Facebook, Inc.

Patent Trials and Appeals Board

Oct 19, 2021

2020005733 (P.T.A.B. Oct. 19, 2021)

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.
15/793,893 10/25/2017 Salahuddin Choudhary 26295-37882/US 9832
87851 7590 10/19/2021
Facebook/Fenwick
Silicon Valley Center
801 California Street
Mountain View, CA 94041
EXAMINER
FAN, HUA
ART UNIT PAPER NUMBER
2449
NOTIFICATION DATE DELIVERY MODE
10/19/2021 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):
fwfacebookpatents@fenwick.com
ptoc@fenwick.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________

Ex parte SALAHUDDIN CHOUDHARY, YI XIAO, and
MIKHAIL LARIONOV

Appeal 2020-005733
Application 15/793,893
Technology Center 2400
____________

Before JAMES R. HUGHES, NORMAN H. BEAMER,
and SCOTT B. HOWARD, Administrative Patent Judges.

BEAMER, Administrative Patent Judge.

DECISION ON APPEAL
Appellant appeals under 35 U.S.C. § 134(a) from the Examiner’s final
rejection of claims 1–6, 8–13, 15–19.1 Claims 7, 14, and 20 are cancelled.2
We have jurisdiction over the pending rejected claims under 35 U.S.C.
§ 6(b).
We REVERSE.

1 We use the word “Appellant” to refer to “applicant” as defined in
37 C.F.R. § 1.42. Appellant identifies Facebook, Inc. as the real party in
interest. (Appeal Br. 2.)
2 The Final Office Action incorrectly also lists cancelled claim 14 as
rejected. (Appeal Br. 6, n.1.)
Appeal 2020-005733
Application 15/793,893

2

THE INVENTION
Appellant’s disclosed and claimed invention is directed to a
messaging system that applies a filtering policy to a message, which has an
associated message tag describing the content of the message, the filtering
policy selectively blocking messages, and allowing a message having an
associated tag to pass through the filter even though the message might
otherwise violate the filtering policy. (Abstr.)
Independent claim 1 is representative:
1. A method comprising:
receiving a message sent by a sender to a recipient using an electronic
messaging system;
identifying a tag associated with the message, the tag describing
content in a body of the message;
analyzing the message using a trained tag model, the trained tag
model operative to receive the message and determine, based at
least in part on one or more features associated with the
message, whether the content of the message conforms with
requirements associated with the tag;
applying a tag-based filtering policy, the tag-based filtering policy
indicating an action for the messaging system to perform on the
message based on the tag and a result of the analysis; and
performing the action indicated by the filtering policy on the message.
(Appeal Br. 13 (Claim App.))
REJECTIONS
The Examiner rejected claims 1, 3, 8, 10, 15, and 17 under 35 U.S.C.
§ 103 as being unpatentable over Wexelblat (US 7,222,158 B2, iss. May 22,
2007) and Gobburu, et al. (US 2007/0117552 A1, pub. May 24, 2007).
(Final Act. 3–5.)
Appeal 2020-005733
Application 15/793,893

3

The Examiner rejected claims 2, 9, and 16 under 35 U.S.C. § 103 as
being unpatentable over Wexelblat, Gobburu, and in view of official notice.
(Final Act. 5–6; Ans. 6.)
The Examiner rejected claims 4, 11, and 18 under 35 U.S.C. § 103 as
being unpatentable over Wexelblat, Gobburu, and Kamma (US
2014/0337413 A1, pub. Nov. 13, 2014. (Final Act. 6; Ans. 6.)
The Examiner rejected claims 4, 5, 11, 12, and 18 under 35 U.S.C.
§ 103 as being unpatentable over Wexelblat, Gobburu, and Gannu, et al. (US
8,166,161 B1, iss. Apr. 24, 2012). (Final Act. 7–8; Ans. 7.)
The Examiner rejected claims 6, 13, and 19 under 35 U.S.C. § 103 as
being unpatentable over Wexelblat, Gobburu, Gannu, and in view of official
notice. (Final Act. 7–8; Ans. 9.)
ISSUE ON APPEAL
Appellant’s arguments present the following dispositive issue:3
Whether the Examiner erred in finding that the combination of
Wexelblat and Gobburu taught or suggested the independent claim 1
limitation, “analyzing the message using a trained tag model, the trained tag
model operative to receive the message and determine, based at least in part
on one or more features associated with the message, whether the content of
the message conforms with requirements associated with the tag,” and the

