Ex Parte Lu et al

Patent Trial and Appeal Board

Feb 27, 2013

11256411 (P.T.A.B. Feb. 27, 2013)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________________

Ex parte LIE LU, WEI-YING MA and
ZHIWEI LI
____________________

Appeal 2010-009168
Application 11/256,411
Technology Center 2100
____________________

Before ROBERT E. NAPPI, HUNG H. BUI, and LYNNE E. PETTIGREW,
Administrative Patent Judges.

BUI, Administrative Patent Judge.

DECISION ON APPEAL
Appellants1 seek our review under 35 U.S.C. § 134(a) of the
Examiner’s final rejections of claims 1-20. We have jurisdiction under 35
U.S.C. § 6(b).
We AFFIRM-IN-PART.2

1 Real Party in Interest is Microsoft Corporation.
2 Our decision refers to Appellants’ Appeal Brief filed October 15,
2009 (“App. Br.”); Reply Brief filed April 5, 2010 (“Reply Br.”);
Examiner’s Answer mailed February 4, 2010 (“Ans.”); Final Office Action
mailed April 21, 2008 (“FOA.”); and the original Specification filed October
21, 2005 (“Spec.”).
Appeal 2010-009168
Application 11/256,411

2

STATEMENT OF THE CASE
Appellants’ Invention
According to Appellants, their invention relates to an automated
presentation of a semantic topic by way of searching multimodal
information (e.g., image, text, audio and/or video) associated with the
semantic topic from a database or the Internet, via a user interface layout of
a storyboard. See Appellants’ Spec., ¶ [0003], and Abstract.

Claims on Appeal
Claims 1, 13 and 18 are independent claims. Claim 1 is representative
of the invention, as reproduced below with disputed limitations emphasized:

1. A computer-implemented method comprising:

determining a semantic topic;

evaluating respective portions of multimodal information
corresponding to the semantic topic to identify events, each
event being associated with one or more of person, time,
location, and keyword;

for each document in the respective portion, calculating
probability that the document belongs to an event of the
events based on a generative model and document distribution
along a time line associated with the event;

for each event in least a subset of the events:

objectively identifying one or more representative
documents that are of greater relevance to the event as
compared to other documents;

Appeal 2010-009168
Application 11/256,411

3

extracting other media corresponding to the
representative documents from the multimodal
information, the representative documents and the other
media being objectively most representative of the
semantic topic; and

wherein the one or more representative documents and
the other media are for presentation to a user in a storyboard.

Evidence Considered
The prior art relied upon by the Examiner in rejecting the claims on
appeal is:
Julien U.S. 2002/0129011 A1 Sep. 12, 2002
Buinevicius U.S. 2004/0093349 A1 May 13, 2004

Microsoft PowerPoint 2003 as described in
http://office.microsoft.com/training/Training.aspx?AssetID=RC01 0713231
033&CTT=6&Origin=RC010713231033, Microsoft Power Point 2003
released in 2003.

Examiner’s Rejections
(1) Claims 1-8 and 13-16 stand rejected under 35 U.S.C. § 102(b)
as being anticipated by Julien. Ans. 3-9.
(2) Claims 9-12 and 17 stand rejected under 35 U.S.C. § 103(a) as
being unpatentable over Julien and Power Point 2003. Ans. 10-13.
(3) Claims 18-19 stand rejected under 35 U.S.C. § 103(a) as being
unpatentable over Julien and Buinevicius. Ans. 14-15.
(4) Claim 20 stands rejected under 35 U.S.C. § 103(a) as being
unpatentable over Julien, Buinevicius and Power Point 2003. Ans. 15-16.

