Ex Parte Chidlovskii et al

Patent Trial and Appeal Board

Mar 15, 2013

11170542 (P.T.A.B. Mar. 15, 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.
11/170,542 06/29/2005 Boris Chidlovskii 20041565USNP-XER01011US01 2935
27885 7590 03/15/2013
FAY SHARPE LLP
1228 Euclid Avenue, 5th Floor
The Halle Building
Cleveland, OH 44115
EXAMINER
PAULA, CESAR B
ART UNIT PAPER NUMBER
2177
MAIL DATE DELIVERY MODE
03/15/2013 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 BORIS CHIDLOVSKII and JEROME FUSELIER
____________

Appeal 2010-007677
Application 11/170,542
Technology Center 2100
____________

Before STEPHEN C. SIU, JOSIAH C. COCKS, and JENNIFER S. BISK,
Administrative Patent Judges.

SIU, Administrative Patent Judge

DECISION ON APPEAL

This is an appeal under 35 U.S.C. § 134 from the Examiner’s rejection
of claims 1-9, 11-14, and 19-23. Claims 10 and 15-18 were cancelled. We
have jurisdiction under 35 U.S.C. § 6(b).

STATEMENT OF THE CASE
The disclosed invention relates generally to annotation of documents
(Spec. 1).
Independent claim 1 reads as follows:
Appeal 2010-007677
Application 11/170,542

2
1. A document processor comprising:
a classifier that classifies fragments of an input document respective
to a set of terminal elements; and
a parser that defines a parsed document structure associating the input
document fragments with terminal elements connected by links of non-
terminal elements conforming with a probabilistic grammar defining
transformation rules operating on elements selected from the set of terminal
elements and a set of non-terminal elements, the parsed document structure
being used to organize the input document, the parser defining the parsed
document structure by optimizing a probability function of the parsed
document structure including at least a probability function term indicative
of a probability of the connecting links of non-terminal elements of the
parsed document structure computed using the probabilistic grammar.

The Examiner relies on the following references:
1) Eila Kuikka, Paula Leinonen, and Martti Penttonen, “Towards
Automating of Document Structure Transformations,” DocEng ’02, 103,
November 8-9 2002) (“Kuikka”);
2) IBM Technical Disclosure Bulletin, “Probabilistic Model of
Link Grammar,” Vol. 37, No. 02B, Feb. 1994 (“IBM”); and
3) Keith Hall and Mark Johnson, “Language Modeling Using
Efficient Best-First Bottom-Up Parsing,” IEEE, 2003 (“Hall”).

The Examiner rejects the claims as follows:
1) Claims 1-4, 6, 8, 11-14, and 19-23 under 35 U.S.C. § 103(a) as
unpatentable over Kuikka and IBM;
1

1
The Examiner deems claim 7 to be allowable with respect to Kuikka and
Appeal 2010-007677
Application 11/170,542

3
2) Claims 5 and 9 under 35 U.S.C. § 103(a) as unpatentable over
Kuikka, IBM, and Hall; and
3) Claims 1-9, 19, and 20 under 35 U.S.C. § 101 as being directed
to non-statutory subject matter.
2

Issue
Did the Examiner err in rejecting claims 1-9, 11-14, and 19-23?
Principles of Law
The question of obviousness is resolved on the basis of underlying
factual determinations including (1) the scope and content of the prior art,
(2) any differences between the claimed subject matter and the prior art, and
(3) the level of skill in the art. Graham v. John Deere Co., 383 U.S. 1, 17-
18 (1966).
“The combination of familiar elements according to known methods
is likely to be obvious when it does no more than yield predictable results.”
KSR Int’l Co. v. Teleflex, Inc., 550 U.S. 398, 416 (2007).

IBM (Ans. 15).
2
The Examiner does not appear to withdraw the rejection of claim 7 under
35 U.S.C. § 101. We assume this rejection still applies to claim 7.
Appeal 2010-007677
Application 11/170,542

4
Analysis
35 U.S.C. § 103(a)
Appellants argue that Kuikka “does not determine the parse document
structure” (App. Br. 13). Claim 1, for example, recites “a parser that defines
a parsed document structure.” We disagree with Appellant for at least the
reasons set forth by the Examiner (see, e.g., Ans. 15).
The Examiner states that Kuikka discloses segmenting a source
document into “various node substructures” (see, e.g., Ans. 15, citing
Kuikka p. 107, Fig. 2). Appellants do not dispute this disclosure in Kuikka.
Both the substructures disclosed by Kuikka and the claimed “parsed
document structure” appear to include segments or fragments of a document
(i.e., “parsed”) that form a “structure” (see, e.g., Kuikka Fig. 2). Appellants
do not point out sufficient differences between the node substructures of
Kuikka and the “parsed document structure” as recited in claim 1, for
example.
Appellants also argue that Kuikka “does not disclose or fairly suggest
a document processor comprising a classifier that classifies fragments of an
input document respective to set of terminal elements” (App. Br. 15). We
disagree with Appellants for at least the reasons set forth by the Examiner
(see, e.g., Ans. 16). As the Examiner indicates, Kuikka discloses “target
substructures” including nodes (representing fragments of a document, see,
e.g., Fig. 2) of various classifications (e.g., “title,” “author,” etc. – see Fig.
2). Appellants do not demonstrate specific differences between Kuikka
Appeal 2010-007677
Application 11/170,542

