Ex Parte Rader

Board of Patent Appeals and Interferences

Jan 24, 2012

11110291 (B.P.A.I. Jan. 24, 2012)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE BOARD OF PATENT APPEALS
AND INTERFERENCES
____________

Ex parte JEFFREY SCOTT RADER
____________

Appeal 2010-011490
Application 11/110,291
Technology Center 3600
____________

Before: JOSEPH A. FISCHETTI, MEREDITH C. PETRAVICK, and
MICHAEL W. KIM, Administrative Patent Judges.

KIM, Administrative Patent Judge.

DECISION ON APPEAL
Appeal 2010-011490
Application 11/110,291

2
STATEMENT OF THE CASE
This is an appeal from the final rejection of claims 1-4 and 6-40. We
have jurisdiction to review the case under 35 U.S.C. §§ 134 and 6 (2002).
The claimed invention is directed to the use of automated sentiment
analysis in securities research (Spec., para. [0001]). Claim 1, reproduced
below, is further illustrative of the claimed subject matter.
1. A computer system for performing financial analysis
using raw textual data stored in one or more electronic data
sources, comprising:
a computer readable memory;
a content mining search agent stored on the computer
readable memory and adapted to be executed on a processor to
search for raw textual data in the one or more electronic data
sources using natural language processing to identify relevant
raw textual data within the one or more electronic data sources
related to a particular financial entity, wherein such one or more
electronic data sources contain at least one electronic data
source that is not a company specific message board;
a sentiment analyzer stored on the computer readable
memory and adapted to be executed on a processor to determine
a nature of sentiment with respect to the particular financial
entity in the relevant raw textual data identified by the content
mining search agent and to assign a value to the nature of the
sentiment in the relevant raw textual data;
a user interface program stored on the computer readable
memory and adapted to be executed on a processor to control
the content mining search agent and the sentiment analyzer and
to display the value of the nature of the sentiment with respect
to the particular financial entity assigned by the sentiment
analyzer; and
an archive database that stores the value of the nature of
the sentiment with respect to the particular financial entity
assigned by the sentiment analyzer.

Appeal 2010-011490
Application 11/110,291

3
Claims 1-4 and 6-38 stand rejected under 35 U.S.C. § 103(a) as
unpatentable over Sanjiv R. Das, et al., Yahoo! For Amazon: Opinion
Extraction from Small Talk on the Web, 1-32 (Aug. 5, 2001) (hereinafter
“Das”) in view of Zangari (US Pub. No. 2002/0184133 A1, pub. Dec. 5,
2002); and claims 39 and 40 stand rejected under 35 U.S.C. § 103(a) as
unpatentable over Das in view of Zangari and Nigam1 (US Pub. No.
2006/0069589 A1, pub. Mar. 30, 2006).
We AFFIRM.

ISSUES
Did the Examiner err in asserting that a combination of Das and
Zangari renders obvious independent claim 12? The issue turns on whether
Das and Zangari teach away from being combined with each other.
Did the Examiner err in asserting that that a combination of Das and
Zangari renders obvious dependent claim 83? The issue turns on whether a
combination of Das and Zangari renders obvious analysis of the sentiment of
pre-selected categories of raw data.
Did the Examiner err in asserting that that a combination of Das and
Zangari renders obvious dependent claim 20?

1 The Examiner withdrew the rejection of claim 11 under 35 U.S.C. § 112,
first paragraph (Exam’r’s Ans. 12-13).
2 As Appellant sets forth the same arguments for independent claim 1 and
dependent claims 2-7, 11, and 22, we choose independent claim 1 as
representative. See 37 C.F.R. § 41.37(c)(1)(vii).
3 As Appellant sets forth the same arguments for claims 8-10, 12-19, 21, and
25-37, we choose dependent claim 8 as representative. See 37 C.F.R. §
41.37(c)(1)(vii).
Appeal 2010-011490
Application 11/110,291

4
Did the Examiner err in asserting that that a combination of Das and
Zangari renders obvious dependent claims 23 and 24?
Did the Examiner err in asserting that that a combination of Das and
Zangari renders obvious dependent claim 38?
Did the Examiner err in asserting that that a combination of Das,
Zangari, and Nigam renders obvious dependent claims 39 and 40?

FINDINGS OF FACT
We adopt the Examiner’s findings of fact, as set forth on pages 4-24
of the Examiner Answer.

