Ex Parte Dean et al

Board of Patent Appeals and Interferences

Jul 25, 2012

10917746 (B.P.A.I. Jul. 25, 2012)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE BOARD OF PATENT APPEALS
AND INTERFERENCES
____________

Ex parte JEFFREY ADGATE DEAN, PAUL G. HAAHR,
OLCAN SERCINOGLU, and AMITABH K. SINGHAL

___________

Appeal 2009-015116
Application 10/917,746
Technology Center 2100
____________

Before JEFFREY S. SMITH, ERIC B. CHEN, and
ANDREW CALDWELL, Administrative Patent Judges.

CHEN, Administrative Patent Judge.

DECISION ON APPEAL

Appeal 2009-015116
Application 10/917,746

2
This is an appeal under 35 U.S.C. § 134(a) from the final rejection of
claims 2, 3, 23, 24, and 26-39. Claims 1, 4-22, and 25 have been cancelled.
We have jurisdiction under 35 U.S.C. § 6(b). We reverse.

STATEMENT OF THE CASE
Appellants’ invention relates to a multi-stage query processing system
and method that enables multi-stage query scoring through incremental
document reconstruction facilitated by a multi-tiered mapping scheme. A
set of relevancy scores are used to select a subset of documents for
presentation as an ordered list to a user. The set of relevancy scores can be
derived in part from one or more sets of relevancy scores determined in prior
stages of the multi-stage query processing system. (Abstract.)
Claim 23 is exemplary, with disputed limitations in italics:
23. A method of processing a query in a multi-stage query
processing system, comprising:
performing first stage processing of the query, including:
retrieving from an index, a list of token positions
for one or more query terms in the query, wherein the
token positions in the list specify positions of individual
tokens that match the query terms;
based on the list of token positions, producing a
first set of document identifiers, and
generating a first set of relevancy scores for a first
set of documents corresponding to the first set of
document identifiers;
performing a second stage processing of the query, including:
generating a second set of relevancy scores for the
documents in the first set of documents based on one or
more of: the list of token positions, distances between
query terms in the documents, attributes of tokens in the
Appeal 2009-015116
Application 10/917,746

3
documents, and a context in which a query term is used
in a document of the first set of documents, and
selecting a subset of the first set of documents
based on the second set of relevancy scores;
automatically generating one or more new query
expansion terms based on documents in the subset of the first
set of documents; and
generating a new query expression based on the query
and the one or more new query expansion terms for processing
by at least the first processing stage to produce at least a second
set of document identifiers corresponding to a second set of
documents.
Claims 2, 3, 23, 24, and 26-39 stand rejected under 35 U.S.C. § 103(a)
as being obvious over Messerly (U.S. Patent No. 6,076,051) and Ballard
(U.S. Patent No. 5,987,457).

ANALYSIS
We are persuaded by Appellants’ arguments (App. Br. 15-16; see also
Reply Br. 6-8) that the combination of Messerly and Ballard would not have
rendered obvious independent claim 23, which includes the disputed
limitation “generating a second set of relevancy scores for the documents in
the first set of documents.”
The Examiner found that the facility of Messerly that processes
information retrieval queries against semantic tokens in target documents
and ranks such target documents corresponds to the claimed “performing
first stage processing of the query.” (Ans. 3.) The Examiner further found
that an additional preferred embodiment of Messerly, in which document
groups are ranked, corresponds to the claimed “generating a second set of
Appeal 2009-015116
Application 10/917,746

4
relevancy scores for the documents in the first set of documents.” (Ans. 4,
15-16.) We do not agree.
Messerly relates to information retrieval, in particular, information
retrieval tokenization. (Col. 1, ll. 6-8.) Figure 3 of Messerly illustrates a
flow diagram of steps performed by an information retrieval tokenization
facility that “indexes the target documents by converting each sentence or
sentence fragment of the target document into a number of [semantic]
tokens” (col. 5, ll. 34-36) such that the facility can process informational
retrieval queries against this index (col. 5, ll. 42-44). Furthermore, “[f]or
each such query received, the facility tokenizes the text of the query” (col. 5,
ll. 44-46) and “then compares these semantic tokens to the semantic tokens
stored in the index . . . and ranks the target documents containing these
semantic tokens in the order of their relevance to the query” (col. 5, ll. 48-
53). In an additional preferred embodiment, “target documents . . . are
organized into document groups each containing one or more documents”
(col. 13, ll. 57-58) such that the document groups and document sections are
ranked by relevancy (col. 13, ll. 54-56).
In other words, in the Figure 3 embodiment of Messerly, the
information retrieval tokenization facility semantically indexes target
documents using semantic tokens. (See col. 5, ll. 34-36, 42-44.) In another
“additional preferred embodiment” of Messerly, the target documents are
“organized into document groups each containing one or more documents.”
(Col. 13, ll. 57-58.) However, Messerly does not provide an adequate
teaching or suggestion that the Figure 3 embodiment of Messerly and the
“additional preferred embodiment” of Messerly can be performed
sequentially. Instead, Messerly teaches that the “additional preferred
Appeal 2009-015116
Application 10/917,746

5
embodiment” of Messerly is an alternative to the Figure 3 embodiment for
situations in which target documents are organized into groups.
Thus, we do not agree with the Examiner that the combination of
Messerly and Ballard would have rendered obvious independent claim 23,
which includes the limitation “generating a second set of relevancy scores
for the documents in the first set of documents.”
Accordingly, we do not sustain the rejection of independent claim 23
under 35 U.S.C. § 103(a). Claims 2, 3, 24, 26, and 27 depend from
claim 23. Therefore, we do not sustain the rejection of claims 2, 3, 24, 26,
and 27 under 35 U.S.C. § 103(a), for the same reasons discussed with
respect to independent claim 23.
Independent claims 28 and 34 recite limitations similar to those
discussed with respect to independent claim 23. We do not sustain the
rejection of claims 28 and 34, as well as claims 29-33 and 35-39, which
depend from claims 28 and 34, for the same reasons discussed with respect
to claim 23.

DECISION
The Examiner’s decision to reject claims 2, 3, 23, 24, and 26-39 is
reversed.

REVERSED

rwk