Knowlege Synthesis

Patent Trials and Appeals Board

Oct 2, 2020

14719246 - (D) (P.T.A.B. Oct. 2, 2020)

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.
14/719,246 05/21/2015 Hugh Salamon 46925-501001US 2687
30623 7590 10/02/2020
Mintz Levin/Boston Office
One Financial Center
Boston, MA 02111
EXAMINER
PHAM, KHANH B
ART UNIT PAPER NUMBER
2166
NOTIFICATION DATE DELIVERY MODE
10/02/2020 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):
IPDocketingBOS@mintz.com
IPFileroombos@mintz.com
mintzdocketing@cpaglobal.com
PTOL-90A (Rev. 04/07)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________

Ex parte HUGH SALAMON and KEN D. YAMAGUCHI

Appeal 2019-003377
Application 14/719,246
Technology Center 2100

Before JOSEPH L. DIXON, DAVID M. KOHUT, and
JON M. JURGOVAN, Administrative Patent Judges.

KOHUT, Administrative Patent Judge.

DECISION ON APPEAL
Pursuant to 35 U.S.C. § 134(a), Appellant1 appeals from the
Examiner’s decision to reject claims 1–20. We have jurisdiction under
35 U.S.C. § 6(b).
We AFFIRM.

1 We use the word “Appellant” to refer to “applicant” as defined in
37 C.F.R. § 1.42(a). Appellant identifies the real party in interest as Hugh
Salamon and Ken D. Yamaguchi (the inventors). Appeal Br. 1.
Appeal 2019-003377
Application 14/719,246

2
INVENTION
Appellant’s invention relates to systems and methods for searching
large data collections to generate large numbers of statistically significant
findings, and ranking and querying statistical analysis results of database
contents to populate search engine query results with novel content.
Spec. ¶ 1, Abstract.
Claims 1 and 10 are illustrative of argued subject matter.
1. A system for searching and analyzing data, comprising:
an engine configured to rank and query statistical
analysis results of database contents that include records, the
database contents comprising:
a scalar data field for each of the records;
a categorization of each of the records or a
network of relationships among the records;
natural language that describes the scalar data field
and the categorization;
a statistical test for each of the records, category of
the records, or vertex of the records in the network;
a rank of results of the statistical tests;
a ranking of the test results across multiple
analyses of data in which results are statistical test
computations normalized by the rank of the results of the
statistical tests and normalized by strength of effect of the
statistical test computations; and
a final ranking of findings consisting of the rank of
the record, category, or network finding, adjusted for
query term hits on the natural language; and
a processor configured to receive a search engine query
input that includes the query term and to provide a search
engine query result to the user based on the final ranking.

Appeal 2019-003377
Application 14/719,246

3
10. A system for searching and analyzing data, comprising:
a memory storing instructions therein; and
a processor configured to receive a search engine query
input by a user that includes a query term, the processor being
configured to execute the instructions and thereby:
retrieve findings from a plurality of different data
sources each storing a plurality of records each including
a text field, the findings being based on a match of the
query term with the text field,
calculate a statistical test result of the findings,
each test result including a p-value and a strength of
effect,
calculate a weighted test result using the strength
of effect,
rank the weighted test results from largest to
smallest,
calculate a relevance score for each weighted test
result based on the ranking,
normalize the relevance scores, and
rank query results to be provided to the user in
response to the search engine query by ranking the
normalized relevance scores from largest to smallest.

Appeal Br. A–D (Claims App.).
REJECTION2
Claims 1–20 stand rejected under 35 U.S.C. § 102(a)(1) as anticipated
by Kupershmidt (US 2007/0162411 A1; July 12, 2007). Final Act. 7–12.

2 Claims 1–20 were rejected under 35 U.S.C. § 101 as being directed to
patent-ineligible subject matter. Final Act. 2–6. However, this rejection was
withdrawn in the Examiner’s Answer and is no longer pending on appeal.
Ans. 3.
Appeal 2019-003377
Application 14/719,246

