Ex Parte Bramlett et al

Patent Trial and Appeal Board

Jan 27, 2017

13107176 (P.T.A.B. Jan. 27, 2017)

United States Patent and Trademark Office
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
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO.
13/107,176 05/13/2011 William L. Bramlett JR. 2591.028US1 6873
25763 7590 01/31/2017
DORSEY & WHITNEY LLP - MINNEAPOLIS
ATTENTION: PATENT PROSECUTION DOCKETING DEPARTMENT
INTELLECTUAL PROPERTY PRACTICE GROUP - PT/16TH EL
50 SOUTH SIXTH STREET, SUITE 1500
MINNEAPOLIS, MN 55402-1498
EXAMINER
SCHEUNEMANN, RICHARD N
ART UNIT PAPER NUMBER
3624
NOTIFICATION DATE DELIVERY MODE
01/31/2017 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):
ip. docket @ dorsey .com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte WILLIAM L. BRAMLETT JR., BRIAN C. GIEDT,
and JOSEPH M. KISTNER
Appeal 2015-0027971
Application 13/107,176
Technology Center 3600
Before MURRIEL E. CRAWFORD, JOSEPH A. FISCHETTI, and
MICHAEL W. KIM, Administrative Patent Judges.
KIM, Administrative Patent Judge.
DECISION ON APPEAL
STATEMENT OF THE CASE
This is an appeal from the final rejection of claims 1—6, 9-24, 27—38,
and 40-42. We have jurisdiction to review the case under 35 U.S.C. §§ 134
and 6.
The invention relates generally to using a neural network to select
employees for a particular job or occupation. Spec. 115.
1 The Appellants identify Profiles International, Inc. as the real party in
interest. Appeal Br. 2.
Appeal 2015-002797
Application 13/107,176
Independent claim 1 is illustrative:
1. A system comprising:
one or more computer processors configured for:
receiving data relating to a plurality of persons, the
plurality of persons employed in the same occupation, a portion
of the plurality of persons comprising top performers in the
occupation, and a portion of the plurality of persons comprising
bottom performers in the occupation, wherein the data relates to
one or more of personal traits and performance traits;
inputting the data into a software-based neural network;
using the neural network to generate models for the
personal traits as a function of the personal traits and the
performance traits of the top performers; and
using the neural network to generate performance models
comprising the personal traits models;
wherein the performance models are configured to
determine that a particular person, who is not one of the plurality
of persons, will likely be a top performer in the occupation, a
bottom performer in the occupation, or neither a top performer
or a bottom performer;
wherein the personal traits models comprise a numeric
range; and
wherein the performance models comprise at least one
personal traits model that includes, among the performance
models, two or more different sub-ranges such that the system is
less likely to treat the particular person as an outlier candidate.
Claims 1—6, 9-24, 27—38, and 40-42 are rejected under 35 U.S.C.
§ 101 as reciting ineligible subject matter in the form of an abstract idea.
Claims 1 and 16 are rejected under 35 U.S.C. 112, first paragraph, as
failing to comply with the written description requirement.
Claims 1, 2, 9-16, 18—23, and 27—37 are rejected under 35 U.S.C.
103(a) as being unpatentable over Thissen-Roe (US 2006/0282306 Al, pub.
Dec. 14, 2006) and Rosner (US 2008/0033792 Al, pub. Feb. 7, 2008).
2
Appeal 2015-002797
Application 13/107,176
Claims 3—5 are rejected under 35 U.S.C. § 103(a) as unpatentable over
Thissen-Roe, Rosner, and Rosen (US 7,805,382 B2, iss. Sept. 28, 2010).
Claims 17 and 24 are rejected under 35 U.S.C. § 103(a) as
unpatentable over Thissen-Roe, Rosner, and Scarborough (US 7,562,059 B2,
iss. July 14, 2009).
Claim 6 is rejected under 35 U.S.C. § 103(a) as unpatentable over
Thissen-Roe, Rosner, Rosen, and Scarborough.
Claim 38 is rejected under 35 U.S.C. § 103(a) as unpatentable over
Thissen-Roe, Rosner, and Bryce (US 2002/0143573 Al, pub. Oct. 3, 2002).
