Ex Parte Lamere et al

Patent Trial and Appeal Board

Feb 18, 2015

12338596 (P.T.A.B. Feb. 18, 2015)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________

Ex parte PAUL B. LAMERE, STEPHEN J. GREEN,
JEFFREY H. ALEXANDER, FRANCOIS MAILLET,
and DOUGLAS ECK
____________

Appeal 2012-011096
Application 12/338,596
Technology Center 2100
____________

Before DENISE M. POTHIER, ERIC B. CHEN, and IRVIN E. BRANCH,
Administrative Patent Judges.

POTHIER, Administrative Patent Judge.

DECISION ON APPEAL
STATEMENT OF THE CASE
Appellants appeal under 35 U.S.C. § 134(a) from the Examiner’s
rejection of claims 1–20. App. Br. 4.1 We have jurisdiction under 35 U.S.C.
§ 6(b). We reverse.

1 Throughout this opinion, we refer to the Appeal Brief (App. Br.) filed
March 19, 2012; (2) the Examiner’s Answer (Ans.) mailed May 24, 2012;
and (3) the Reply Brief (Reply Br.) filed July 24, 2012.
Appeal 2012-011096
Application 12/338,596

2
Invention
Appellants’ invention relates to a process of a user dynamically
interacting with a recommender, which receives user inputs through a user
interface and steers recommendations towards more relevant and
personalized content. The recommender may receive an input that specifies
a tag cloud, which may include one or more tags and weights for the tags.
See Spec. ¶¶ 3–4. Claim 1 is reproduced below with the key disputed
limitations emphasized:
1. A computer-implemented method, comprising:
displaying a tag cloud in response to user input
specifying the tag cloud, wherein the tag cloud comprises one
or more tags, wherein each tag specifies a descriptive word or
phrase;
determining a set of recommended items from a
collection of items in response to the user input, wherein the set
of recommended items is determined according to similarity of
tag clouds associated with the items in the collection to the
displayed tag cloud, wherein the set of recommended items
comprises items for which a determined similarity value is
above a similarity threshold; and
displaying the set of recommended items.

The Examiner relies on the following as evidence of unpatentability:
Chakra US 2009/0287989 A1 Nov. 19, 2009
(filed May 16, 2008)
Kirby US 2010/0114907 A1 May 6, 2010
(filed Oct. 31, 2008)

THE REJECTION
The Examiner rejected claims 1–20 under 35 U.S.C. § 103(a) as
unpatentable over Chakra and Kirby. Ans. 5–10.

Appeal 2012-011096
Application 12/338,596

3
THE CONTENTIONS
The Examiner finds that Chakra teaches the step of “determining a set
of recommended items from a collection of items” according to the
similarity of tag clouds associated with the items in the collection to the
displayed tag cloud as recited in claim 1. Ans. 5–6 (citing Chakra ¶¶ 25–27;
Fig. 4). Specifically, the Examiner finds Figure 4 of Chakra shows a set of
recommended items (e.g., flowers, museum, restaurants, and Eiffel Tower)
that are determined according to similarity of tag clouds to the displayed tag
cloud (e.g., tag manager assigns weights and priorities to different categories
and uses this information to determine most relevant tags to the user based
on the user input). Ans. 5, 13 (citing Chakra ¶¶ 11–12, 22).
Among other things, Appellants argue that “the initial and subsequent
tag clouds in Chakra are based on a current user profile, and not based on a
similarity of tag clouds to each other.” App. Br. 10 (underlining omitted).
Given that Appellants contend Chakra lacks this feature, Appellants assert
that Chakra fails to teach the recited step of determining the recommended
set of items “according to similarity of tag clouds associated with the items
in the collection to the displayed tag cloud” as recited. App. Br. 9–11.

ISSUE
(1) Under § 103, has the Examiner erred in rejecting claim 1 by
finding that Chakra and Kirby would have taught or suggested “determining
a set of recommended items from a collection of items in response to the
user input, wherein the set of recommended items is determined according to
similarity of tag clouds associated with the items in the collection to the
displayed tag cloud”?
Appeal 2012-011096
Application 12/338,596

4
ANALYSIS
Based on the record before us, we find error in the Examiner’s
rejection of claim 1. We begin by construing the recited phrase, “tag cloud.”
The disclosure states that the “tag cloud may include one or more tags and
weights for the tags” and tags are “descriptive words or phrases.” Spec.
¶¶ 4, 6. Also, the prior art, Chakra, states “[t]ag clouds typically contain a
set of related tags, where the tags can be text, such as keywords, that
describe web content.” Chakra ¶ 12. However, neither of these documents
defines a tag cloud.
Appellants urge that a tag cloud conveys “information beyond what
can be conveyed if the keywords in the tag cloud are simply considered a
collection of keywords[,]” including “how and where keywords are
displayed.” Reply Br. 2. However, Appellants fail to provide evidence to
support this position. Given the record and giving the recited phrase its
broadest reasonable interpretation in light of the disclosure as it would be
interpreted by an ordinarily skilled artisan, a tag cloud can include a single
tag and weight for the tag. Additionally, given this understanding, the
recited step of “determining a set of recommended items . . . according to
similarity of tag clouds . . . to the displayed tag cloud” includes the set being
determined according to similarities of tag clouds, each having a single tag
(e.g., descriptive word) and weight, associated with the collection items to
the displayed tag cloud having a single tag and weight.
Turning to the cited art, Chakra teaches that each tag cloud has a tag,
and some tags are assigned weights indicating their popularities, which can
be accomplished by displaying more popular tags with a striking color or
bigger font (e.g., “EIFFEL TOWER” is larger and more bold than
Appeal 2012-011096
Application 12/338,596