Appeal 2010-009168
Application 11/256,411

4

ISSUES
Under § 102(b), the dispositive issue on appeal is whether Julien
discloses the feature “for each document in the respective portion,
calculating probability that the document belongs to an event of the events
based on a generative model and document distribution along a time line
associated with the event,” as recited in independent claims 1 and 13. App.
Br. 7-21; Reply Br. 3-7.
Under § 103(a), the dispositive issue on appeal is whether the
combination of Julien and Buinevicius discloses or suggests the feature
“identifying, from multi-modal data sources, representative multi-modal
content of a semantic topic, the representative multi-modal content
comprising one or more of images, news article summaries, video, and
locations corresponding to one or more salient events associated with the
semantic topic,” as recited in independent claim 18. App. Br. 22-24; Reply
Br. 7-8.

ANALYSIS
We have reviewed the Examiner’s rejections in light of Appellants’
arguments that the Examiner has erred. Only those arguments actually made
by Appellant in the Appeal Brief have been considered. See 37 C.F.R.
§ 41.37(c)(1)(vii).

§ 102(b) Rejection of Claims 1-8 and 13-16
Appellants contend that Julien does not disclose the feature “for each
document in the respective portion, calculating probability that the document
belongs to an event of the events based on a generative model and document
Appeal 2010-009168
Application 11/256,411

5

distribution along a time line associated with the event,” as recited in
independent claims 1 and 13. App. Br. 7-21; Reply Br. 3-7. In particular,
Appellants argue that:
(1) Julien describes a system for automatically finding
and aggregating contact information from web pages . . . . [It]
allows a user to specify a business and in response receive
corresponding contact information listed on the business's
website . . . . [T]he identification unit of Julien determines Web
pages associated with a particular URL "that are most likely to
contain contact information."
(App. Br. 14) (emphasis added);

(2) Fig. 3 [of Julien] illustrates a process for searching
for contact information. As shown above, this process includes:
(1) a user submitting a URL address; (2) the identification unit
establishing a connection with the submitted URL; (3) the
identification unit identifying contact information on the
retrieved Web page(s); (4) the extractor unit extracting the
contact information; (5) the aggregator unit compiling a list of
the extracted contact information, and (6) the aggregator unit
transmitting the list to the user . . . . [T]his process fails to
relate to any sort of “document distribution along a timeline.”
(App. Br. 18-19); and

(3) Julien teaches time-stamping of the contact
information that is mapped to the URL address of the Web
pages.
(App. Br. 19).

According to Appellants, Julien simply fails to disclose or suggest
“calculating probability that the document belongs to an event of the events”
based on (1) a “generative model” and (2) “document distribution along a
time line associated with the event,” as recited in independent claims 1 and
13. Reply Br. 6-7.
Appeal 2010-009168
Application 11/256,411

6

In response thereto, the Examiner has taken a broad interpretation of
the disputed limitations of claims 1 and 13. For example, the Examiner has
broadly interpreted “calculating probability that the document belongs to an
event of the events” as encompassing the identification unit 30 of Julien to
examine data contained in each Web page connected to the URL address and
tag in each Web page any information elements relevant to contact
information. Ans. 18 (citing ¶ [0066] of Julien). In addition, the Examiner
has broadly interpreted Appellants’ claimed two conditions for such a
calculation, i.e., (1) “generative model” as encompassing the same
identification unit 30 of Julien to determine the probability of Web pages
containing contact information and treat a set of Web pages as randomly
generated data, and (2) “document distribution along a time line associated
with the event” as encompassing time-stamping of contact information that
is mapped to the URL addresses of the Web pages. Ans. 18-19 (citing
¶ [0014] and ¶ [0073] of Julien).
However, the Examiner’s broad interpretation is unreasonable and
inconsistent with the language of the claims and Appellants’ Specification.
We agree with Appellants that neither the identification unit 30 as described
in ¶ [0014] and ¶ [0073] of Julien nor Fig. 3 of Julien performs any
probability calculation ... based on (1) a “generative model” and (2)
“document distribution along a time line associated with the event,” as
recited in independent claims 1 and 13. Nor does the timestamp associated
with contact information of Julien meet Appellants’ claimed “document
distribution along a time line associated with the event.” App. Br. 15-16;
Reply Br. 3-7.
Appeal 2010-009168
Application 11/256,411

7

For the reasons set forth above, Appellants have persuaded us of error
in the Examiner’s rejection of independent claims 1 and 13. Therefore, we
cannot sustain the Examiner’s rejection of claims 1 and 13 and their
respective dependent claims 2-8 and 14-16 under 35 U.S.C. § 102(b) as
being anticipated by Julien.

