Facebook, Inc.

Patent Trials and Appeals Board

Aug 31, 2020

2020002802 (P.T.A.B. Aug. 31, 2020)

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.
15/476,725 03/31/2017 Ido Wakrat 108282.021380 5060
169063 7590 08/31/2020
BakerHostetler / Facebook
Cira Centre 12th Floor
2929 Arch Street
Philadelphia, PA 19104-2891
EXAMINER
WHITAKER, ANDREW B
ART UNIT PAPER NUMBER
3629
NOTIFICATION DATE DELIVERY MODE
08/31/2020 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):
eofficemonitor@bakerlaw.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte IDO WAKRAT, TAO CUI BAECKLUND,
JUAN FELIPE COIROLO TESTA, and DMITRY KOTEROV

Appeal 2020-002802
Application 15/476,725
Technology Center 3600
Before ST. JOHN COURTENAY III, LARRY J. HUME, and
PHILLIP A. BENNETT, Administrative Patent Judges.
BENNETT, Administrative Patent Judge.
DECISION ON APPEAL
STATEMENT OF THE CASE
Pursuant to 35 U.S.C. § 134(a), Appellant1 appeals from the
Examiner’s decision to reject claims 1, 5–11, and 15–20, which are all of the
claims pending in this application. Claims 2–4 and 12–14 have been
cancelled. We have jurisdiction under 35 U.S.C. § 6(b).
We AFFIRM.

1 We use the word “Appellant” to refer to “applicant” as defined in
37 C.F.R. § 1.42(a). Appellant identifies the real party in interest as
Facebook, Inc. Appeal Br. 2.
Appeal 2020-002802
Application 15/476,725

2
CLAIMED SUBJECT MATTER
The claims are directed to customizing content items by filtering a
matrix of temporal values. Spec. ¶ 1. Claim 1, reproduced below, is
illustrative of the claimed subject matter:
1. A method comprising:
receiving, by a processor of an online system from a third
party system, (i) a matrix of temporal values for a plurality of
lodgings, each temporal value of the matrix of temporal values
corresponding to at least a start date and an end date and (ii) one
or more parameters describing the plurality of lodgings;
determining, by the processor of the online system, a target
start date and a target end date for a user of the online system
responsive to at least one action performed by the user;
determining, by the processor of the online system, a
subset of temporal values by filtering the matrix of temporal
values for temporal values having a corresponding start date and
corresponding end date matching the target start date and the
target end date, respectively;
determining, by the processor of the online system, a
subset of lodgings of the plurality of lodgings, wherein each
lodging of the subset of lodgings corresponds to at least one of
the subset of temporal values;
retrieving, from a data store of the online system, actions
performed by the user on the online system or the third party
system;
for each lodging of the subset of lodgings, generating, by
the processor of the online system, a likelihood that the user will
select the lodging by:
extracting feature vectors from the retrieved actions
performed by the user and the corresponding one or more
parameters of the lodging for a machine learning model trained
by:
Appeal 2020-002802
Application 15/476,725

3
deriving feature vectors from various content items and
from actions performed by a plurality of other users of the online
system with the content items, and
training the machine learning model by using the derived
feature vectors, and
inputting the extracted feature vectors into the machine
learning model, the machine learning model outputting the
likelihood that the user will select the lodging;
selecting, by the processor of the online system, a lodging
of the plurality of lodgings based on the generated likelihoods;
and
providing a content item for presentation on a client device
of the user, wherein the content item includes the corresponding
temporal value of the subset of temporal values of the selected
lodging.
Appeal Br. 17–18 (Claims App.).
REFERENCES
The prior art relied upon by the Examiner as evidence is:
Name Reference Date
Krishnareddy et al. US 2013/0006777 Al Jan. 3, 2013
Fishberg US 2016/0225108 Al Aug. 4, 2016
Zhang et al. US 2017/0048664 Al Feb. 16, 2017
REJECTIONS2
Claims 1, 5, 7–11, 15, and 17–20 stand rejected under 35 U.S.C. § 103
as being unpatentable over Fishberg and Zhang. Final Act. 8.
Claims 6 and 16 stand rejected under 35 U.S.C. § 103 as being
unpatentable over Fishberg, Zhang and Krishnareddy. Final Act. 13.

