Henkin, Assaf et al.

Patent Trials and Appeals Board

Dec 16, 2019

13234679 - (D) (P.T.A.B. Dec. 16, 2019)

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.
13/234,679 09/16/2011 Assaf Henkin 4487.033US1 8991
21186 7590 12/16/2019
SCHWEGMAN LUNDBERG & WOESSNER, P.A.
P.O. BOX 2938
MINNEAPOLIS, MN 55402
EXAMINER
MACASIANO, MARILYN G
ART UNIT PAPER NUMBER
3688
NOTIFICATION DATE DELIVERY MODE
12/16/2019 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):
SLW@blackhillsip.com
uspto@slwip.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
____________________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________________

Ex parte ASSAF HENKIN, ITAI BRICKNER, and
STANISLAV KRICHEVSKY
__________________

Appeal 2019-000484
Application 13/234,679
Technology Center 3600
____________________

Before JAMES P. CALVE, MICHAEL J. FITZPATRICK, and
PAUL J. KORNICZKY, Administrative Patent Judges.

CALVE, Administrative Patent Judge.

DECISION ON APPEAL

STATEMENT OF THE CASE
Pursuant to 35 U.S.C. § 134(a), Appellant1 appeals from the decision
of the Examiner to reject claims 1, 4–14, and 17–30. We have jurisdiction
under 35 U.S.C. § 6(b).
We REVERSE.

1 “Appellant” refers to “applicant” as defined in 37 C.F.R. § 1.42. Appellant
identifies Amobee, Inc. as the real party in interest. Appeal Br. 2.
Appeal 2019-000484
Application 13/234,679

2
CLAIMED SUBJECT MATTER
Claims 1 and 14 are independent, with claim 1 reproduced below with
disputed limitations highlighted in italics.
1. A method comprising:
receiving, at a server computer, text from a client computing
device, the text representing a user’s web search query
that has been submitted to a search engine during a web
browsing session of the user;
accessing a target web page that the user has requested to view
subsequent to the submitting of the user’s web search
query;
before communicating the target web page for presentation on
the client computing device, selecting a content item in
ordinary text of the target web page based on a
combination of a first related score, a second related
score, and a third related score, the first related score
representing a relatedness of the target web page to the
content item, the second related score representing a
relatedness of the content item to advertising content, the
third related score representing a relatedness of the
target web page to the advertising content;
selecting the advertising content based on the selecting of the
content item;
modifying the target web page in real-time to convert the
content item in the web page to an interactive link such
that when the interactive link is invoked by the client
computing device, the client computing device presents
the advertising content along with the web page; and
communicating to the modified target web page to client
computing device for presentation on the client
computing device.

REJECTION
Claims 1, 4–14, and 17–30 are rejected under 35 U.S.C. § 103(a) as
unpatentable over Sravanapudi (US 2007/0174255 A1, pub. July 26, 2007)
and Kraft (US 7,603,349 B1, iss. Oct. 13, 2009).
Appeal 2019-000484
Application 13/234,679

3
ANALYSIS
Independent Claims 1 and 14
The Examiner finds that Sravanapudi identifies content related to
input content (a uniform resource locator, text from a web page), creates
hyperlinks (e.g., “in-line” text ads) to receive content, extracts text from a
search query or web page, and analyzes the text to select ads related to the
extracted text, as recited in claims 1 and 14, but Sravanapudi lacks “before
communicating the target web page for presentation on the client computing
device, selecting a content item in ordinary text of the target web page based
on a combination of a first related score, a second related score, and a third
related score” and modifying the target web page to convert the content item
to an interactive link. Final Act. 3–8. The Examiner relies on Kraft to teach
these features as pop-up elements that display additional information with a
query entry area, links to a page, and a query response module that enhances
search results with additional information such as links and/or advertising
copy obtained from a sponsor content database. Id. at 5–7; Ans. 3–4.
Appellant argues that Kraft does not teach “selecting a content item in
ordinary text of a web page for modification based on a combination of three
relatedness scores,” or the details of the three relatedness scores recited in
claims 1 and 14. Appeal Br. 15. Appellant argues that Kraft’s enhancement
of search results with additional information such as links and/or advertising
copy obtained from a sponsored content database does not teach or suggest
first, second, and third scores or selecting a content item in ordinary text of a
target web page based on a combination of those scores. Id. Appellant also
argues that there is no citation to Kraft for a “third related score representing
a relatedness of the target web page to the advertising content.” Id.
Appeal 2019-000484
Application 13/234,679

4
In response, the Examiner asserts that Kraft’s “query response module
may enhance search result information (from terms search from content
vector) with additional information such as links and/or advertising copy
(first and second scores) obtained from a sponsor content database.” Ans.
4. The Examiner finds that Kraft does so by displaying web pages with links
to hit pages, sites, or advertising from a sponsor content database. Id. at 3–5.
The Patent Laws set forth the standard for obviousness as follows:
A patent may not be obtained though the invention is not
identically disclosed or described as set forth in section 102, if
the differences between the subject matter sought to be patented
and the prior art are such that the subject matter as a whole
would have been obvious at the time the invention was made to
a person having ordinary skill in the art to which said subject
matter pertains.