5
(fragments of a document classified in a node substructure) and classifying
fragments of an input document, as recited in claim 1, for example.
Appellants argue that while IBM discloses “link grammar” (i.e., a
“probabilistic grammar” as recited in claim 1), the “link grammar” of IBM,
according to Appellants, “does not use non-terminal elements” (Reply Br.
5). IBM discloses, for example, that in “the most common probabilistic
model . . . parameters . . . [include] nonterminals” (p. 99, 3
rd
paragraph).
Since the most common model includes nonterminals (i.e., “non-terminal
elements” according to IBM), one of ordinary skill in the art would have
known that non-terminal elements are used (and used most commonly).
Appellants argue that it would not have been obvious to one of
ordinary skill in the art to have combined Kuikka and IBM (App. Br. 17-19;
Reply Br. 7-8). We disagree with Appellants for at least the reasons set
forth by the Examiner (see, e.g., Ans. 5). As the Examiner states, Kuikka
discloses transformation of documents (i.e., language and grammar) from
one structure to another while IBM also discloses conversion of language
and grammar models using “link grammar.” The combination of Kuikka
and IBM (i.e., two methods of language/grammar conversion of structure)
would have entailed no more than the mere combination of known
components, each performing its respective known function, to achieve the
predictable result of the conversion of document structures. “The
combination of familiar elements according to known methods is likely to be
obvious when it does no more than yield predictable results.” KSR Int’l Co.,
Appeal 2010-007677
Application 11/170,542

6
550 U.S. at 416.
Regarding claim 3, the Examiner states that IBM discloses “that the
link grammar is used to calculate an approximate probability function –
optimizing a probability function – for a predetermined probability function
of the terminal and nonterminal words, and their respective links or
linkage” (Ans. 17-18) (emphasis in original). Claim 3 recites a probabilistic
classifier that assigns probability values corresponding to elements of the set
of terminal elements and a probability function optimized by a parser.
While Appellants argue that “the cited pages 6-7 of [the] ‘IBM reference’
[is] indecipherable” (App. Br. 20), Appellants do not sufficiently
demonstrate a difference between the probability function of IBM (as cited
by the Examiner), for example, and the probability function recited in claim
3 (see, also, Ans. 6-7).
Regarding claim 4, Appellants argue that Kuikka and IBM both fail to
disclose or suggest “a parser including a joint probability optimizer that
optimizes the parsed document structure respective to a joint probability . . .”
(App. Br. 23) while the Examiner states that IBM discloses “link grammar”
that is “used to calculate an approximate probability function – optimizing a
probability function – for a predetermined probability function of the
terminal and nonterminal words, and their respective links or linkages
leading to the terminal and nonterminal words” (Ans. 18) (emphasis in
original). Claim 4 recites, for example, optimizing a parsed document
structure respective to a joint probability of probability values of associated
Appeal 2010-007677
Application 11/170,542

7
terminal elements and probabilities of connecting links of non-terminal
elements derived from the probabilistic grammar. Appellants do not point
out a difference between calculating the probability function of IBM (as
indicated by the Examiner), for example, and optimizing a parsed document
structure respective to a joint probability as recited in claim 4. We therefore
agree with the Examiner (see, e.g., Ans. 7-8).
Regarding claim 6, Appellants argue that Kuikka fails to disclose or
suggest “a classifier trainer” (App. Br. 23) while the Examiner states that
Kuikka discloses a heuristic function that “uses the parameters found in the
structure from the translated source documents, to determine – training to
find – the best source document translations” (Ans. 19). Appellants do not
dispute this disclosure of Kuikka and do not demonstrate a difference
between determining (or “training to find,” per the Examiner) source
document translations based on parameters and the disputed claim feature of
training a classifier to a set of training documents (see, also, Ans. 8).
Appellants do not provide additional arguments in support of claims 5
or 9 or arguments with respect to the Hall reference.

35 U.S.C. § 101
The Examiner states that claims 1-9, 19, and 20 recite “software
instructions” that are “devoid of any physical articles or objects” and are
therefore directed to non-statutory subject matter (Ans. 3). Appellants
disagree with the Examiner because, according to Appellants, claim 1 and
Appeal 2010-007677
Application 11/170,542

8
claim 4 recite a “processor” and “[a] processor is not software” (App. Br.
26). We disagree with Appellants for at least the reasons set forth by the
Examiner (Ans. 3 and 19-20).
Conclusion of Law
The Examiner did not err in rejecting claims 1-9, 11-14, and 19-23.

SUMMARY
We affirm the Examiner’s rejection of claims 1-6, 8, 11-14, and 19-23
under 35 U.S.C. § 103(a) as unpatentable over Kuikka and IBM; claims 5
and 9 under 35 U.S.C. § 103(a) as unpatentable over Kuikka, IBM, and Hall;
and claims 1-9, 19, and 20 under 35 U.S.C. § 101.

AFFIRMED

ack