Claim 40 is rejected under 35 U.S.C. § 103(a) as unpatentable over
Thissen-Roe, Rosner, and “Prevue Client Service System”, Leader’s Guide
(Profiles International, Inc., Waco, Texas), (1995), 6 pgs. (hereinafter
“Prevue”).
Claims 41 and 42 are rejected under 35 U.S.C. § 103(a) as
unpatentable over Thissen-Roe, Rosner, and Levin (US 2003/0101091 Al,
pub. May 29, 2003).
We AFFIRM.
ANALYSIS
Rejection of Claims 1—6, 9—24, 27—38, and 40—42 under 35 U.S.C.
mi
We are persuaded by the Appellants’ argument that the pending
claims do not merely recite abstract ideas, because they are instead directed
to systems that use machine intelligence to analyze human data. Reply Br.
9. The Examiner finds the subject matter of the claims to be abstract ideas
that “require no more than a generic computer to perform generic computer
3
Appeal 2015-002797
Application 13/107,176
functions that are well-understood, routine and conventional activities
previously known to the industry.” Answer 5—6. The Examiner provides no
evidence or reasoning, however, to support how the claimed use of a “neural
network,” a complex artificial intelligence software application (Spec.
1113—14) that is recited in each of the independent claims, is a “generic
computer” that performs “generic computer functions.”
For this reason, we do not sustain the rejection of claims 1—6, 9—24,
27-38, and 40-42 under 35 U.S.C. § 101.
Rejection of Claims 1 and 16 under 35 U.S.C. § 112, First Paragraph,
as New Matter
We are persuaded by the Appellants’ argument that there is support in
the application for multiple sub-ranges that make it less likely to treat a
candidate as an outlier, as claimed. Appeal Br. 10-12; see also Reply Br. 1—
3.
Specifically, the Examiner finds “multiple sub-ranges are not
disclosed, and multiple sub-ranges to reduce the likelihood that a candidate
will be treated as an outlier are not disclosed.” Final Act. 5.
The Appellants’ Specification states that “FIG. 4 further illustrates
how a particular person compares with each of the personal trait models 435,
generated by the neural network using the personal trait data of the top and
bottom performers.” Spec. 121. Figure 4, shown below, is annotated to
point out element 435, which shows a series of traits with associated ranges,
as follows:
4
Appeal 2015-002797
Application 13/107,176
The Appellants’ Figure 4 annotated to highlight traits with
associated ranges shown in element 435.
The Specification provides an example that demonstrates the use of
multiple sub-ranges for a single trait model:
[NJeural network determined that the data for the top and bottom
performers indicate that top bank tellers display an assertiveness
ranking of 3-5 and 7-9, and an independence ranking of 1-3 and
6- 9. Consequently, the neural network generated a performance
model 400 to identify potential top bank tellers, wherein the
assertiveness and independence rankings are 3-5 and 6-9
respectively, and a performance model 450 to identify top bank
tellers, wherein the assertiveness and independence rankings are
7- 9 and 1-3 respectively.
Spec. 123. As the Appellants point out, this discloses multiple sub-ranges,
being 7—9 and 1—3, within a single trait model, as claimed. Appeal Br. 11.
We agree with the Appellants that the ordinary artisan would understand
from this as follows:
If there is a first sub-range and a second sub-range, and a person
falls within the second sub-range, the presence of that second
5
Appeal 2015-002797
Application 13/107,176
sub-range makes it less likely that that person will be treated as
an outlier (because the second sub-range expands the acceptable
scores that are considered as meeting the personality trait for the
job position). The presence of the second sub-range also makes
it less likely that an outlier will be missed for consideration of a
job position.
Appeal Br. 12. The existence of multiple sub-ranges means more candidates
may fall within one or the other, which may be more inclusive than having
only one range.
We are persuaded that the ordinary artisan would, thus, have
recognized from the Appellants’ original Specification that the Appellants
had possession of “two or more different sub-ranges such that the system is
less likely to treat the particular person as an outlier candidate,” as claimed.
For these reasons, we do not sustain the rejection of claims 1 and 16
under 35U.S.C. § 112, first paragraph.
Rejection of Claims 1, 2, 9—15, and 34—37 under 35 U.S.C. f 103(a)
The Appellants argue independent claims 1, and 34—37 together as a
group. Appeal Br. 19. We select claim 1 as representative. See 37 C.F.R.
§ 41.37(c)(l)(iv).
