Google Inc.

Patent Trials and Appeals Board

Dec 9, 2020

2019004940 (P.T.A.B. Dec. 9, 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.
13/830,526 03/14/2013 John R. Provine 16113-4276001 5684
26192 7590 12/09/2020
FISH & RICHARDSON P.C.
PO BOX 1022
MINNEAPOLIS, MN 55440-1022
EXAMINER
HEFFERN, JAMES E
ART UNIT PAPER NUMBER
2158
NOTIFICATION DATE DELIVERY MODE
12/09/2020 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):
PATDOCTC@fr.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
____________________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________________

Ex parte JOHN R. PROVINE
____________________

Appeal 2019-004940
Application 13/830,526
Technology Center 2100
____________________

Before JEREMY J. CURCURI, JUSTIN BUSCH, and
ROBERT J. SILVERMAN, Administrative Patent Judges.

BUSCH, Administrative Patent Judge.

DECISION ON APPEAL
Pursuant to 35 U.S.C. § 134(a), Appellant1 appeals from the
Examiner’s decision to reject claims 1–20, 22, and 23, which constitute all
the claims pending. We have jurisdiction over the pending claims under
35 U.S.C. § 6(b).
We reverse.

1 We use the term “Appellant” to refer to “applicant” as defined in 37 C.F.R.
§ 1.42. Appellant identifies the real party in interest as Google LLC.
Appeal Br. 1.
Appeal 2019-004940
Application 13/830,526

2
STATEMENT OF THE CASE
Introduction
The invention generally relates to determining whether to display
data, such as “knowledge panels,” along with search results. Spec. 1:7–8,
5:16–17. “A knowledge panel is a user interface element that provides a
collection of information or other content related to a particular entity
referenced by a search query.” Spec. 5:17–18. Claims 1, 11, and 18 are
independent claims. Claim 1 is reproduced below:
1. A method performed by data processing apparatus,
the method comprising:
receiving a query from a client device;
obtaining search results that are responsive to the received
query;
identifying two different factual entities that each
correspond to the received query and that are both identified
using a same word;
determining, for each of the two identified factual entities,
a topicality score that specifies a measure of topical relatedness
between the factual entity and content presented by resources
linked to by the obtained search results;
selecting, from among the two different factual entities, a
particular factual entity for which a knowledge panel is to be
provided with the obtained search results based, at least in part,
on the topicality score for each of the two factual entities, the
knowledge panel for the particular factual entity including a
plurality of different types of content for the particular factual
entity, the plurality of different types of content including at least
one content item, for the particular factual entity, received from
a first resource and at least one content item, for the particular
factual entity, received from a second resource different than the
first resource;
determining that the knowledge panel for the particular
factual entity is to be provided with the obtained search results,
the determination being based, at least in part, on content of the
knowledge panel and characteristics of the obtained search
Appeal 2019-004940
Application 13/830,526

3
results, the characteristics including a performance of one or
more resources referenced by the identified search results when
search results that reference the one or more resources were
provided in response to the received query; and
providing, to the client device, data that presents the
knowledge panel at the client device.
The Pending Rejections
Claims 1, 2, 6, 9–12, 15, 18, 19, 22, and 23 stand rejected under
35 U.S.C. § 103 as obvious in view of Azzam (US 2008/0027888 A1; Jan.
31, 2008), Donmez (US 2013/0159272 A1; June 20, 2013), Baron (US
2010/0076972 A1; Mar. 25, 2010), and Berntson (US 2012/0059838 A1;
Mar. 8, 2012). Final Act. 3–39, 42–44.
Claims 3–5, 13, 14, and 20 stand rejected under 35 U.S.C. § 103 as
obvious in view of Azzam, Donmez, Baron, Berntson, and Ando (US
2005/0096897 A1; May 5, 2005). Final Act. 39–42.
Claims 7, 8, 16, and 17 stand rejected under 35 U.S.C. § 103 as
obvious in view of Azzam, Donmez, Baron, Berntson, and Vinay (US
2014/0059042 A1; Feb. 27, 2014). Final Act. 45–48.
ANALYSIS
The Examiner finds the combination of Azzam, Donmez, Baron, and
Berntson teaches or suggests every limitation recited in independent claims
1, 11, and 18. Final Act. 3–33. Of particular relevance to this Appeal, the
Examiner finds Azzam teaches or suggests “determining that the knowledge
panel . . . is to be provided . . . based, at least in part, on . . . a performance of
one or more resources referenced by the identified search results when
search results that reference the one or more resources were provided in
response to the received query” (the “disputed limitation”), as recited in
Appeal 2019-004940
Application 13/830,526

