Ex Parte Chen et al

Patent Trial and Appeal Board

Jul 30, 2013

10626875 (P.T.A.B. Jul. 30, 2013)

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.
10/626,875 07/25/2003 Francine R. Chen A3053USNP-XER1564US01 1845
61962 7590 07/31/2013
FAY SHARPE LLP / XEROX - PARC
1228 EUCLID AVENUE, 5TH FLOOR
THE HALLE BUILDING
CLEVELAND, OH 44115
EXAMINER
BURKE, JEFF A
ART UNIT PAPER NUMBER
2159
MAIL DATE DELIVERY MODE
07/31/2013 PAPER
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.
PTOL-90A (Rev. 04/07)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________________

Ex parte FRANCINE R. CHEN, AYMAN O. FARAHAT and
THORSTEN H. BRANTS
____________________

Appeal 2010-005592
Application 10/626,875
Technology Center 2100
____________________

Before ALLEN R. MacDONALD, ROBERT E. NAPPI and
KRISTEN L. DROESCH, Administrative Patent Judges.

MacDONALD, Administrative Patent Judge.

DECISION ON APPEAL

Appeal 2010-005592
Application 10/626,875

2
STATEMENT OF CASE
Introduction
Appellants appeal under 35 U.S.C. § 134(a) from a rejection of claims
1-88. We have jurisdiction under 35 U.S.C. § 6(b).
Exemplary Claims
Exemplary claims 1 and 77 under appeal read as follows (emphasis
added):
1. A computer-implemented method of determining predictive
models for a linked event detection system comprising the steps of:
determining source-identified training stories;
determining inter-story similarity vectors in a memory for at
least one story-pair of the source-identified training stories;
determining link label information for the at least one story-
pair, the link label information indicating the existence of at least one
link between a pair of stories in the source-identified training stories
and that the linked source-identified stories are related to the same
event; and
determining and storing at least one predictive model in the
memory based on the inter-story similarity vectors and the link label
information.
77. A method of determining a stopword list comprising the
steps of:
determining a source-identified training corpus of text
information;
determining a verified first source-mode transformation of the
source-identified training corpus text from a first mode to a second
mode based on at least one of a verified transcription and a verified
translation;
determining an un-verified second source-mode transformation
of the source-identified training corpus text from a first mode to a
second mode;
Appeal 2010-005592
Application 10/626,875

3
determining at least one transformation error associated with
distribution differences between the first and second transformations
and identified sources;
determining and storing at least one source-specific
transformation action for the determined transformation errors in a
memory; and
identifying and transforming transformation errors in other
transformed source-identified texts based on the source-specific
transformation actions in the memory.
Rejections on Appeal
The Examiner rejected claims 77-80 under 35 U.S.C. § 101 as being
directed to non-statutory subject matter. (Non-final Rej. 4-5).
1

The Examiner rejected claims 77-80 under 35 U.S.C. § 103(a) as
being unpatentable over the combination of Wayne (“Topic Detection and
Tracking in English and Chinese”) and Brown (“Dynamic Stopwording for
Story Link Detection”). (Ans. 31-34).
2

The Examiner rejected claims 1-76 and 81-88 under 35 U.S.C.
§ 103(a) as being unpatentable over either the combination of Sundaresan
(US 6,606,620 B1), Pirolli (US 5,835,905), and Maybury (US 6,961,954
B1); or these references in further combination with Gange
(US 2004/0006559 Al) or Zhou (US 2004/0002849 Al). (Ans. 6-31).

1
As to this rejection, separate patentability is not argued for claims 78-80.
Except for our ultimate decision, the rejection of these claims is not
discussed further herein.
2
As to this rejection, separate patentability is not argued for claims 78-80.
Except for our ultimate decision, the rejection of these claims is not
discussed further herein.
Appeal 2010-005592
Application 10/626,875

4
Appellants’ Contentions
1. Appellants contend that the Examiner erred in rejecting claims
1-76 and 81-88 under § 103(a) because “Neither Sundaresan Nor [the]
Combination of References Teach Use of Source Identification.” (App. Br.
15-17).
2. Appellants contend that the Examiner erred in rejecting claims
1-76 and 81-88 under § 103(a) because the “Neither Maybury Nor [the]
Combination of References Teach Indicating Existence of Stories Linked to
the Same Event.” (App. Br. 17-18).
3

3. Appellants contend that the Examiner erred in rejecting claim
77 under § 103(a) because:
Also with regard to claim 77, the Examiner's Answer
(page 37) maintains the 35 U.S.C. §103(a) rejection by stating
that "nowhere in the claim language does the claim require the
transformation to occur in the 'actual' text or the 'body' of the
text" and further that it is enough "that the transformations
merely occur, it is irrelevant where they occur according to the
claim." The Appellant respectfully disagrees, and points to the
claim language that identifies and transforms transformation
errors in other transformed source-identified texts.
(Reply Br. 6).
4. Appellants contend that the Examiner erred in rejecting claim
77 under § 103(a) because:
[T]he cited portion of Brown explicitly teaches away from this
transforming source-identified texts:
Each test document then has the weights of the
stopwords for the cluster containing it set to zero
in its term vector prior to computing the cosine
similarity, effectively removing the stopwords