We are unpersuaded by the Appellants’ argument that Thissen-Roe
does not disclose two or more sub-ranges within a personal traits model,
using a neural network. Appeal Br. 13—16; see also Reply Br. 5—6.
Thissen-Roe discloses “a trait predictor can predict any of a variety of
personality traits such as assertiveness, conscientiousness, diligence,
integrity, responsibility, honesty, reliability, ambition, resilience,
compliance, and the like.” Thissen-Roe, para. 88. Thissen-Roe also
discloses “[i]n any of the examples herein, a predictive model can be a
6
Appeal 2015-002797
Application 13/107,176
neural network, expert system, or other artificial intelligence model.” Id. at
1117. We are persuaded that the ordinary artisan would, thus, have
recognized that Thissen-Roe discloses “using the neural network to generate
models for the personal traits as a function of the personal traits and the
performance traits of the top performers,” as claimed.
In addition, Thissen-Roe discloses the output of its trait predictor can
be multiple ranges, in that its “trait predictor can apply pre-processing to its
inputs to provide the output (e.g., to a predictive model), which can take the
form of an estimate of where within a bell curve the candidate lies (e.g., a
distribution from -3 to 3, with a standard deviation of 1).” Thissen-Roe,
para. 90. Here, the ordinary artisan would have recognized the distribution
of a range of -3 to 3 includes a sub-range from -3 to zero, and a second sub­
range from zero to 3, meeting the claim language of “at least one personal
traits model that includes, among the performance models, two or more
different sub-ranges.” Insofar as the Appellants may be asserting that the
two sub-ranges must be explicitly identified in the prior art and/or cannot be
overlapping in any respect, such limitations are not recited in the claim.
We are not persuaded by the Appellants’ argument that Thissen-Roe
fails to disclose “any system or means that is less likely to treat a candidate
as an outlier,” specifically in paragraphs 176—77. Appeal Br. 14—15; see
also Reply Br. 4.
The Specification describes that “the neural network generates a
plurality of performance models. These performance models include one or
more different models for the models that make up the performance models.
The plurality of performance models makes it less likely that an outlier
candidate will be missed.” Spec. 123. The Specification, thus, discloses
7
Appeal 2015-002797
Application 13/107,176
that the method by which “the system is less likely to treat the particular
person as an outlier candidate,” is to include multiple, different models. The
Specification does not explain how multiple sub-ranges leads to it being less
likely to treat a candidate as an outlier. Reading claim 1 broadly, in light of
the Specification {Id.), we construe the language to mean that having sub­
ranges that cover more traits makes them more inclusive, and, thus, helps to
avoid candidates being treated as outliers by having more candidates
included with more, or wider, ranges of acceptable traits. In addition,
because Thissen-Roe discloses the use of multiple, different trait models (see
Thissen-Roe, para. 88), Thissen-Roe is also “less likely to treat the particular
person as an outlier candidate.” Similarly, multiple, wide sub-ranges would
tend to be inclusive by including more traits in the acceptable range, so
Thissen-Roe’s six-point-wide zone covered by its inclusive range would be
more inclusive, we assume, than a two-point-wide zone, or two one-point­
wide zones, such as from -1 to zero and zero to 1.
The Appellants refer to the Declaration of William Bramlett as
evidence of patentability, especially as to the purpose of Thissen-Roe, and
the lack of neural networks at use in the industry. Appeal Br. 16—18.
The Declaration asserts that “the contribution of the Thissen-Roe
system seems to be the use of a neural network to implement an adaptive
assessment system.” Deck, para. 4. It is not necessary for the prior art to
serve the same purpose as that disclosed in the Appellants’ Specification in
order to support the conclusion that the claimed subject matter would have
been obvious. See In reLinter, 458 F.2d 1013, 1016. Although Thissen-
Roe does discuss adaptive questioning techniques (see Thissen-Roe 13), the
adaptive questioning is in addition to other disclosures, such as the use of a
8
Appeal 2015-002797
Application 13/107,176
model that predicts traits as inputs to a model that makes candidate
predictions (Thissen-Roe 1 85) (“a predictive model 1410 includes a trait
predictor 1420 that accepts some of the inputs directed to the predictive
model 1410 and generates a trait predictor output 1482, which is fed to the
prediction engine (e.g., a neural network) 1430, which then generates the
output OUT.”).