2 The Examiner has withdrawn the rejection of claims 1, 5–11, and 15–20
under 35 U.S.C. § 101. Ans. 3.
Appeal 2020-002802
Application 15/476,725

4
ISSUE
Issue: Has the Examiner erred in finding Fishberg and Zhang teach or
suggest:
for each lodging of the subset of lodgings, generating, by the
processor of the online system, a likelihood that the user will
select the lodging by:
extracting feature vectors from the retrieved actions performed
by the user and the corresponding one or more parameters of the
lodging for a machine learning model trained by:
deriving feature vectors from various content items and from
actions performed by a plurality of other users of the online
system with the content items, and training the machine learning
model by using the derived feature vectors,
as recited in claim 1?
ANALYSIS
The Examiner rejects claim 1 as obvious over Fishberg and Zhang.
The Examiner finds that Fishberg teaches most of the limitations of claim 1,
but states that
[Fishberg] does not expressly teach for each lodging of the subset
of lodgings, generating, by the processor of the online system, a
likelihood that the user will select the lodging by: extracting
feature vectors from the retrieved actions performed by the user
and the corresponding one or more parameters of the lodging for
a machine learning model trained by: deriving feature vectors
from various content items and from actions performed by a
plurality of other users of the online system with the content
items, and training the machine learning model by using the
derived feature vectors, and inputting the extracted feature
vectors into the machine learning model, the machine learning
model outputting the likelihood that the user will select the
lodging; selecting, by the processor of the online system, a
lodging of the plurality of lodgings based on the generated
likelihoods.
Appeal 2020-002802
Application 15/476,725

5
Final Act. 10. The Examiner cites Zhang to address this deficiency. Id. at
10–11 (citing Zhang ¶¶ 51–59, 80–81).
Appellant asserts that Zhang’s teachings are deficient because:
[T]he claimed machine learning model is also different from the
coefficient taught in Zhang because the claimed model is trained
based on feature vectors derived from various content items.
This allows for a more robust model for use in the particular
application that relates to serving the most temporally-relevant
content about lodgings to users. Zhang does not teach or suggest
the step of (or the sub-steps under) generating a likelihood of a
user selecting a lodging by using a machine learning model
trained based on ‘feature vectors from various content items and
from actions performed by a plurality of other users of the online
system with the content items,’ as recited in claim 1.
Appeal Br. 15. Furthermore, Appellant argues “[t]he machine learning
model in claim 1 is trained based on feature vectors derived from various
content items and actions performed by other users with the content items.
In contrast, Zhang's models are generated merely based on the target user’s
information (e.g., the user looking for lodgings) but not based on other
users’ information and other content items and actions on those other items.”
Appeal Br. 14 (emphasis omitted).
We are not persuaded by Appellant’s arguments because we agree
with the Examiner that Zhang teaches or at least suggests the disputed
limitations. We emphasize that the standard for determining whether a
claim is obvious is “an expansive and flexible approach.” KSR International
Co. v. Teleflex, Inc., 550 U.S. 398, 415 (2007). Zhang’s vector or model that
is created based on collected information of what may be of interest to the
user (¶ 59) combined with Zhang’s social networking system 160
determining coefficients using machine-learning algorithms trained on
historical actions and past user responses, or data farmed from users by
Appeal 2020-002802
Application 15/476,725

6
exposing them to various options and measuring responses (Zhang ¶ 81),
teaches or at least suggests the limitation “deriving feature vectors from
various content items and from actions performed by a plurality of other
users of the online system with the content items, and training the machine
learning model by using the derived feature vectors.” That is, we find
Zhang’s users teach or at least suggest “a plurality of other users” as the
source of “various content items” as recited in the disputed limitation of
claim 1. We agree with the Examiner that in light of the teachings of Zhang,
a person of ordinary skill in the art would have found it obvious to derive
feature vectors from actions performed by a plurality of other users of the
online system with the content items. Ans. 4–5.
We, therefore, disagree with Appellant’s argument that the cited
references fail to teach or otherwise the disputed limitation and we sustain
the rejection of claim 1. Appellant does not present separate arguments on
independent claim 11 or dependent claims 5, 7–10, 14, 15, and 17–20.
Arguments not made are waived. Accordingly, we affirm the Examiner’s
obviousness rejection of claims 5, 7–10, 14, 15, and 17–20.
Appellant does not present arguments on the Examiner’s obviousness
rejection of claims 6 and 16 over the combination of Fishberg, Zhang, and
Krishnareddy and we affirm the Examiner’s rejection of these claims also.
CONCLUSION
We affirm the Examiner’s rejections under 35 U.S.C. § 103.

Appeal 2020-002802
Application 15/476,725

7
DECISION SUMMARY
Claims
Rejected
35 U.S.C.
§
Reference(s)/Basis Affirmed Reversed
1, 5, 7–11,
15, 17–20
103 Fishberg, Zhang 1, 5, 7–11,
15, 17–20

6, 16 103 Fishberg, Zhang,
Krishnareddy
6, 16
Overall
Outcome
1, 5–11,
15–20

TIME PERIOD FOR RESPONSE
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a)(1)(iv). See 37
C.F.R. § 41.50(f).
AFFIRMED