Ex Parte Boguraev et al

Patent Trial and Appeal Board

Feb 27, 2017

13610267 (P.T.A.B. Feb. 27, 2017)

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.
13/610,267 09/11/2012 Branimir K. Boguraev YOR920120433US1 1032
29154 7590
FREDERICK W. GIBB, III
GIBB & RILEY, LLC
844 West Street
SUITE 200
ANNAPOLIS, MD 21401
EXAMINER
SECK, ABABACAR
ART UNIT PAPER NUMBER
2122
NOTIFICATION DATE DELIVERY MODE
03/01/2017 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):
support @ gibbiplaw .com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte BRANIMIR K. BOGURAEV,
DAVID W. BUCHANAN, JENNIFER CHU-CARROLL,
DAVID A. FERRUCCI, ADITYA A. KALYANPUR,
JAMES W. IV MURDOCK, and
SIDDHARTH A. PATWARDHAN
Appeal 2016-005025
Application 13/610,267
Technology Center 2100
Before JEAN R. HOMERE, JON M. JURGOVAN, and
PHILLIP A. BENNETT, Administrative Patent Judges.
JURGOVAN, Administrative Patent Judge.
DECISION ON APPEAL
Appeal 2016-005025
Application 13/610,267
STATEMENT OF THE CASE
Appellants1 seek review under 35 U.S.C. § 134(a) from a Final
Rejection of claims 1—25. We have jurisdiction under 35 U.S.C. § 6(b).
We affirm.2
CLAIMED INVENTION
The claims are directed to generating secondary questions in an
introspective question answering system. (Spec. Title.) Claim 1,
reproduced below, is illustrative of the claimed subject matter:
1. A method of generating secondary questions in a question-
answer system, said method comprising:
obtaining candidate answers for a first question posed to
said question-answer system;
identifying missing information from a corpus of data, said
missing information comprising any information that improves
confidence scores for said candidate answers to said first
question;
automatically generating a plurality of hypotheses
concerning said missing information;
automatically generating at least one secondary question
based on each of said plurality of hypotheses concerning said
missing information;
ranking said hypotheses based on relative utility to
determine an order in which to output said at least one secondary
question to external sources to obtain responses.
1 The Appeal Brief identifies International Business Machines Corporation
as the real party in interest. (App. Br. 3.)
2 Our Decision refers to the Specification (“Spec”) filed Sept. 11, 2012, the
Final Office Action (“Final Act.”) mailed Oct. 30, 2014, the Appeal Brief
filed July 7, 2015 (“App. Br.”), the Examiner’s Answer (“Ans.”) mailed Feb
22, 2016, and the Reply Brief filed April 8, 2016 (“Reply Br.”).
2
Appeal 2016-005025
Application 13/610,267
REJECTIONS
Claims 1—25 stand rejected under 35 U.S.C. § 101 as being directed to
non-statutory subject matter. (Final Act. 2—3.)
Claims 1, 6—8, 13—15, 18—20, 24, and 25 stand rejected under 35
U.S.C. § 103(a)3 based as being unpatentable over the combination of
Kupiec (US 5,519,608, May 21, 1996) and Wang (US 2010/0191686 Al,
July 29, 2010). (Final Act. 3-12.)
Claims 2—5, 9-12, 16, 17, and 21—23 stand rejected under 35 U.S.C.
§ 103(a) as being unpatentable over the combination of Kupiec, and Ferruci
(David Ferruci et al., IBM Research Report, Watson: Beyond Jeopardy/,
Computer Science, RC25270 (W1106-019) (June 2, 2011)). (Final Act.
12-22.)
ANALYSIS
Rejection of Claims 1,6-8,13-15,18-20,24, and 25 under 35 U.S.C.
§ 1 03(a) based on Kupiec and Wang
Missing Information
Appellants argue the combination of Kupiec and Wang fails to
disclose “identifying missing information” as recited in independent claims
1, 8, 15, and 20. (App. Br. 9, Reply Br. 2.) According to Appellants,
“missing information” as claimed is any information that improves
confidence scores for candidate answers to a “first question.” (Id.)
3 A paragraph in the Final Office Action misidentifies the statutory basis for
the rejection as under § 102. (Final Act. 4.) However, the beginning of the
rejection correctly identifies § 103(a) as the statutory basis. (Final Act. 3.)
As we are aware of no evidence indicating Appellants did not apprehend the
nature of the rejection, we view the error to be harmless. (Final Act. 3.)
3
Appeal 2016-005025
Application 13/610,267
Appellants contend Kupiec, the reference relied on by the Examiner to teach
this feature, does not disclose “missing information” as so defined. (Id.)
