Ex Parte Kasravi et al

Patent Trial and Appeal Board

Jun 7, 2016

12511547 (P.T.A.B. Jun. 7, 2016)

UNITED STA TES p A TENT AND TRADEMARK OFFICE
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR
12/511,547 0712912009 Kas Kasravi
56436 7590 06/09/2016
Hewlett Packard Enterprise
3404 E. Harmony Road
Mail Stop 79
Fort Collins, CO 80528
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
ATTORNEY DOCKET NO. CONFIRMATION NO.
82257800 4025
EXAMINER
PELLETT, DANIEL T
ART UNIT PAPER NUMBER
2122
NOTIFICATION DATE DELIVERY MODE
06/09/2016 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):
hpe.ip.mail@hpe.com
mkraft@hpe.com
chris.mania@hpe.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte KAS KASRA VI and MARIE RISOV
Appeal2015-001112
Application 12/511,547
Technology Center 2100
Before DEBRA K. STEPHENS, KEVIN C. TROCK, and
JESSICA C. KAISER, Administrative Patent Judges.
KAISER, Administrative Patent Judge.
DECISION ON APPEAL
Introduction
Appellants 1 appeal under 35 U.S.C. § 134(a) from a final rejection of
claims 1---6, 8-13, 15-17, and 19. We have jurisdiction under 35 U.S.C.
§ 6(b ).
We affirm.
1 According to Appellants, the real party in interest is Hewlett-Packard
Development Company, LP. (App. Br. 1.)
Appeal2015-001112
Application 12/511,547
EXEMPLARY CLAIM
Claim 1, reproduced below, is illustrative of the claimed subject
matter with disputed limitations emphasized:
1. A method comprising:
extracting, by a processor, plural characteristics of text
documents, wherein at least one of the text documents is a target
patent, wherein the plural characteristics include citations to or
from other text documents and at least one other characteristic,
and wherein each of the text documents is associated with a
corresponding collection of the plural characteristics;
creating vectorized representations of the characteristics
of the target patent and the corresponding collection of the plural
characteristics of each of the text documents, wherein each
element of the vector quantifies a single characteristic including
at least one concept derived from content of the target patent;
comparing a vector distance between concepts of the
target patent and collection of concepts and other characteristics
of text documents without performing any comparisons between
any text documents of a subset of the text documents, wherein the
subset includes all of the text documents except the target patent;
determining similarity measures based on vector distance
comparison of vectorized concepts such that a shorter distance
indicates a stronger similarity; and
identifj;ing which of the text documents are relevant to the
target patent while not requiring any of the text documents to be
the same by generating, by the processor, an output using the
collections of the plural characteristics, wherein the output
indicates similarities among at least a subset of the text
documents and the target patent based on the similarity measures.
REJECTIONS
The Examiner made the following rejections:
2
Appeal2015-001112
Application 12/511,547
Claims 1---6, 8-13, 15, 17, and 19 stand rejected under 35 U.S.C.
§ 103(a) as being unpatentable over Uchida (Hideyuki Uchida et al., Patent
Map Generation Using Concept-Based Vector Space Model, NTCIR-4 (June
2004)) and Jiang (US 2008/0162165 Al; published July 3, 2008). (Final
Act. 2-15.)
Claim 16 stands rejected under 35 U.S.C. § 103(a) as being
unpatentable over Uchida, Jiang, and Barney (US 7,716,226 B2; issued May
11, 2010). (Final Act. 15-16.)
ISSUES
Issue 1: Did the Examiner err in finding the combination of Uchida
and Jiang teaches or suggests "identifying which of the text documents are
relevant to the target patent while not requiring any of the text documents to
be the same," as recited in claim 1 and similarly recited in claims 10 and 17?
Issue 2: Did the Examiner err in finding Uchida teaches or suggests
comparing "without performing any comparisons between any text
documents of a subset of the text documents, wherein the subset includes all
of the text documents except the target patent," as recited in claim 1 and
similarly recited in claims 10 and 17?
Issue 3: Did the Examiner err in finding Uchida provides an enabling
disclosure for "comparing a vector distance between concepts of the target
patent and collection of concepts and other characteristics of text
documents," as recited in claim 1 and similarly recited in claims 10 and 17?
Issue 4: Did the Examiner improperly combine Uchida and Jiang?
3
Appeal2015-001112
Application 12/511,547
ANALYSIS
We have reviewed the Examiner's rejections in light of Appellants'
arguments that the Examiner has erred. We disagree with Appellants'
conclusions. We adopt as our own the findings and reasons set forth by the
Examiner in the Final Action (Final Act. 2-21) from which this appeal is
taken and the findings and reasons set forth in the Examiner's Answer in
response to Appellants' Appeal Brief (Ans. 17-28). We highlight and
address specific findings and arguments for emphasis as follows.