3 Rather than reiterate the arguments of Appellant and the legal conclusions
and findings of the Examiner, we refer to the Appeal Brief, filed Apr. 8,
2020 (“Appeal. Br.”); the Reply Brief, filed Aug. 4, 2020 (“Reply Br.”); the
Final Office Action, mailed Sept. 17, 2019 (“Final Act.”); and the
Examiner’s Answer, mailed June 9, 2020 (“Ans.”) for the respective details.
Appeal 2020-005733
Application 15/793,893

4

commensurate limitations in independent claims 8 and 15. (Appeal Br. 4–
7.)
ANALYSIS
In rejecting the independent claims over the combination of Wexelblat
and Gobburu, the Examiner relies solely on Wexelblat, determining that,
“checking black-list and/or white-list and updating blacklist and white-list
using sender’s identify is equivalent to a trained tag model.” (Final Act. 4.)
The Examiner’s Answer elaborates:
[C]hecking of black-list and/or white-list and then updating the
black-list and white-list using sender’s identify [sic]
accordingly is equivalent to training a tag model wherein the
sender ID is considered a tag. This is because an ordinary
skilled in the art [sic] would understand that a black-list and/or
white-list containing sender IDs (tag) is equivalent to a tag
model, used, e.g., for filtering messages, and that updating such
a black-list/white list (tag model) based on newly processed
sender IDs is equivalent to training such a tag model.
(Ans. 11.)
Appellant argues that the handling of black-lists and white-lists
disclosed in Wexelblat is not related to training a tag model. (Appeal Br. 6.)
Appellant relies on the ordinary meaning of “training,” and on the
explanation in the Specification of training a tag model involving “using
‘supervised learning’ in which a ‘messaging learning module 330 uses the
results of the analyses conducted by the analysis module 320 to build
training sets containing examples of correctly and incorrectly tagged
messages and associated information about the messages.’” (Reply Br. 3
(citing Spec. ¶ 52).)
We agree with Appellant. The portions of Wexelblat relied on by the
Examiner do not relate to training a tag model. Moreover, the sender ID of a
Appeal 2020-005733
Application 15/793,893

5

message is not related to the tag of a message, which the claims define as
“describing content in a body of the message.” In addition, nothing in the
cited portions of Wexelblat discloses an analysis of “whether the content of
the message conforms with requirements associated with the tag,” as
required by the claims.
Accordingly, we do not sustain the Examiner’s rejection of
independent claims 1, 8, and 15 as obvious over Wexelblat and Gobburu. In
rejecting the dependent claims, the Examiner relied only on Wexelblat for
the claim limitation at issue (Final Act. 6–8), and therefore those rejections
are not sustained for the same reasons as set forth above.
DECISION SUMMARY
In summary:
Claim(s)
Rejected
35 U.S.C. § Reference(s)/Basis Affirmed Reversed
1, 3, 8,
10, 15, 17
103 Wexelblat, Gobburu 1, 3, 8,
10, 15, 17
2, 9, 16 103 Wexelblat, Gobburu, 2, 9, 16
4, 11, 18 103 Wexelblat, Gobburu,
Kamma
4, 11, 18
4, 5, 11,
12, 18
103 Wexelblat, Gobburu,
Gannu
4, 5, 11,
12, 18
6, 13, 19 103 Wexelblat, Gobburu,
Gannu
6, 13, 19
Overall
Outcome
1–6, 8–
13, 15–19

REVERSED