5
“RESTAURANTS”). Chakra ¶¶ 12, 25; Fig. 4. As such, Chakra teaches tag
clouds (e.g., a single tag, such as Eiffel Tower, assigned a weight indicated
by its font size) as construed broadly, but reasonably.
The Examiner acknowledges that Chakra teaches the tag cloud can be
based on a user profile but further finds Chakra’s tag clouds can also be
based on the “similarity of tag clouds.” Ans. 13 (citing Chakra ¶ 22).
Chakra teaches a content manager and a tag manager work together given a
request to determine user’s interests, find relevant tags, and display the tags
in a tag cloud. Chakra ¶ 22; Ans. 13. Thus, Chakra at least suggests to an
ordinarily skilled artisan determining a set of recommended items (e.g.,
relevant tags) for display based on a user’s interests. However, neither
Chakra nor the Examiner explains how “user interests” are analogous to “tag
clouds,” such that the set of recommended items is determined “according to
similarity of tag clouds . . . to the displayed tag cloud” as recited in claim 1.
Chakra also teaches receiving a user request when a tag manager has
no access to a user profile (e.g., block 202 to block 203) or receiving an
input to modify a user profile (e.g., block 208 to block 203) temporarily.
Chakra ¶ 27; Fig. 2. Here, the user of Chakra indicates his or her interests
by clicking on boxes or buttons, entering keywords, selecting from a drop
down menu, and/or indicating the amount of emphasis (e.g., weight) on each
interest category (e.g., finance, marketing, software test, astrology). Chakra
¶ 27; Fig. 3. Based on the user interests and each category’s weight,
relevant tags are selected (e.g., block 206). Chakra ¶¶ 27–28; Fig. 2.
Accordingly, Chakra teaches or suggests determining a set of recommended
items (e.g., relevant tags) at least in part according to a category’s weight.
Also, because a category includes a descriptive word (e.g., finance or
Appeal 2012-011096
Application 12/338,596

6
astrology) and the category has a weight, Chakra teaches determining a set
of recommended items according to tag clouds. Id. Additionally, given that
the tags are relevant to the category and its weight, the tag clouds are
associated with collection items as recited. Id.
From these teachings in Chakra, the Examiner extrapolates that
Chakra demonstrates the recommended items are determined according to
the similarity of tag clouds to the displayed tag cloud. Ans. 13.
Specifically, the Examiner finds that tag clouds that contain a related tag set
“is inherited [sic] to similarity ( i.e., ‘related tags’ in the ‘tag clouds’)” and
that paragraph 22 in Chakra teaches finding “relevant tags’ in other tag
clouds as known by a skill [sic] artisan in searching information (e.g., search
keywords, query, etc.).” Ans. 13. As best understood, we disagree.
As discussed in Chakra, one skilled in the art would have recognized
that the related tags are related to or have some type of similarity to tag
clouds (e.g., descriptive words or categories with assigned weights). See
Chakra ¶ 22. However, these teachings fail to discuss or suggest
determining related tags (e.g., recommended items) according to the
similarity of these tag clouds to the displayed tag cloud (e.g., tag cloud
shown in Fig. 4) as required by claim 1. Moreover, the Examiner has not
explain adequately how any incidental similarities between the tag clouds
discussed in Chakra and the displayed tag cloud are used to determine the
recommended items recited in claim 1. See id. Given the findings presented
in the record, we find that the Examiner has failed to explain how Chakra
teaches the recited phrase, “determining a set of recommended items from a
collection of items in response to the user input, wherein the set of
recommended items is determined according to similarity of tag clouds
Appeal 2012-011096
Application 12/338,596

7
associated with the items in the collection to the displayed tag cloud” in
claim 1.
Lastly, the Examiner does not rely on Kirby to cure the
above-noted deficiencies. See Ans. 6–7.
Independent claims 10 and 16 include similar limitations to the
above-discussed recitation of claim 1. Compare App. Br. 21 with App. Br.
23, 25. Accordingly, for the above-noted reasons, we are also persuaded of
error for the rejection presented for claims 10 and 16.
For the foregoing reasons, Appellants have persuaded us of error in
the rejection of: (1) independent claim 1; (2) independent claims 10 and 16,
which recite commensurate limitations; and (3) dependent claims 2–9,
11–15, and 17–20 for similar reasons.

CONCLUSION
The Examiner erred in rejecting claims 1–20 under § 103.

DECISION
The Examiner’s decision rejecting claims 1–20 is reversed.

REVERSED

msc