Ex Parte Sayal

Patent Trial and Appeal Board

Nov 21, 2014

10873556 (P.T.A.B. Nov. 21, 2014)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________________

Ex parte MEHMET SAYAL
____________________

Appeal 2012-007003
Application 10/873,556
Technology Center 2100
____________________

Before: JOSEPH L. DIXON, JAMES R. HUGHES, and ERIC S. FRAHM,
Administrative Patent Judges.

DIXON, Administrative Patent Judge.

DECISION ON APPEAL

Appeal 2012-007003
Application 10/873,556
2
STATEMENT OF CASE
Appellant appeals under 35 U.S.C. § 134 from a rejection of claims 1–
3, 5, 6, 8–14, and 17–26. Claims 4, 7, 15, and 16 have been canceled. We
have jurisdiction under 35 U.S.C. § 6(b).
We affirm.

The claims are directed to system and method for correlation of time-
series data. Claim 1, reproduced below, is illustrative of the claimed subject
matter:
1. A processor-based method for discovering time
correlations among data, comprising:

inputting time-series data including multiple time-series
data streams from a processor-readable medium;

summarizing the time-series data at different time
granularities;

detecting change points in the time-series data;

reducing a many-to-many comparison or a many-to-one
comparison of the multiple time-series data streams to a one-to-
one comparison;

performing the one-to-one comparison of the multiple
time-series data streams based on the change points to generate
correlation rules; and

detecting correlations between the time-series data based
on the correlation rules.

Appeal 2012-007003
Application 10/873,556
3

REFERENCES
The prior art relied upon by the Examiner in rejecting the claims on
appeal is:
Whight US 2001/0003821 A1 June 14, 2001
Baker US 6,336,103 B1 Jan.1, 2002
McGee US 2003/0088542 A1 May 8, 2003
Landau US 2004/0027349 A1 Feb.12, 2004
Murakami US 2004/0044613 A1 Mar. 4, 2004
Kumar US 6,834,266 B2 Dec. 21, 2004

REJECTIONS
The Examiner made the following rejections:
Claims 1, 3, 5, 6, 8–11, 13, 14, 17, 18, 20–22, 24, and 25 stand
rejected under 35 U.S.C. § 103(a) as being unpatentable over Landau, in
view of McGee, further in view of Kumar.
Claims 2, 12, and 19 stand rejected under 35 U.S.C. §103(a) as being
unpatentable over Landau, McGee and Kumar, further in view of Baker.
Claim 23 stands rejected under 35 U.S.C. § 103(a) as being
unpatentable over Landau, McGee and Kumar, further in view of Murakami.
Claim 26 stands rejected under 35 U.S.C. § 103(a) as being
unpatentable over Landau, McGee and Kumar, further in view of Whight.

ANALYSIS
Appellant elected to group independent claims 1, 11, 18, and 22
together. (App. Br. 6-13). We select independent claim 1 as the
representative claim for the group and will address Appellant’s arguments
thereto.
Appeal 2012-007003
Application 10/873,556
4
Appellant contends that Landau, McGee, and the Kumar references do
not teach all of the limitations recited in claim 1, including "detecting
change points in the time-series data." (App. Br. 8).
Appellant contends that Landau reference does not detect change
points in time-series data wherein "[i]n contrast, the present claims are
directed to finding time correlations among time-series data through the
detection of specific change points within the time-series data sets, for
example, using statistical methods such as CUSUM." (App. Br. 8). We note
that the Examiner has not relied upon the Landau reference to detect change
points (Ans. 5–6), and we further find Appellant's argument is not
commensurate in scope with the express language of independent claim 1
regarding the statistical methods.
Appellant further contends that the McGee reference does not disclose
detection of change points and specifically distinguishes "detection of
specific change points within time-series data." (App. Br. 9). Appellant
further contends:
the teachings of McGee differ entirely from the method of the
present claims, because McGee fails to disclose any sort of
statistical method for tracking change points between time-
series data sets. The mention of "tracking changes" in the
histograms generated according to the teachings of McGee is
explicitly for the purpose of determining how the data values
change for each step in the histogram. See id., paras. [0046],
[0151]. In contrast, the method of the current invention
involves the detection of specific change points within time-
series data. Therefore, the Appellant respectfully asserts that
the teachings of McGee also fail to disclose detecting change
points in time-series data, as generally recited.