4
independent claims 1, 11, and 18. Final Act. 4 (citing Azzam ¶¶ 25–27, 37,
40, 41, Fig. 6, Table 1), 14, 24.
Specifically, the Examiner finds Azzam assigns scores to factual
descriptions from multiple documents based on linguistic constituents and
semantics and, if a particular factual description’s score exceeds a threshold,
presents the factual description in a presentation operation. Final Act. 4, 14,
24; Ans. 7–8 (finding that, because Azzam’s fact extraction tool recognizes
factual descriptions in documents based on determining linguistic
constituents of the factual descriptions, Azzam teaches the disputed
limitation). The Examiner finds Azzam “associate[s] a performance of the
resources” because Azzam teaches associating words in a table with an
appropriate part of speech “to provide optimal processing performance.”
Ans. 4 (quoting Azzam ¶ 27) (emphasis omitted). The Examiner finds that
Azzam’s processing performance teaches the recited “performance of one or
more resources referenced by the identified search results” and Appellant
argues limitations not recited in the claims. Ans. 6 (citing Azzam ¶¶ 27, 37).
The Examiner finds Azzam’s list of extracted facts provided with the search
results teaches the recited “when search results that reference the one or
more resources were provided in response to the received query.” Ans. 7
(citing Azzam, Fig. 6).
Appellant argues the Examiner dissects the disputed limitation to find
that Azzam teaches discrete portions of the disputed limitation in isolation,
but the rejection fails to address the disputed limitation as a whole. Reply
Br. 1–2.
Among other arguments, Appellant asserts Azzam’s processing
performance “is a measure of performance of a ‘fact extraction tool’ that
Appeal 2019-004940
Application 13/830,526

5
uses the ‘table’ to ‘find words or phrases’ in a document that matches ‘those
of the fact-word-table,’” which does not teach or suggest the performance of
a resource when a search result referencing the resource is provided. Appeal
Br. 9–10. Appellant also argues the cited “processing performance” does
not teach the disputed limitation because Azzam’s processing performance is
not used to determine anything, let alone determine that a knowledge panel
is to be provided, as recited in the claims. Appeal Br. 10. Appellant
contends Azzam cannot teach using the processing performance to
determine anything because, although Azzam mentions optimizing
performance, Azzam never actually evaluates or determines the performance
such that the performance could be a condition for determining any other
action—namely providing a knowledge panel. Appeal Br. 10.
Appellant also asserts the Examiner’s findings—that (1) Azzam’s
cited processing performance teaches the performance portion of the
disputed limitation and (2) Azzam’s list of extracted facts teaches “when
search results that reference the one or more resources were provided in
response to the received query”—further demonstrate that the Examiner
improperly dissects the disputed limitation. In particular, Appellant argues
the Examiner’s finding that the extracted facts teach the latter half of the
disputed limitation ignores the entire first portion of the disputed
limitation—i.e., that the claim conditions display of its knowledge panel on
determining that the “performance of one or more resources referenced by
the identified search results.” Reply Br. 1–2. Moreover, Appellant asserts
that Azzam does not determine whether to provide section 608 (the set of
extracted facts that Azzam presents with the search result), which the
Examiner finds teaches the knowledge panel, see Final Act. 4 (citing Azzam
Appeal 2019-004940
Application 13/830,526

6
¶ 41, Fig. 6), based on the performance of resources referenced by Azzam’s
“list of documents 604,” which the Examiner finds teaches the resources
referenced by the identified search results, see Final Act. 4 (“The extraction
tool attempts to recognize a set of factual descriptions (factual entities) from
multiple documents (first and second resources),” (emphasis added)).
We agree with Appellant. Even accepting the Examiner’s findings
that Azzam’s extracted factual descriptions teach or suggest the recited
factual entities and Azzam’s multiple documents teach or suggest the recited
one or more resources, the rejection is problematic. First, it is unclear from
the Examiner’s findings and explanation what disclosures in Azzam teach or
suggest the disputed language. “When a reference is complex or shows or
describes inventions other than that claimed by the applicant, the particular
part relied on must be designated as nearly as practicable” and “[t]he
pertinence of each reference, if not apparent, must be clearly explained and
each rejected claim specified.” 37 C.F.R. § 1.104(c)(2) (emphases added).
An agency is bound by its own regulations. See Service v. Dulles, 354 U.S.
363, 388 (1957). Specifically, in reviewing the Examiner’s findings and
explanations, we see nothing in Azzam’s cited disclosures that teaches
basing any decision on a performance of any resource, let alone the specific
language recited in the disputed limitation. Nevertheless, we address the
two disclosed features in Azzam that it appears the Examiner may find teach
or suggest the disputed limitation.
To the extent the rejection depends on a finding that Azzam’s
disclosed optimal processing performance, see Azzam ¶ 27 (disclosing that
associating words in a fact-word table with a most appropriate part of speech
provides optimal processing performance), teaches the disputed limitation,
Appeal 2019-004940
Application 13/830,526