Issue 1
Appellants contend Jiang does not teach or suggest "identifying which
of the text documents are relevant to the target patent while not requiring
any of the text documents to be the same," as recited in claim 1 and similarly
recited in claims 10 and 17. (App. Br. 6-8, 11-13, 15-18.) Specifically,
Appellants argue Jiang requires identified text documents to be the same
because Jiang "compar[ es] titles of the documents with each other" in order
to "determin[e] which documents in a set of documents are the same." (Id.
at 7-8 (citing Jiang Abstract, i-fi-14, 23).)
We are not persuaded. The Examiner finds, and we agree, Uchida
teaches identifying relevant text documents that are not the same by
grouping different patents into clusters based on the similarity of the
different patents' text. (Final Act. 4 (citing Uchida Abstract, 2).)
Appellants' arguments against Jiang (App. Br. 6-8, 11-13, 15-18) do not
address the Examiner's findings regarding Uchida (Ans. 17; Final Act. 4).
Accordingly, we are not persuaded the Examiner erred in finding the
4
Appeal2015-001112
Application 12/511,547
combination of Uchida and Jiang teaches or suggests "identifying which of
the text documents are relevant to the target patent while not requiring any
of the text documents to be the same," as recited in claim 1 and similarly
recited in claims 10 and 17.
Issue 2
Appellants contend Uchida does not teach or suggest comparing
"without performing any comparisons between any text documents of a
subset of the text documents, wherein the subset includes all of the text
documents except the target patent," as recited in claim 1 and similarly
recited in claims 10 and 17. (App. Br. 8-10, 13-15, 18-20; Reply Br. 3-5.)
Specifically, Appellants argue Uchida "require[s] comparing every
document in a collection with every other document in that collection" (App.
Br. 9) because in Uchida, "all of the words from a first document would
have to be placed into [a word frequency] matrix and then each of those
words compared to all of the words in a second document" (Reply Br. 4).
We are not persuaded. Initially, we note that the claims do not
preclude comparisons between words of a word frequency matrix and a
document, as argued by Appellants; rather, the claims recite comparing
"without any comparisons between any text documents" (emphasis added).
Appellants contend that a comparison between a set of documents and words
extracted from that set of documents are precluded by the claim (App. Br.
23-24; Reply Br. 3--4), but Figure 6 in Appellants' Specification, which
Appellants identify as representative of the claimed subject matter (App. Br.
3), describes a similar comparison between concepts extracted from a set of
patents, e.g., "engine," and patent documents (Spec. ,-r 35, see Fig. 6
5
Appeal2015-001112
Application 12/511,547
(comparing patent documents 400, 606, and 608 to extracted concept
"ENGINE")). Accordingly, the claims do not preclude a comparison
between extracted words and the documents the words are extracted from.
The Examiner finds, and we agree, Uchida "compares vector
representations of text documents to other vector representations" using a
similarity calculation module. (Final Act. 3--4, 18 (citing Uchida 2-3).) The
Examiner further finds, and we agree, Uchida' s similarity calculation is
based on the creation of word frequency matrix which tracks the frequency
of words recited in a collection of patent documents. (Ans. 19; see Uchida
2.) That is, Uchida compares the words of the word frequency matrix with
documents in the set of documents, rather than comparing documents in the
set of documents with one another. (Uchida 2.) Accordingly, we are not
persuaded the Examiner erred in finding Uchida teaches or suggests
comparing "without performing any comparisons between any text
documents of a subset of the text documents, wherein the subset includes all
of the text documents except the target patent," within the meaning of claims
1, 10, and 17.
Issue 3
The Examiner finds Uchida' s clustering algorithm, which determines
distance between vectors, teaches or suggests "comparing a vector distance
between concepts of the target patent and collection of concepts and other
characteristics of text documents," as recited in claim 1 and similarly recited
in claims 10 and 17. (Final Act. 3--4 (citing Uchida 2-3); Ans. 20.)
Appellants argue Uchida's clustering algorithm is not enabled because the
algorithm references "data points" and "elements in L," but Uchida does not
6
Appeal2015-001112
Application 12/511,547
"state what Uchida's data points are" and does not "define what Uchida's
'L' represents." (App. Br. 10 (citing Uchida 2), 15, 19-20.)
During prosecution, "a prior art printed publication cited by an
examiner is presumptively enabling." In re Antor Media Corp, 689 F.3d
1282, 1288 (Fed. Cir. 2012). Thus, Appellants must show Uchida is not
enabled. Id. at 1288-89 (citing Amgen, Inc. v. Hoechst Marion Roussel,
Inc., 314 F.3d 1313, 1355 (Fed. Cir. 2003)).