4
OPINION
Claims 1–9
With respect to claim 1, Appellant argues Kupershmidt does not
disclose
database contents comprising: a scalar data field for each of the
records; a categorization of each of the records or a network of
relationships among the records; natural language that describes
the scalar data field and the categorization; . . . and a final
ranking of findings consisting of the rank of the record, category,
or network finding, adjusted for query term hits on the natural
language.
Appeal Br. 5–6. Appellant argues, although Kupershmidt presents a user’s
query results in a ranked order, “Kupershmidt provides no disclosure that the
ranking is adjusted in using the user query,” and “[q]uery term hits are never
mentioned in Kupershmidt, much less that the ranking presented in response
to the user’s query is in any way adjusted for query terms hits on natural
language or on anything else.” Appeal Br. 6; Reply Br. 3. More
particularly, Appellant asserts Kupershmidt’s ranking is based only on
statistics (correlation scores) that were previously computed during the
Knowledge Base generation process of Figure 1 “before any query has been
input” and “before the query term is known and thus also before any hits on
the query term are known.” Appeal Br. 6; Reply Br. 2–3 (citing
Kupershmidt ¶¶ 162–164, Figs. 1 and 13A).
Appellant’s arguments do not persuade us of Examiner error. The
Examiner finds, and we agree, that Kupershmidt’s processor receives a
query input that includes a query term, and provides a search engine query
result to the user based on a final ranking, as required by claim 1. Ans. 3–4
(citing Kupershmidt ¶ 164, Fig. 13A (showing receipt of a query input at
Appeal 2019-003377
Application 14/719,246

5
step 1301, and output of a search engine query result to user at step 1307,
based on a final ranking from step 1305)); see Kupershmidt ¶ 164 (“as
indicated at block 1305 in FIG. 13A, a comparison of the query Feature Set
against all other Feature Sets in the field of search is used to produce a
ranked list of the other Feature Sets,” and “[t]his ranked list can be used to
display the other Feature Sets from the field of search in descending order,
with the most highly correlated (or otherwise most relevant) other Feature
Set listed first, at the top of the list,” “the resulting ranked list . . . presented
as a result of the query via a user interface.”).
Appellant’s arguments that Kupershmidt is deficient (because
Kupershmidt’s ranking is based only on statistics pre-computed “before any
query has been input” and the ranking is not “adjusted in using the user
query”) are not commensurate with the scope of claim 1. See Appeal Br. 6
and Reply Br. 2–3 (emphases added); see also In re Self, 671 F.2d 1344,
1348 (CCPA 1982) (“[A]ppellant’s arguments fail from the outset because
. . . they are not based on limitations appearing in the claims.”); In re Hiniker
Co., 150 F.3d 1362, 1369 (Fed. Cir. 1998) (“[The] proffered facts . . . are not
commensurate with the claim scope and are therefore unpersuasive.”).
Claim 1 recites “a final ranking of findings [in the database’s contents]
consisting of the rank of the record, category, or network finding, adjusted
for query term hits on the natural language” (emphasis added). However,
claim 1 does not require the statistics of the claimed “final ranking . . .
adjusted for query term hits” to be calculated in real-time based on a user’s
input, such that the claim does not preclude a “final ranking” that is based on
pre-computed statistics (e.g., statistics pre-computed for curated, pre-
selected query terms, as in Kupershmidt). Ans. 4; see Kupershmidt ¶ 164.
Appeal 2019-003377
Application 14/719,246

6
Claim 1 is therefore broad enough to include a “final ranking of
findings” that is pre-computed based on possible query terms (query terms
that are made available to users for searching after the pre-computed
statistics of final rankings are completed, as in Kupershmidt, see ¶¶ 162–
164). More particularly, as the Examiner finds, Kupershmidt’s “correlation
scores” between Feature Sets and Groups disclose the claimed “final
ranking.” Final Act. 8 (citing Kupershmidt ¶¶ 111–112, 164–165); Ans. 3–4
(citing Kupershmidt Fig. 1 (step 108–Perform correlation scoring of each
imported Feature Set to all other Feature Sets and Feature Groups in the
Knowledge Base)). Thus, Kupershmidt’s Knowledge Base contents include
final rankings of findings consisting of ranks of records, categories, or
network findings (correlation scores between Feature Sets and Groups, see
Kupershmidt ¶¶ 111–112, 127, 164) adjusted for query term hits (identified
instances of “overlapped features” or terms “in common” between Feature
Sets and Groups, such as instances of overlapping signaling pathways,
genes, and other features that become user-searchable after the Knowledge
Base is generated, see id. ¶¶ 64, 109 (emphasis added), 111 (emphasis
added), 127 (emphasis added), 162, 164) on the natural language (tags and
keywords identifying features of Feature Sets, see id. ¶¶ 57, 66, 106, 162), as
recited in claim 1.
Accordingly, we sustain the Examiner’s anticipation rejection of
independent claim 1, and dependent claims 2–9 not separately argued.