On the contrary, the Examiner finds that the claimed “missing
information” are the guess words disclosed in Kupiec. (Ans. 27 citing
Kupiec 3:44—67, Final Act. 4 citing Kupiec 2:24—51, Section 7. Hypothesis
Ranking, 34:60-35:15, 38:8—67.) In particular, Kupiec states:
In the present invention, the answer hypotheses generated are
used in new queries that the system constructs without user aid.
These secondary queries are based on guesses — words that in
the vast majority of cases are not present in the user’s original
question.
(Kupiec 3:44-48.)
The Examiner apparently considers the guess words to be “missing
information” because Kupiec describes them as generally not being present
in the original question (which is the claimed “first question”). The
Examiner also appears to find that, in order to be used in secondary queries,
Kupiec’s guess words must be identified, which corresponds to the claimed
“identifying.” Secondary queries based on Kupiec’s guess words are used to
retrieve secondary documents from a data corpus. (Kupiec 5:3—6, 6:66—67 —
definition of “secondary documents,” Figs. 1 and 5.) The Examiner finds
that Kupiec’s refining of the ranking of retrieved documents improves
confidence scores. (Ans. 27.) In particular, Kupiec discloses:
Narrowing is performed to attempt to reduce the number of hits.
It can also serve to refine the ranking of the documents.
(Kupiec 34:66—35:1.)
We agree with the Examiner’s findings concerning Kupiec,
specifically, that the claimed “missing information” corresponds to Kupiec’s
guess words. Kupiec’s guess words are automatically generated as part of
4
Appeal 2016-005025
Application 13/610,267
secondary queries designed to refine ranking of documents and improve
confidence scores, thus meeting Appellants’ definition of the claimed term.
Appellants further argue that Kupiec’s guess words are not the
claimed “missing information” because they could as easily decrease as
increase confidence scores. (Reply Br. 2). We disagree. The purpose of
Kupiec’s narrowing to refine ranking of documents is to generate more
relevant search results, which necessarily improves confidence scores. We
also note that the argument itself admits there are circumstances under which
the confidence of results generated by the guess words increases, even
supposing that in some cases they decrease. In the cases in which
confidence increases, the claimed limitation is satisfied by Kupiec’s
disclosure. Thus, we do not find this argument persuasive.
Generating Hypotheses
Appellants argue Kupiec does not disclose “generating a plurality of
hypotheses concerning said missing information’'' as claimed. (App. Br.
10-11.) Specifically, Appellants argue that “Kupiec is not looking at
developing the next question. Nothing in Kupiec suggests generating and
evaluating hypotheses after the first question has been answered.” (Id. at
11.)
We disagree with Appellants’ argument. As the Examiner correctly
notes, Kupiec clearly states “the answer hypotheses generated are used in
new queries that the system constructs without user aid.” (Ans. 27 citing
Kupiec 3:4AA8, supra', see also Fig. 5 [265], [266], [270].) This teaching
unequivocally discloses the claimed limitation.
Automatically Generating Secondary Question
Appellants further argue Kupiec fails to disclose the claim limitation
of “automatically generating at least one secondary question based on each
5
Appeal 2016-005025
Application 13/610,267
of said plurality of hypotheses.” (App. Br. 12—13, Reply Br. 3.) According
to the Examiner, the broad claim language does not define a “secondary
question” in a manner that differentiates from Kupiec’s secondary queries.
(Ans. 27 citing Kupiec 3:44^48, supra.)
We agree with the Examiner. It is well-established that “the PTO is
obligated to give claims their broadest reasonable interpretation during
examination.” See In re Am. Acad. ofSci. Tech. Ctr., 367 F.3d 1359, 1369
(Fed. Cir. 2004). As the Examiner correctly notes, the claimed limitation is
disclosed by Kupiec, which states “the answer hypotheses generated are
used in new queries that the system constructs without user aid.'” (Final Act.
4 citing Kupiec 3:44^48.)(Emphasis added). Kupiec’s new queries
correspond to the claimed “secondary question” and the phrase “without
user aid” means “automatically” as claimed. Thus, Kupiec’s teaching is
sufficient to show the claimed limitation under the broadest reasonable
interpretation.
Ranking Hypotheses
According to Appellants, Kupiec does not rank hypotheses to
determine an order to output a secondary question, as claimed. (App. Br.
14.) To the contrary, the Examiner finds this limitation disclosed in Kupiec.
(Final Act. 5 citing Kupiec 21:2—25:16 (Section 7. Hypothesis Ranking, 7.1
Scoring, and 7.2 Merging Scores).)