(App. Br. 9). Again, Appellant’s argument is not commensurate in scope
with the express language of independent claim 1 where "specific change
Appeal 2012-007003
Application 10/873,556
5
points" are not expressly recited, and Appellant asserts that the McGee
reference fails to disclose detecting change points in time-series data as
generally recited in the independent claims. We find Appellant’s general
assertion to be unpersuasive of error in the Examiner's conclusion of
obviousness of representative independent claim 1.
Appellant further contends in addition to the above differences that:
In addition to the differences discussed above, independent
claim 1 recites "reducing a many-to-many comparison or a
many-to-one comparison of the multiple time-series data
streams to a one-to-one comparison." … As discussed in the
Specification, a "one-to-one" comparison may be defined as the
simplest form of time-series comparison, whereby two time-
series data streams are compared with each other to find
correlations. See Specification, paras. [0027]-[0031]. This type
of comparison can be simpler than many-to-many comparisons
and many-to-one comparisons of multiple time-series data
streams. See id. The reduction of many-to-many comparisons
or many-to-one comparisons to one-to-one comparisons, as
recited by the present claims, may be accomplished using
reduction techniques such as convolution. See id., paras.
[0031]-[0033].

(App. Br. 10). Appellant summarizes the disclosures of the references and
generally contends that the references fail to disclose reducing a many-to-
many comparison or a many-to-one comparison of the multiple time-series
data streams to a one-to-one comparison, as broadly claimed. (App. Br. 11-
12).
The Examiner maintains that the technique of convolution is not
recited in the express language of independent claim 1 and therefore
Appellant's argument is not persuasive of error in the Examiner's conclusion
of obviousness. (Ans. 17–18). The Examiner maintains that the broadest
reasonable interpretation of the claimed "reducing" step is taught and
Appeal 2012-007003
Application 10/873,556
6
suggested by Kumar’s clustering and weighted average. (Ans. 18). We
agree with the Examiner's claim interpretation and findings and adopt them
as our own. Therefore, Appellant has not shown error in the Examiner's
underlying factual findings and ultimate conclusion of obviousness, and we
sustain the Examiner's rejection of representative independent claim 1 and
independent claims 11, 18, and 22 and their respective dependent claims
grouped therewith by Appellant.
With respect to dependent claims 2, 12, and 19, Appellant argues the
claims as a group. (App. Br. 13). Therefore, we select dependent claim 2 as
the representative claim for the group and will address Appellant's
arguments thereto. Appellant relies upon the arguments advanced with
respect to independent claim 1 and further contends that Baker does not
detect change points and reducing reduce a many-to-many or many-to-one
comparison to a one-to-one comparison of time-series data sets. (App. Br.
13-14). The Examiner does not rely upon Baker to teach or suggest
detecting change points. The Examiner relies upon the Baker reference to
teach and suggest the use of convolution to achieve optimum correlation.
(Ans. 19). Therefore, Appellant’s argument does not show error in the
Examiner's conclusion of obviousness of representative dependent claim 2.
With respect to dependent claim 23, Appellant relies upon the
arguments advanced with respect to independent claim 1 which we found
unpersuasive of error (supra). (App. Br. 14–15). Since we found those
arguments unpersuasive in the ultimate conclusion of obviousness, we
sustain the rejection of dependent claim 23.
With respect to dependent claim 26, Appellant contends that "[t]o the
extent that Whight discloses the use of the CUSUM statistical method,
Whight does not disclose the detection of specific change points according
Appeal 2012-007003
Application 10/873,556
7
to such a method" and repeats the arguments advanced with respect to
independent claim 1. (App. Br. 17–18). As discussed above, we find
Appellant's general arguments concerning "specific change points" to be of
incommensurate scope with the express claim language, and we find
Appellant's arguments to independent claim 1 to be unpersuasive of error.
Therefore, we sustain the rejection of dependent claim 26.

CONCLUSION
The Examiner did not err in rejecting claims 1–3, 5, 6, 8–14, and 17–
26 based upon obviousness.

DECISION
For the above reasons, the Examiner’s obviousness rejections of
claims 1–3, 5, 6, 8–14, and 17–26 are sustained.

No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a)(1)(iv) (2011).

AFFIRMED

lv