§ 103(a) Rejection of Claims 18-20
Appellants contend that the combination of Julien and Buinevicius
does not disclose or suggest the feature “identifying, from multi-modal data
sources, representative multi-modal content of a semantic topic, the
representative multi-modal content comprising one or more of images, news
article summaries, video, and locations corresponding to one or more salient
events associated with the semantic topic,” as recited in independent claim
18. App. Br. 22-23; Reply Br. 7-8. In particular, Appellants solely argue:
the Julien system merely attempts to find particular web pages
associated with the user-specified URL or keyword(s) that are
most likely to contain the requested contact information. As
such, Julien fails to teach or suggest “identifying ...
representative multi-modal content of a semantic topic ...
corresponding to one or more salient events associated with the
semantic topic.” In fact, Julien fails to relate to “salient events
associated with a semantic topic” at all. Again, Julien simply
retrieves Web pages that contain contact information.

App. Br. 23 (emphasis added).
We are not persuaded by Appellants’ sole argument. As expressly
acknowledged by Appellants’ own Specification:
a semantic topic may be one or more keywords (e.g., input by a
user as part of a search query, etc.) representing one or more
events, a person's name, or anything else. For example,
Appeal 2010-009168
Application 11/256,411

8

respective semantic topics include the “World Cup 2002”,
“USA election”, “Halloween”, “Harry Potter”, etc. In some
cases, a semantic topic may represent a target topic and an
event. . . . For example, “Halloween” can be both a semantic
topic and an event.
Spec. ¶ [0013].

The Examiner finds that ¶ [0014] and ¶ [0020] of Julien describe the
identification of multi-modal content of a semantic topic from multi-modal
data sources. Ans. 14. Appellants have not contested, and we see no reason
to disturb the Examiner’s specific findings regarding Julien.
Moreover, we agree with the Examiner that, contrary to Appellants’
argument, Julien discloses that “multi-modal information is extracted to
retrieve information that is the most representative of the semantic topic . . . .
[D]ocument sources that represent a semantic topic are examined and only
those documents are identified that correspond to one or more salient events
(contact information) associated with the semantic topic.” Ans. 20-21.
For the reasons set forth above, Appellants have not persuaded us of
error in the Examiner’s rejection of claim 18 as well as claims 19-20, which
were not separately argued, under 35 U.SC § 103(a).

CONCLUSION
On the record before us, we conclude that the Examiner has erred in
rejecting claims 1-8 and l3-l6 under 35 U.S.C. § 102(b) and their respective
dependent claims 9-12 and 17 under 35 U.S.C. § 103(a). We also conclude
that the Examiner has not erred in rejecting claims 18-20 under 35 U.S.C.
§ 103(a).

Appeal 2010-009168
Application 11/256,411

9

DECISION
As such, we REVERSE the Examiner’s final rejections of (1) claims
1-8 and 13-16 under 35 U.S.C. § 102(b) as being anticipated by Julien; and
(2) claims 9-12 and 17 under 35 U.S.C. § 103(a) as being unpatentable over
Julien and Power Point 2003
However, we AFFIRM the Examiner’s final rejections of (3) claims
18-19 under 35 U.S.C. § 103(a) as being unpatentable over Julien and
Buinevicius; and (4) claim 20 under 35 U.S.C. § 103(a) as being
unpatentable over Julien, Buinevicius and Power Point 2003.
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) (2011).

AFFIRMED-IN-PART

ELD