Claims 10–20
With respect to independent claim 10, Appellant argues Kupershmidt
does not disclose
Appeal 2019-003377
Application 14/719,246

7
the processor being configured to execute the instructions and
thereby: retrieve findings from a plurality of different data
sources each storing a plurality of records each including a text
field, the findings being based on a match of the query term with
the text field, [and] calculate a statistical test result of the
findings, each test result including a p-value and a strength of
effect.
Appeal Br. 6; see also Reply Br. 4. Appellant argues Kupershmidt does not
“calculat[e] a statistical test result of findings that are based on a match of a
query term with a text field” and does not “calculate a weighted test result
using the strength of effect,” because Kupershmidt’s statistical calculations
“do not occur after a query is received.” Appeal Br. 6–7. Instead,
Kupershmidt’s statistical calculations are pre-computed during the
Knowledge Base production process of Figure 1 “before any query is or can
be submitted by a user.” Id. According to Appellant, “the only calculation
of any statistics in Kupershmidt occurs before any query is received and
therefore necessarily does not and cannot be a calculation of findings that
are only known after a search engine query is input by a user.” Id.
Appellant’s arguments do not persuade us of Examiner error. The
Examiner finds, and we agree, that Kupershmidt’s processor receives a
search engine query input by a user that includes a query term, and ranks
query results to be provided to the user in response to the search engine
query, as required by claim 10. Ans. 5 (citing Kupershmidt ¶ 162, Fig. 13A
(showing receipt of a search engine query input by a user at step 1301, and
output of ranked query results to the user at step 1307, in response to the
search engine query)); Final Act. 10–11 (citing Kupershmidt ¶¶ 164–165).
Appellant’s arguments (that Kupershmidt does not disclose the
claimed calculations because “the only calculation of any statistics in
Appeal 2019-003377
Application 14/719,246

8
Kupershmidt occurs before any query is received and therefore necessarily
does not and cannot be a calculation of findings that are only known after a
search engine query is input by a user”) are not commensurate with the
scope of claim 10. See Appeal Br. 7 (emphases added). Claim 10 does not
recite that the “findings” (retrieved and used in the claimed calculations) are
based on a match of a query term obtained from a query previously
submitted by a user, the claim does not require the “findings” to be “only
known after a search engine query is input by a user,” and the claim does not
require the claimed statistical and weighted test calculations to occur at the
time the user submits a query, as Appellant argues. See id.
Accordingly, we conclude claim 10 is broad enough to include
calculations (of statistical and weighted test results) that are pre-computed
based on curated, pre-selected query terms (terms that are made available to
users for searching, after pre-computing the statistics of the records in a
database, as in Kupershmidt). Ans. 5 (citing Kupershmidt ¶¶ 77, 80); Final
Act. 10–11 (citing Kupershmidt ¶¶ 72, 121–127, 162, Fig. 9B). More
particularly, Kupershmidt retrieves findings, the findings being based on a
match (identified instances of “overlapped features” or terms “in common”
between Feature Sets and Groups, such as instances of overlapping signaling
pathways, genes, and other features, see Kupershmidt ¶¶ 64, 109, 111, 127)
of the query term (e.g., a signaling pathway, gene, or other Feature Set’s
feature that becomes user-searchable after the Knowledge Base is generated,
see id. ¶¶ 64, 66, 68, 162, 164) with text fields of records (records in
Kupershmidt’s Feature Sets and Groups), as required by claim 10.
Kupershmidt then performs the claimed calculations (of statistical and
Appeal 2019-003377
Application 14/719,246

9
weighted test results) based on the findings, to determine correlation scores
between Feature Sets and Groups. See id. ¶¶ 111–112, 121–127.
Accordingly, we sustain the Examiner’s anticipation rejection of
independent claim 10. We also sustain the Examiner’s anticipation rejection
of independent claim 15, reciting features similar to claim 10, and argued for
the same reasons as claim 10. Appeal Br. 7; see claim 15 (“receiving at a
processor a search engine query input by a user that includes a query term,”
“the processor retrieving findings from a plurality of different data sources
each storing a plurality of records each including a text field, the findings
being based on a match of the query term with the text field,” and “the
processor calculating a statistical test result of the findings, each test result
including a p value and a strength of effect”); see also Altiris, Inc. v.
Symantec Corp., 318 F.3d 1363, 1369–70 (Fed. Cir. 2003) (where the
language of method claims did not impose a specific order on the
performance of the method steps, finding reversible error to interpret the
method claims as requiring a specific order even though the patent’s
Specification disclosed only a single embodiment). We further sustain the
Examiner’s anticipation rejection of dependent claims 11–14 and 16–20, not
separately argued. See Appeal Br. 7.

CONCLUSION
The Examiner’s decision rejecting claims 1–20 under
35 U.S.C. § 102(a)(1) is affirmed.
DECISION SUMMARY
In summary:

Appeal 2019-003377
Application 14/719,246

10
Claims
Rejected 35 U.S.C. §
Reference(s)/
Basis Affirmed Reversed
1–20 102(a)(1) Kupershmidt 1–20

TIME PERIOD FOR RESPONSE
No time period for taking any subsequent action in connection with
this Appeal may be extended under 37 C.F.R. § 1.136(a). See
37 C.F.R. § 1.136(a)(1)(iv).
AFFIRMED