3
Appellants’ remaining contentions as to dependent claims are not
reproduced herein as our decision renders these arguments moot.
Appeal 2010-005592
Application 10/626,875

5
without actually altering the document stored in
memory. (Brown, page 2, column 1, lines 2-6;
emphasis added).
Thus, the cited portion of Brown never actually transforms
transformation errors in transformed source-identified texts
based on source-specific transformation actions in memory.
Additionally, Wayne does not transform transformation errors
in source-identified texts either. Therefore, the combination of
Wayne and Brown does not address each and every limitation
of independent claim 77.
(App. Br. 23-24).
5. Appellants also contend that the Examiner erred in rejecting
claim 77 under § 103(a) because:
Independent claim 77 recites a method of determining a
stopword list that includes identifying and transforming
transformation errors in transformed source-identified texts
based on source-specific transformation actions in memory. The
process of transforming of transformation errors in transformed
source-identified texts transforms particular articles (source-
identified texts), and thus is statutory under § 101. As
acknowledged by the Interim Examination Instructions for
Evaluating Subject Matter Eligibility Under 35 U.S.C. §101 of
August 24, 2009:
An article can be electronic data that represents a
physical object or substance. ... The data should be
more than an abstract value. Data can be
specifically identified by indicating what the data
represents, the particular type or nature of the data,
and/or how or from where the data was obtained.
(page 5, next-to-last paragraph)[now MPEP § 2106
II.B.1.(b)(second full paragraph); 8th Ed., Rev. 9].
The source-identified texts recited in independent claim
77 are considered articles because they are electronic data that
represents a physical text, such as an [sic] news article or
story.
(App. Br. 22)(emphasis added).
Appeal 2010-005592
Application 10/626,875

6
6. Appellants also contend that the Examiner erred in rejecting
claim 77 under § 101 because:
Here, with respect to claim 77, a particular article (source-
identified texts) is transformed into a different state or thing
through the process of "transforming transformation errors in
other transformed source-identified texts based on the source-
specific transformation actions in the memory." The source-
identified texts are not merely abstract data. The source-
identified texts represent a story or article, which is used to
communicate information to a reader. Thus, they are articles
within the scope of In re Bilski. Furthermore, the articles are
transformed to a different state through the process of
correcting errors in the articles. Accordingly, claim 77 is
appropriate subject matter under 35 U.S.C. §101.
(Reply Br. 6).

ANALYSIS
We have reviewed the Examiner’s rejections in light of Appellants’
arguments that the Examiner has erred.
As to Appellants’ above contention 1, we disagree for the reasons set
forth by the Examiner at pages 33-35 of the Answer.
As to above contention 2, we agree with Appellants. This renders
moot Appellants’ remaining arguments as to claims 1-76 and 81-88.
As to Appellants’ above contention 3, we agree with Appellants. This
renders moot Appellants’ contention 4.
As to Appellants’ above contentions 5 and 6, we disagree. Contrary to
the foundation of Appellants’ argument, “text” is not “physical.” Text is not
a tangible article or commodity. Cf. In re Nuijten, 500 F.3d 1346, 1356
(Fed. Cir. 2007). Rather, Appellants’ text is merely the printed matter on the
physical substrate (i.e., document). Appellants data in memory is not a data
Appeal 2010-005592
Application 10/626,875

7
representation of the document (e.g., a scanned image of the physical
document and its abstract text information), but rather a data representation
of only the abstract information printed on the document. As such we find
no basis for Appellants to argue that the claimed transformation of text is
transformation of an article which renders the claim patentable under 35
U.S.C. §101.
Additionally, unlike claim 1, Appellants’ claim 77 is not limited to a
“computer implemented method” and instead encompasses performance of
the “determining”, and “identifying and transforming” steps as mental
activity by a human being. Even the act of using “a memory” as the storage
for the text information encompasses storage in a human memory. Even if
claim 77 were construed as requiring the memory to be some tangible
machine recording medium, the essence of claim 77 would encompass a
method using mental activity alone with insignificant “extra-solution”
activity in the form of storage in a recording medium.

CONCLUSIONS
(1) Appellants have established that the Examiner erred in rejecting
claims 1-76 and 81-88 as being unpatentable under 35 U.S.C. § 103(a).
(2) Appellants have established that the Examiner erred in rejecting
claims 77-80 under 35 U.S.C. § 103(a).
(3) The Examiner did not err in rejecting claims 77-80 under
35 U.S.C. § 101.
Appeal 2010-005592
Application 10/626,875

8
DECISION
The Examiner’s rejections of claims 1-88 under 35 U.S.C. § 103(a)
are reversed.

The Examiner’s rejection of claims 77-80 under 35 U.S.C. § 101 are
affirmed.

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).

AFFIRMED-IN-PART

ELD