We have carefully considered Appellants' arguments that Uchida does
not provide an enabling disclosure for the claimed subject matter and
conclude that Appellants have not overcome the presumption that Uchida is
enabled. Appellants' assertion that "Appellants do not understand [what]
Uchida['s]" data points and "L" represent do not provide sufficient evidence
or argument that one of ordinary skill in the art would have been unable to
practice Uchida's clustering algorithm. (App. Br. 10.) While a reference
that "appears to not be enabling on its face" can be challenged for
enablement "without resort to expert assistance," (In re Morsa, 713 F.3d
104, 110 (Fed. Cir. 2013)), we are not persuaded that Uchida is non-enabling
on its face (see Uchida 2, 5). Moreover, Appellants do not address the
Examiner's finding that, when reviewing the entirety of the Uchida
disclosure, one of ordinary skill in the art would understand that Uchida' s
"data points" are data points in set S and that "L" is the total set of clusters.
(See Ans. 20.)
Accordingly, Appellants have not persuaded us Uchida does not teach
or suggest "comparing a vector distance between concepts of the target
patent and collection of concepts and other characteristics of text
7
Appeal2015-001112
Application 12/511,547
documents," as recited in claim 1 and similarly recited in claims 10 and 17,
because Uchida is not enabled.
Issue 4
Appellants argue "there is no motivation to combine Uchida and Jiang
because Jiang and Uchida teach away from each other." (App. Br. 6, 8, 11,
13, 15-16, 18, 22 (emphasis omitted); Reply Br. 1-3.) Specifically,
Appellants argue Jiang "teach[ es] away from 'identifying which of the text
documents are relevant to the target patent while not requiring any of the
text documents to be the same."' (App. Br. 6 (emphasis omitted).)
We are not persuaded. The Examiner finds, and we agree, Uchida
discloses a system which compares patent documents based on extracted
characteristics of the patent documents. (Final Act. 2-3 (citing Uchida
Abstract, 2).) The Examiner further finds, and we agree, Jiang teaches
patent document characteristics include citations to other references and
"other patent information such as author, application date, classification, and
assignee." (Id. at 5 (citing Jiang i-f 20).) The Examiner combines Uchida
and Jiang, resulting in a system which compares patent documents based on
characteristics including citations to other references and other patent
information. (Id.)
Appellants' arguments discussing Jiang' s method of identifying patent
documents (App. Br. 7-8, 21-22; Reply Br. 1-3) do not persuade us that
Jiang teaches away from modifying Uchida's comparison of patent
documents to compare the patent document characteristics taught by Jiang.
The Examiner's combination relies on Jiang to teach various patent
document characteristics, but Appellants' arguments discuss a different
8
Appeal2015-001112
Application 12/511,547
feature of Jiang, one that the Examiner does not rely on in the combination.
Accordingly, Appellants' arguments do not persuasively show that either
reference criticizes, discredits, or discourages the combination found by the
Examiner (i.e., including Jiang's patent document characteristics in Uchida's
patent document comparison). See In re Fulton, 391 F.3d 1195, 1201 (Fed.
Cir. 2004). Furthermore, while Appellants argue Jiang teaches identification
of the same document, we agree with the Examiner's finding that "Jiang
does not require text documents to be the same." (Ans. 17-18 (citing Jiang
i-f 23).) Indeed, Jiang teaches that identified documents may be similar, but
not the same. (See Jiang i-f 23 (teaching similarity values less than 1.0,
where a similarity value of 1.0 indicates an identical identification).)
Additionally, Appellants argue Uchida teaches away from its
combination with Jiang because Uchida does not compare vector distances
"without performing any comparisons between any text documents of a
subset of the text documents, wherein the subset includes all of the text
documents except the target patent." (App. Br. 8-10, 13-14, 18-20.) As
discussed supra, we agree with the Examiner's finding that Uchida teaches
comparing vector distances without comparisons between text documents.
(Final Act. 3--4, 18 (citing Uchida 2-3).) Furthermore, as discussed above,
Appellants have not persuasively identified any criticism, discrediting, or
discouragement in Uchida which teaches away from the consideration of
Jiang's patent characteristics in Uchida's patent comparison. In re Fulton,
391 F.3d at 1201.
Accordingly, we are not persuaded the Examiner improperly
combined Uchida and Jiang.
9
Appeal2015-001112
Application 12/511,547
Remaining Claims
Appellants do not argue separate patentability for dependent claims 2-
6, 8, 9, 11-13, 15, 16, and 19 which depend directly or indirectly from
claims 1, 10, or 17. (See App. Br. 10, 15, 20-21.) For the reasons set forth
above, therefore, we are not persuaded the Examiner erred in rejecting these
claims. Accordingly, we sustain the Examiner's rejections of claims 2-6, 8,
9, 11-13, 15, 16, and 19. See 37 C.F.R. § 41.37(c)(l)(iv).
DECISION
For the above reasons, the Examiner's decision to reject claims 1-6,
8-13, 15-1 7, and 19 is 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)(l )(iv).
AFFIRMED
10