7
we disagree. As Appellant argues, Azzam merely describes how to provide
optimal processing performance of the fact extraction process, but does not
actually teach or suggest determining some level of actual processing
performance. Accordingly, without determining the processing
performance, Azzam necessarily fails to teach or suggest basing a
determination to provide a knowledge panel (i.e., Azzam’s extracted set of
factual descriptions) on the processing performance.
To explain our interpretation of the possible alternative finding
regarding the recited performance, we first summarize the findings with
respect to the related limitations. The Examiner finds that: 1) Azzam’s
“electronic documents of interest” teach or suggest the recited “one or more
resources”; 2) Azzam’s “factual descriptions [extracted] from the electronic
documents of interest” teach or suggest the recited “factual entities”;
3) Azzam’s “various web site links 604 available from the Internet” teach or
suggest the recited “search results that are responsive to the received query”;
and 4) Azzam’s paragraphs 25 through 27, 37, 39, 40, 41, and Figure 6 teach
or suggest the disputed limitation. Final Act. 3–4.
Given these findings, we presume the Examiner implicitly finds,
notwithstanding failing to make such a finding clearly or explicitly, that
Azzam’s “total score” assigned to a particular factual description teaches or
suggests the recited “performance of one or more resources referenced by
the identified search results” when such search results were provided in
response to the query. See Final Act. 4 (explaining that the extraction tool
identifies factual entities exceeding a threshold “based on a determination of
linguistic constituents and semantics” (citing Azzam ¶ 37)); Azzam ¶¶ 36–
37 (describing rules for assigning scores to candidate factual descriptions
Appeal 2019-004940
Application 13/830,526

8
and comparing a determined “total score” for a factual description to a
threshold to determine whether to present or discard the factual description).
Even assuming this finding, however, we still find the rejection problematic.
Specifically, the claim recites using the “performance of one or more
resources referenced by the identified search results” as a basis for providing
the knowledge panel (Azzam’s set of factual descriptions according to the
Examiner’s mapping). Even to the extent that a total score may be
considered some measure of a “performance,” Azzam assigns a total score to
a particular factual description, not to the electronic documents, which is
what the Examiner finds teaches or suggests the recited “one or more
resources.” Moreover, we disagree that Azzam’s total score, which indicates
the likelihood that a particular factual description is a fact rather than an
opinion, teaches or suggests the “performance” of that factual description
(let alone of a resource). On this record, we fail to see how identifying
whether content in a document is a fact (rather than an opinion) based on
linguistic analysis teaches the performance of that fact in the sense of the
disputed limitation—i.e., the performance of the factual description “when
search results that reference the [factual description] were provided in
response to the received query.”
For these reasons, on this record, we do not sustain the rejection of
independent claims 1, 11, and 18, which recite the disputed limitation as
obvious over Azzam, Donmez, Baron, and Berntson. Nor do we sustain the
rejection of claims 2, 6, 9, 10, 12, 15, 19, 22, and 23, which depend
therefrom and incorporate the disputed limitation, as obvious over Azzam,
Donmez, Baron, and Berntson. The Examiner does not find either Ando or
Vinay cures this deficiency. Therefore, we do not sustain the rejection of
Appeal 2019-004940
Application 13/830,526

9
claims 3–5, 13, 14, and 20 as obvious over Azzam, Donmez, Baron,
Berntson, and Ando or the rejection of claims 7, 8, 16, and 17 as obvious
over Azzam, Donmez, Baron, Berntson, and Vinay because each of these
claims depend ultimately from one of claims 1, 11, and 18 and, therefore,
incorporate the disputed limitation. Because we reverse the rejection of the
independent claims and claim 23, we need not address Appellant’s other
arguments.
DECISION SUMMARY
Claims
Rejected
35 U.S.C. § References Affirmed Reversed
1, 2, 6, 9–
12, 15, 18,
19, 22, 23
103 Azzam, Donmez,
Baron, Berntson
1, 2, 6, 9–
12, 15, 18,
19, 22, 23
3–5, 13, 14,
20
103 Azzam, Donmez,
Baron, Berntson, Ando
3–5, 13, 14,
20
7, 8, 16, 17 103 Azzam, Donmez,
Baron, Berntson, Vinay
7, 8, 16, 17
Overall
Outcome
1–20, 22,
23

REVERSED