We agree with the Examiner this limitation is taught by Kupiec. For
example, one excerpt from the Examiner’s citation states:
Hypothesis ranking can be done by assigning a score to each
hypothesis and ordering the hypotheses according to the scores
thus assigned. The highest-ranked hypotheses are considered the
best answers. Any number of scoring methods can be used.
6
Appeal 2016-005025
Application 13/610,267
Scoring can be based in whole or in part on the verification
evidence collected in step 260.
(Kupiec 21:9—14.)
Thus, Kupiec discloses the argued limitation and we are not persuaded by
Appellants’ argument.
Rejection of Claims 2-5,9-12,16,17, and 21-23 under 35 U.S.C.
§ 103(a) based on Kupiec and Ferrucci
Appellants contend the Examiner errs in the rejection for reasons
similar to those addressed previously. (App. Br. 15—16.) We agree with the
Examiner that Kupiec is not deficient (Final Act. 12—22, Ans. 31—32), and
we sustain the Examiner’s rejection for the reasons stated.
Rejection of Claims 1-25 under 35 U.S.C. § 101
The Examiner finds the claims are directed to the abstract idea of
organizing human activities without significantly more. (Ans. 36.) The
Examiner further finds that the claimed limitation of obtaining candidate
answers amounts to no more than insignificant data gathering activities.
(Id.)
Appellants disagree with the Examiner’s findings, arguing that “the
present claims specifically apply the “idea” of automated question answering
to a new and useful end because the claims provide such new and useful
ends as finding missing information from external sources, which the system
can use to improve its ability to understand and answer questions.” (App.
Br. 17—18, Reply Br. 4—5.) Appellants also argue the claims do not pre-empt
all practical uses of an idea because the claims are constrained to generating
secondary questions. (Id.) Further, Appellants argue that recitation of some
structure, such as a machine, or the recitation of some transformative
component, and that a computer is integral to the claimed invention. (Id.)
7
Appeal 2016-005025
Application 13/610,267
The test for subject matter eligibility is set forth in Alice Corp. Pty.
Ltd. v. CLS Bank Inti, 134 S. Ct. 2347 (2014), Mayo Collaborative Services
v. Prometheus Laboratories, Inc., 566 U.S. 66 (2012), and Association for
Molecular Pathology v. Myriad Genetics, 133 S. Ct. 2107 (2013). From the
nature of the Examiner’s findings and Appellants’ arguments, it appears to
be conceded the claims recite an abstract idea, and the question remaining to
be answered is whether the claim recites significantly more than the judicial
exception to be subject-matter eligible under § 101. Factors pro and con
used to decide this question are set forth in the USPTO’s 2014 Interim
Guidance on Patent Subject Matter Eligibility, Vol. 79, No. 241, p. 74624,
and the Examiner’s findings and Appellants’ arguments are judicially
recognized as factors for use in deciding this question.
We do not agree with Appellants’ argument that the claims, as broadly
recited, are directed to a new and useful end for the reasons stated with
respect to the obviousness rejections. Specifically, while Appellants appear
to regard the “missing information” as related only to the data corpus that is
the target of the questions, as the Examiner demonstrated, the term is
disclosed by Kupiec’s guess words as broadly claimed. Thus, we are not
persuaded this factor weighs in Appellants’ favor in the eligibility analysis.
However, concerning Appellants’ pre-emption argument, we consider
the claims not to pre-empt all practical uses of the recited abstract idea.
Constraining a system to answer secondary questions does make the system
less abstract. Furthermore, the claims mention “generating secondary
questions in a question-answering system” and state that hypotheses and
secondary questions are generated “automatically.” We find these
limitations do tie the claimed invention to a machine, tending to show they
are patent-eligible claims.
8
Appeal 2016-005025
Application 13/610,267
We also note that, apart from the mere assertion, the Examiner has not
demonstrated how the claims are directed to organizing human activities.
We do agree with the Examiner, however, that “obtaining candidate
answers” may be considered a data gathering step.
Thus, on the balance of these factors, considering the claims as a
whole, we agree with Appellants that the claimed invention is significantly
more than the abstract idea itself, and we do not sustain the Examiner’s
rejection of the claims under § 101.
DECISION
We affirm the Examiner’s rejections of claims 1—25 under 35 U.S.C.
§ 103(a).
We reverse the rejection of claims 1—25 under 35 U.S.C. § 101.
Because we have affirmed at least one ground of rejection with
respect to each claim on appeal, the Examiner’s decision is affirmed. See
37 C.F.R. § 41.50(a)(1).
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
9