ANALYSIS
Independent Claim 1
We are not persuaded the Examiner erred in asserting that a
combination of Das and Zangari renders obvious independent claim 1 (App.
Br. 11-15). Appellant asserts that Das and Zangari teach away from being
combined with each other because
[n]othing in either reference provides a basis to conclude there
would be value to a potential investor in applying the data
collection system of Zangari with the sentiment analysis of Das.
Instead, both references teach away from that combination, Das
by indicating that sentiment lags stock price and Zangari by
emphasizing the importance of using validated data
(App. Br. 14). Appellant’s arguments are not persuasive.
Appellant asserts that Das concludes that “sentiment is reactive to
stock movements,” thus precluding the use of sentiment as predictive (App.
Br. 11-12). However, no such predictive aspect is recited in independent
claim 1. See CollegeNet, Inc. v. ApplyYourself, Inc., 418 F.3d 1225, 1231
Appeal 2010-011490
Application 11/110,291

5
(Fed. Cir. 2005) (while the specification can be examined for proper context
of a claim term, limitations from the specification will not be imported into
the claims). Indeed, independent claim 1 appears to recite analyzing and
storing past sentiments, much like that disclosed in Das. Moreover, Zangari
is directed to “verifying the integrity of data in a data warehouse,” which
also does not require predictive abilities.
Appellant also asserts that “the Das model uses only a single company
specific message board as its data source. Das further discourages the
analysis of expanded volumes of data because ‘the overwhelming amount of
data would challenge the performance of any algorithm’ for sentiment
analysis” (App. Br. 12). However, Appellant omits the disclosure
immediately preceding context of the above quotation of Das, which recites
“if all stocks are taken together.” Like independent claim 1, Das discloses
sentiment searches for a subset of “all stocks,” for which the amount of data
would not be as “overwhelming,” even when combined with other data
sources, such as that in Zangari. Moreover, page 31 of Das further discloses
that “this is now possible with the enormous amount of data that may be
thrown at the problem,” showing that, by itself, sheer volume of data is not
problematic. The problem Appellant sets forth is one of raw computational
power. In other words, Das implicitly discloses that more data is preferable
if the computational power exists, showing that expanding data and data
sources is obvious.
Appellant further asserts that
[o]ne of ordinary skill in the art would be deterred by this
explicit emphasis on only using validated financial data in the
predictive Zangari invention from combining Zangari with data
gathered from individual internet posters measuring feelings
Appeal 2010-011490
Application 11/110,291

6
and emotions and which one of ordinary skill in the art would
believe lags behind stock performance as disclosed in Das
(App. Br. 13). Appellant misinterprets Zangari. The portions of Zangari
cited by Appellant are concerned with the integrity of the data itself, i.e.,
whether the data is accurate and unaltered. Applied to Das, Zangari would
be concerned with whether the message board posts were genuine as
opposed to corrupted, and not whether the substance of the opinions
themselves are accurate in their predictions. Specifically, Zangari is
concerned with ensuring the sentiment of “I believe stock X will increase” is
accurately transferred to the data set, as opposed to whether stock X actually
increases or decreases. Given this, we are not persuaded Zangari teaches
away.
Appellant asserts that
the Examiner's rationale for combining Zangari and Das (as
well as Nig[]am) is the mere conclusory statement that it would
be obvious to one skilled in the art at the time of the invention
to combine the sentiment analysis of Das with the data
collection of Zangari because the claimed invention is simply a
combination of old elements and, in the combination, each
element performs the same function as it does separately, and
the results of the combination are predictable. The Examiner's
analysis does not include any discussion of what those
functions are or what result from the combination would
predictabl[y]arise.
(App. Br. 14). However, Zangari discloses that using various data sources is
obvious (para. [0041]). Thus, the rationale to combine was implicit within
Zangari itself.
Appellant additionally asserts that Das and Zangari do not render
obvious “to control the content mining search agent and the sentiment
analyzer and to display the value of the nature of the sentiment with respect
Appeal 2010-011490
Application 11/110,291

7
to the financial entity assigned by the sentiment analyzer,” as recited in
independent claim 1 (App. Br. 15). We adopt the Examiner’s reasoning and
rationale, as set forth on page 15 of the Examiner’s Answer.
Appellant asserts that the rejections of dependent claims 2-7 and 13
should not be sustained for the same reason the rejection of independent
claim 1 should not be sustained (App. Br. 15-16, 19). For the reasons set
forth above, these rejections are sustained.

