Ex Parte Lee et al

Patent Trial and Appeal Board

Mar 27, 2017

14009907 (P.T.A.B. Mar. 27, 2017)

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.
14/009,907 10/04/2013 Chae Hyun Lee 4900-0106 4992
22429 7590 03/29/2017
HAUPTMAN HAM, LLP
2318 Mill Road
Suite 1400
ALEXANDRIA, VA 22314
EXAMINER
TRAN, MAI T
ART UNIT PAPER NUMBER
2124
NOTIFICATION DATE DELIVERY MODE
03/29/2017 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):
docketing @ ipfirm. com
pair_lhhb @ firsttofile. com
EAnastasio @ IPFirm.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte CHAE HYUN LEE, MIN SOENG KIM, and JUN SUP LEE
Appeal 2017-004186
Application 14/009,907
Technology Center 2100
Before THU A. DANG, NORMAN H. BEAMER, and
MATTHEW J. McNEILL, Administrative Patent Judges.
DANG, Administrative Patent Judge.
DECISION ON APPEAL
I. STATEMENT OF THE CASE
Appellants appeal under 35 U.S.C. § 134(a) from the Examiner’s
Final Rejection of claims 1—10, which constitute all the claims pending in
this application. We have jurisdiction under 35 U.S.C. § 6(b).
We affirm.
A. INVENTION
According to Appellants, the claimed invention provides “a system
and a method, which calculate and provide a silhouette coefficient for
Appeal 2017-004186
Application 14/009,907
objectively analyzing a result of clustering massive data by using Hadoop”
(Spec. 3,11. 11-16).
B. REPRESENTATIVE CLAIM
Claim 1 is exemplary:
1. A system for analyzing a result of clustering massive data, the
system comprising:
a task management apparatus configured to divide a clustered target
file into blocks of a pre-designated size, and generate an input split
corresponding to a task pair for a reduce task for reducing input data by
combining the divided blocks;
at least one distance calculation apparatus configured to receive
allocation of the input split, and calculate a distance sum for each record
between blocks included in the input split;
at least one index coefficient calculation apparatus configured to
calculate a clustering significance verification index coefficient for each
record by using the distance sum for each record received from the at least
one distance calculation apparatus; and
an analysis apparatus configured to calculate a final significance
verification index coefficient of a corresponding cluster, by averaging the
clustering significance verification index coefficient for each record.
C. REJECTIONS
Claims 1—6 stand rejected under 35 U.S.C. § 103(a) as unpatentable
over the teachings of Lee et al. (US 2012/0182891 Al; pub. July 19, 2012)
(“Lee”), and Handley (US 2008/0256230 Al; pub. Oct. 16, 2008).
Claims 7—10 stand rejected under 35 U.S.C. § 103(a) as unpatentable
over the teachings of Dean and Ghemawat (MapReduce: Simplified Data
Processing on Larger Cluster, OSDI ’04 6th Symposium on Operating
Systems Design and Implementation 137—149 USENIX Association (2004))
(“Dean”), and Handley.
2
Appeal 2017-004186
Application 14/009,907
II. ISSUES
The principal issues before us are whether the Examiner erred in
finding that the combination of Lee and Handley teaches or suggests a
calculation apparatus configured to “calculate a distance sum for each record
between blocks included in the input split,” and a calculation apparatus
configured to “calculate a clustering significance verification index
coefficient for each record by using the distance sum for each record
received” (claim 1).
III. FINDINGS OF FACT
The following Findings of Fact (FF) are shown by a preponderance of
the evidence.
Lee
1. Lee discloses enabling cluster nodes to process in parallel a large
quantity of packets collected in a network in an open source Hadoop
distribution system (Abst.), wherein the large size data processed by Hadoop
is not stored in one computer but split into several blocks and distributed
into and stored in several computers flflf 8—10).
Handley
2. Handley is directed to “clustering algorithms that use a distance or
similarity measure to determine the proper clustering of objects” (14),
wherein a Silhouette Coefficient for a device is determined based on the
average distance of a device to other devices in its cluster and the minimum
distance of the device to a device not in its cluster (| 43). The Silhouette
Coefficient is determined by summing the distance between each pair of
devices and dividing it by the number of devices in the cluster (| 46). The
3
Appeal 2017-004186
Application 14/009,907
Silhouette Coefficient for a clustering is the average of the Silhouette
Coefficients for the devices in the clustering (| 47).
IV. ANALYSIS
Appellants contend “Handley does not disclose the feature ‘calculate a
distance sum for each record between blocks included in the input split’”
(App. Br. 8). Although Appellants concede Handley discloses that “the
Silhouette Coefficient for a device may be determined based on the average
distance of a device to other devices in its cluster,” Appellants contend
“distances between all devices included in the cluster is calculated” in
Handley, thus, “Handley has problem of increased amount of calculation”
(id.). According to Appellants, “the present invention does not calculate a
distance sum for all clustered files” but instead “calculates a distance sum
for each record between blocks included in the input split, thereby the
amount of calculation being able to be reduced” (id. at 8—9).
Appellants disagree with the Examiner’s inherency finding as to the
average distance (id. at 11). In particular, Appellants contend claim 1 recites
calculating “a clustering, significance verification index for each record by
using the distance sum for each record,” whereas “Handley merely states, in
paragraph [0043], that the Silhouette Coefficient for a device may be
determined based on the average distance of a device to the other devices in