The Declaration also repeats the argument that “Thissen-Roe does not
seem to disclose using a neural network to generate personal trait models,
and then using the neural network to generate a performance model from
those personal trait models.” Decl., para. 5. We find this unpersuasive,
because Thissen-Roe discloses the use of a neural network (| 117) to create
traits models (1 88) that are fed into a predictive model (190), thus, meeting
the claim language.
We are also unpersuaded by the Declaration’s assertion that its author
is “unaware” of any competitors “using a neural network” for the claimed
steps. Deck, para. 6. Although this may be true, we have no evidence in the
record other than the assertion to use in evaluating the statement.
Ultimately, this is unimportant, however, because the rejection of claim 1 is
based instead on the disclosures of Thissen-Roe and Rosner, rather than on
what methods were in use in commerce by the Appellants’ competitors.
For these reasons, we sustain the rejection of independent claims 1
and 34—37, as well as dependent claims 2 and 9-15 that were not separately
argued.
9
Appeal 2015-002797
Application 13/107,176
Rejection of Claims 16, 18—23, and 27—33 under 35 U.S.C. f 103(a)
The Appellants generally assert the language of claim 16 has specific
benefits, but do not provide any further analysis. Appeal Br. 18—19. A
general allegation that the art does not teach any of the claim limitations is
no more than merely pointing out the claim limitations. A statement which
merely points out what a claim recites will not be considered an argument
for separate patentability of the claim. 37 C.F.R. § 41.37(c)(l)(vii).
The Appellants also generally assert that Thissen-Roe does not
disclose claim language similar to that recited in independent claim 1.
Appeal Br. 18—19. Our analysis is the same as set forth above for
independent claim 1, and need not be repeated here.
For this reason, we sustain the rejection of independent claim 16 and
claims 18—23 and 27—33 that depend from claim 16, and which are not
separately argued.
Rejection of Claim 17 under 35 U.S.C. § 103(a)
We are not persuaded by the Appellants’ argument that Scarborough
teaches away from “wherein the set of questions is developed by an
industrial psychologist independently of the generation of the models and
the performance model,” because Scarborough discloses disadvantages of
using psychologist-provided questions, apparently citing to column 2, lines
1—10 of Scarborough. Appeal Br. 19.
Scarborough continues that “a study can be conducted to verify
whether the factors chosen by the psychologist have been successful in
identifying suitable applicants,” but that such studies are expensive.
Scarborough col. 2,11. 11—18. We are unpersuaded that “expense” here is a
10
Appeal 2015-002797
Application 13/107,176
disadvantage sufficient to constitute a teaching away. Indeed, instead, we
discern that it actually discloses that the aforementioned claim limitation is
known, albeit expensive. Moreover, amongst various embodiment
examples, Scarborough discloses, without reservation: “[w]hen selecting
new questions, it may be advantageous to employ the services of an
industrial psychologist who can evaluate the job and determine appropriate
job skills. The psychologist can then determine an appropriate question to
be asked to identify a person who will fit the job.” Id. at col. 10,11. 38-42.
“A reference may be said to teach away when a person of ordinary
skill, upon reading the reference, would be discouraged from following the
path set out in the reference, or would be led in a direction divergent from
the path that was taken by the applicant.” In re Kahn, 441 F.3d 977, 990
(Fed. Cir. 2006) (citations and internal quotation marks omitted).
Scarborough, thus, discloses concerns with using psychologists as a source
of questions, but provides methods to overcome those cautions, and relies on
psychologists in some examples without reservation.
For this reason, we sustain the rejection of claim 17.
Rejections of Claims 3—6, 24, 38, and 40—42 under 35 U.S.C. § 103(a)
The rejections of these dependent claims were not argued separately
with any specificity. We therefore sustain the rejections of claims 3—6, 24,
38, and 40-42.
11
Appeal 2015-002797
Application 13/107,176
DECISION
We REVERSE the rejection of claims 1—6, 9—24, 27—38, and 40-42
under 35 U.S.C. § 101.
We REVERSE rejection of claims 1 and 16 under 35 U.S.C. § 112,
first paragraph.
We AFFIRM the rejections of claims 1—6, 9—24, 27—38, and 40-42
under 35 U.S.C. § 103(a).
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a)(l)(iv).
AFFIRMED
12