Dependent Claim 8
We are not persuaded the Examiner erred in asserting that a
combination of Das and Zangari renders obvious dependent claim 8 (App.
Br. 16-17). Appellant asserts Das and Zangari do not render obvious
analysis of the sentiment of pre-selected categories of raw data. We adopt
the Examiner’s reasoning and rationale, as set forth on pages 16-17 of the
Examiner’s Answer.

Dependent Claim 20
We are not persuaded the Examiner erred in asserting that a
combination of Das and Zangari renders obvious dependent claim 20 (App.
Br. 21-22). Appellant asserts that
[a]s specifically defined in the original specification, financial
performance is related to perceived market performance for a
specific security as opposed to actual market performance as
measured by stock price[ ]. The Das reference to stock price as
a category relates to the stock price data itself, not as a category
of data for which sentiment was to be analyzed
(emphasis original). We adopt the Examiner’s reasoning and rationale, as
set forth on page 18 of the Examiner’s Answer.
Appeal 2010-011490
Application 11/110,291

8
Dependent Claims 23 and 24
We are not persuaded the Examiner erred in asserting that a
combination of Das and Zangari renders obvious dependent claims 23 and
24 (App. Br. 23-24). Appellant asserts that “Das does not disclose allowing
a user to select the particular financial entity for which relevant raw textual
data will be retrieved and analyzed.” We adopt the Examiner’s reasoning
and rationale concerning dependent claims 23 and 24, as set forth on pages
19-21 of the Examiner’s Answer. In particular, page 26 of Das discloses
particular stocks.

Dependent Claim 38
We are not persuaded the Examiner erred in asserting that a
combination of Das and Zangari renders obvious dependent claim 38 (App.
Br. 27-28). Appellant asserts that “[w]hile the Examiner alleges without
citation that Das discloses blogs and user groups, Appellant found reference
only to stock market message boards and chat rooms.” We adopt the
Examiner’s reasoning and rationale, as set forth on page 21 of the
Examiner’s Answer.
Appellant also asserts that
the Appellant's combination of these disparate sources is both
novel and nonobvious. The Examiner's explanation that these
data sources, known in the prior art for use in different ways
(with validated financial date used to predict market behavior
and small investor sentiment reacting to market behavior) when
combined would perform as expected is nonsensical and
conclusory.
For that, we reference our analysis concerning independent claim 1.

Appeal 2010-011490
Application 11/110,291

9
Dependent Claims 39 and 40
We are not persuaded the Examiner erred in asserting that a
combination of Das, Zangari, and Nigam renders obvious dependent claims
39 and 40 (App. Br. 28-32). Appellant asserts that
Nigam's disclosed algorithm is used solely to count the number
of times a relevant word appears in a document and account for
appearance of those words in the sentiment value assigned to
that single document for purposes of determining relevance.
Nigam's weighting is an automatic function of the programming
language that determines relevance or irrelevance. As described
in the pending application, Claim 39 is directed to an entirely
different function.
We adopt the Examiner’s reasoning and rationale, as set forth on pages 23-
24 of the Examiner’s Answer. Specifically, Nigam is solely cited for
disclosing weighted analysis in the context of text analysis, which is even
supported by paragraphs [0044]-[0045] of Nigam’s provisional application.
That Nigam’s weighting is functionally different from the claimed invention
is irrelevant, as the Examiner has cited Das and Zangari for disclosing the
balance of the function. See In re Merck & Co. Inc., 800 F.2d 1091, 1097
(Fed. Cir. 1986) (the argument that a single reference alone does not disclose
the recited claimed steps is not persuasive because nonobviousness cannot
be established by attacking the references individually when the rejection is
predicated upon a combination of prior art disclosures); In re Keller, 642
F.2d 413, 426 (CCPA 1981) (“one cannot show non-obviousness by
attacking references individually where, as here, the rejections are based on
combinations of references”).

Appeal 2010-011490
Application 11/110,291

10
DECISION
The decision of the Examiner to reject claims 1-4 and 6-40 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). See 37 C.F.R.
§ 1.136(a)(1)(iv).
AFFIRMED

hh