process to activate business rules specified by the merchant
(id. ,-r 100). For example, one or more business rules may specify that
3
Appeal2014-003757
Application 12/040,796
customers whose requested return transaction is assigned a score above a
pre-determined threshold value may have their return accepted while
customers whose requested return transaction is assigned a score below the
pre-determined threshold value may have their return denied (id. i-f 101 ).
In rejecting claim 1under35 U.S.C. § 103(a), the Examiner takes the
position that variables 530 constitute "a first score being a first likelihood of
the transaction being associated with fraud" (Final Act. 2; see also Ans. 2).
However, we agree with Appellants that variables 530 "are merely values of
number of items being returned, pattern of travel, consumer profitability,
sales associated with the product being returned, and return seasonality
associated with the product being returned" (App. Br. 12). These values,
thus, are, at best, merely indications of fraud rather than representing a
likelihood, i.e., a probability, of a transaction being fraudulent (id. at 12-13).
Moreover, even accepting the Examiner's interpretation of variables 530 as
"a first likelihood of the transaction being associated with fraud," Hammond
does not disclose or suggest "generating ... using an adaptive model ... a
second score to represent a second likelihood of the transaction being
associated with fraud" when "the first likelihood [i.e., variables 530] is more
than a threshold."
The Examiner acknowledges that "[t]he method/program of
Hammond differs from the claimed invention in that the particular score
compared with the threshold is not the first score and in that the steps taken
if the threshold is exceeded are not shown to be calculation of a second
score" (Final Act. 3). The Examiner takes Official Notice that "it is well-
known to compare values to a threshold and take certain steps if the
threshold is exceeded" (id.). And the Examiner concludes that it would have
4
Appeal2014-003757
Application 12/040,796
been obvious to a person of ordinary skill in the art at the time of
Appellants' invention to
modify the method/program of Hammond so that the score
compared with the threshold is the first score and the steps taken
if the threshold is exceeded is calculation of a second score, as
suggested by official notice, because modifying the teachings of
the prior art with these well-known features would merely yield
predictable results
(id. at 4). Yet the Supreme Court has made clear that"' [r]ejections on
obviousness grounds cannot be sustained by mere conclusory statements;
instead, there must be some articulated reasoning with some rational
underpinning to support the legal conclusion of obviousness."' See KSR
Int'! Co. v. Teleflex, Inc., 550 U.S. 398, 418 (2007) (quoting In re Kahn, 441
F.3d 977, 988 (Fed. Cir. 2006)).
The Examiner takes Official Notice that comparing values to a
threshold is well-known, and summarily concludes that it would have been
obvious "to modify the method/program of Hammond so that the score
compared with the threshold is the first score and the steps taken if the
threshold is exceeded is calculation of a second score" (Final Act. 4). But
even if comparing values to a threshold is well-known, there is nothing in
the record to indicate that it also is well-known and/or a matter of common
knowledge to use an adaptive model to compare the likelihood of a
transaction being associated with fraud (as generated by a base model) to a
threshold to determine whether to generate a second score, as recited in
claim 1.
The Examiner concludes here, based on the taking of Official Notice,
that "it is well-known to compare values to a threshold and take certain steps
if the threshold is exceeded," and that it would have been obvious to modify
5
Appeal2014-003757
Application 12/040,796
Hammond to arrive at Appellants' claimed invention. Final Act. 3.
However, the Examiner does not explain why or how the modification
necessarily follows from this allegedly well-known fact and, therefore, does
not provide the articulated reasoning with rational underpinnings required to
support the obviousness determination.
In view of the foregoing, we do not sustain the Examiner's rejection
of claim 1under35 U.S.C. § 103(a). For the same reasons, we also do not
sustain the rejection of dependent claims 2-10 and 20. Cf In re Fritch,
972 F.2d 1260, 1266 (Fed. Cir. 1992) ("dependent claims are nonobvious if
the independent claims from which they depend are nonobvious").
Independent claim 11 and dependent claims 12-19
Independent claim 11 includes language substantially similar to the
language of claim 1 and stands rejected based on the same findings and
rationale applied with respect to claim 1. Therefore, we do not sustain the
Examiner's rejection under 35 U.S.C. § 103(a) of independent claim 11 and
claims 12-19, which depend therefrom, for the same reasons set forth above
with respect to claim 1.
DECISION
The Examiner's rejection of claims 1-20 under 35 U.S.C. § 103(a) is
reversed.
REVERSED
6