35 U.S.C. § 103(a). Obviousness is a question of law based on fact findings
as to (1) the scope and content of the prior art, (2) differences between the
prior art and the claims at issue, (3) the level of ordinary skill in the pertinent
art, and (4) secondary indicia of obviousness or nonobviousness. Graham v.
John Deere Co. of Kansas City, 383 U.S. 1, 17–18 (1966). “[W]hen
evaluating claims for obviousness under 35 U.S.C. § 103, all the limitations
of the claims must be considered and given weight.” In re Gardner, 449 F.
App’x 914, 916 (Fed. Cir. 2011) (non-precedential) (citing Stratoflex, Inc. v.
Aeroquip Corp., 713 F.2d 1530 (Fed.Cir.1983)); see In re Glatt Air Techs.,
Inc., 630 F.3d 1026, 1030 (Fed. Cir. 2011) (holding obviousness was not
established where the prior art failed to teach the claimed shielding); In re
Mettke, 570 F.3d 1356, 1361 (Fed. Cir. 2009) (finding prima facie case of
obviousness where prior art taught every element of the claim performing
the same function as in the claim with a reason to combine the elements).
Appeal 2019-000484
Application 13/234,679

5
We agree with the Examiner that Kraft teaches “a first related score
representing a relatedness of the target web page to the content item” as
claimed. As part of this feature, Kraft teaches “a content item in ordinary
text of the target web page” as words of interest. Kraft provides an example
of such a content item using the word “Radio” 1910 in the phrase “Sirius
Satellite Radio” in Figure 19 (item 1910). Kraft, 26:37–50. Kraft describes
this feature as created by a text selection tool called a “term extrapolator.”
Id. Kraft teaches that a user selects this content item via an interactive link.
The user then clicks on a desired location (e.g., a word of
interest) in a page 1908 that is being displayed in browser
window 1906. For example, the user may click on the word
“Radio” at 1910. An approximately circular region 1912
around the selected word is highlighted and used to derive
context. A pop-up menu 1914 also appears, offering various
choices to the user, such as related searches, sponsored content,
and the like.

Id. at 26:43–50. The pop-up menu includes links to pages with information
about people and companies based on a “context vector” that is determined
from text contained in the web page and “sponsored content” related to the
subject of the article, which Kraft teaches elsewhere includes advertising
content. Id. at 26:51–60, 8:59–62 (query response module 164 may enhance
search result information with additional information such as links and/or
advertising copy obtained from sponsored content database 162).
The context vector is created for a web page based on keywords and
terms (words or phrases) that appear in web page content with a frequency
count of how many times the term occurs in the content. Id. at 9:1–15. The
context vector provides a first related score of the relatedness of the target
web page to the content item (word/phrase) contained in the web page.
Appeal 2019-000484
Application 13/234,679

6
Kraft also teaches that search providers maintain a database of web
pages that associate the URL of each page with keywords, category data,
and the like reflecting the content of the web page and then use algorithms to
rank (score) the relatedness of each page to a query based on frequency of
occurrence of keywords and other information. Id. at 1:38–63. This score
also relates a content item to a target web page as recited for the first score.
The context vector also provides additional information with the list
of terms from the web page content to specify each term’s frequency, which
reflects the number of times the term occurs in content 302 of the web page
as well as document frequency, weight, and term type. Kraft 10:35–46. The
context vector thus provides a first related score of the “content item” (word
or phrase) in a target web page to the target web page as claimed.
We also agree with the Examiner that Kraft teaches a “second related
score representing a relatedness of the content item to advertising content”
by associating sponsored content (e.g., advertising) with keywords extracted
from a received context vector, where the items of sponsored content are
selected from a sponsored content database 162 based on those keywords.
Kraft, 22:30–46. The selection of sponsored content items can be based on
“the number or importance of the keywords matched by each item.” Id. at
22:40–45. The sponsored/special content items also include shopping and
advertising information. Id. at 22:65. A “Shop in Context” content feature
provides links and text in web page content (Fig. 14) that allows users to buy
merchandise related to the subject of a news story 602, e.g., about Bono. Id.
at 21:43–63; Ans. 3–5. Kraft thus provides a second related score that
represents the relatedness of the content item (extracted from web page text
by a context vector) to sponsored advertising content as claimed.
Appeal 2019-000484
Application 13/234,679

7
However, the Examiner has not provided any findings to explain how
Kraft teaches a “third related score representing a relatedness of the target
web page to the advertising content” as claimed. Appellant’s Specification
discloses that the similarity in a score of a web page and a score of a content
item such as an advertisement may be used to match these two items. Spec.
¶ 45. Content enhancement server 12 can select a word/phrase in a web
page that relates to a topic of the web page (first related score) and to an
advertisement (second related score), and the word or phrase may hyperlink
the two pages. See id.
Because Kraft scores the relatedness of a content item to a web page
and to an advertisement, comparing the first and second scores indicates the
relatedness of a target web page and advertisement. See Kraft, 4:6–17 (pop-
up element links to a page with content related to the search context), 8:5–67
(web pages are ranked by the occurrence of search terms in a search query
and context identifiers, and keywords are used to trigger the display of
sponsored content (links or ads)); 9:5–35 (special content selected for web
page based on context vector for that page); 22:19–52 (web page augmented
with sponsored content associated with keywords extracted from context
vector for that web page).
For the foregoing reasons, we agree with the Examiner’s findings that
Kraft teaches first and second related scores. Ans. 4. However, because the
Examiner does not address whether Kraft also teaches a third related score
(see Ans. 4), we do not sustain the rejection of independent claims 1 and 14
or their dependent claims 4–13 and 17–30.

Appeal 2019-000484
Application 13/234,679

8
CONCLUSION
Claims
Rejected
35 U.S.C.
§
Reference(s)/Basis Affirmed Reversed
1, 4–14, 17–
30
103(a) Sravanapudi, Kraft

1, 4–14, 17–
30
Overall
Outcome
1, 4–14, 17–
30
REVERSED