its cluster . . .” (id., emphasis omitted).
We have considered all of Appellants’ arguments and evidence
presented. However, we disagree with Appellants’ contentions regarding the
Examiner’s rejections of the claims. We agree with the Examiner’s findings,
4
Appeal 2017-004186
Application 14/009,907
and find no error with the Examiner’s conclusion that the claims would have
been obvious over the combined teachings.
Appellants’ contentions are mainly directed to what Handley “does
not disclose,” what Handley “has [a] problem” with, or what Handley
“merely states” {id. at 8—11). However, the test for obviousness is what the
combination of references cited by the Examiner teaches or suggests to one
of ordinary skill in the art. See In re Merck & Co., 800 F.2d 1091, 1097
(Fed. Cir. 1986). Based on the record before us, we are unpersuaded the
Examiner erred in finding the combination of Lee and Handley teaches or at
least suggests the contested limitation.
Here, we agree with the Examiner’s reliance on Lee for teaching the
step of dividing “a clustered target file into blocks of a pre-designated size,”
and generating an “input split corresponding to a task pair for a reduce task
for reducing input data by combining the divided blocks” (Final Act. 4,
citing Lee 10, 16, 30, 37). In particular, similar to Appellants’ invention,
Lee discloses splitting large size data in a Hadoop system into multiple
blocks (FF 1). Similar to Appellants’ invention, where splitting large size
data processed by Hadoop into multiple blocks reduces the amount of
calculation (App. Br. 8—9), Lee also discloses such input split of large size
data processed by Hadoop. (FF 1).
Furthermore, we agree with the Examiner’s reliance on Handley to
teach and suggest the step of calculating “a distance sum” and calculating “a
clustering significance verification index coefficient for each record by using
the distance sum for each record received” (Final Act. 4, citing Handley Tflf
43 47). Similar to Appellants’ invention, Handley discloses determining a
“clustering significance verification index coefficient” (Silhouette
5
Appeal 2017-004186
Application 14/009,907
Coefficient) for objectively analyzing a result of a clustering (FF 2). In
particular, “Handley is directed to clustering algorithms that use a distance
or similarity measure to determine the proper clustering of objects, wherein
a Silhouette Coefficient for a device is determined based on an average
distance” (FF 2). We agree with the Examiner that “[a]n average distance is
the sum of distances divided by the number of distances measure,” wherein
“[i]n order to calculate an average distance, the distance sum, inherently was
calculated” (Ans. 13—14, emphasis omitted). In fact, Handley explicitly
discloses that the Silhouette Coefficient is determined by summing the
distances and dividing the sum by the number of distances measured. (FF
2).
Accordingly, we find no error with the Examiner’s finding the
combination of Lee and Handley teaches or at least suggests calculating “a
distance sum for each record between blocks included in the input split” and
calculating “a clustering significance verification index coefficient for each
record by using the distance sum for each record received” as recited in
claim 1.
The Supreme Court has stated, “[t]he 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). That is, when considering obviousness of a
combination of known elements, the operative question is thus “whether the
improvement is more than the predictable use of prior art elements according
to their established functions.” Id. at 417. The skilled artisan is “a person of
ordinary creativity, not an automaton.” Id. at 421.
6
Appeal 2017-004186
Application 14/009,907
Here, we agree with the Examiner’s reliance on Lee for teaching and
suggesting splitting a target file into a plurality of blocks and the Examiner’s
reliance on Handley for teaching and suggesting calculating a distance sum
and then determining a Silhouette Coefficient using the distance sum (Final
Act. 4). We conclude that it would have been well within the skill of one
skilled in the art to combine Handley’s teaching and suggestion of
calculating distance sum and Silhouette Coefficient in clustering into Lee’s
Hadoop system that splits a clustered file into multiple blocks with multiple
records therebetween, thereby calculating a Silhouette Coefficient for each
record between the blocks using the distance sum for each record received.
Such a combination would have been well within the skill of the art. See
KSR, 550 U.S. at 417.
Appellants have presented no evidence that combining Handley’s
calculations with Lee’s InputSplit would have been “uniquely challenging or
difficult for one of ordinary skill in the art.” Leapfrog Enters., Inc. v.
Fisher-Price, Inc., 485 F.3d 1157, 1162 (Fed. Cir. 2007) (citing KSR, 550
U.S. at 420). Instead, we find Appellants’ invention is simply a
modification of familiar prior art teachings (as taught or suggested by the
cited references) that would have realized a predictable result. KSR, 550
U.S. at 416.
Based on this record, we find no error in the Examiner’s rejection of
independent claim 1.
With respect to claim 2, Appellants repeat that “[a]s set forth above
with respect to claim 1, applied references fail to disclose the recited ‘at least
one distance calculation apparatus’” (App. Br. 12). Appellants do not
7
Appeal 2017-004186
Application 14/009,907
provide substantive arguments for claims 3—6 {id. at 13—14). Accordingly,
we also affirm the rejection of claims 2—6 over Lee and Handley.
As for claim 7, Appellants similarly disagree with the Examiner’s
finding that the distance sum is inherent in Handley’s average distance. Id at
15. However, similar to the Examiner’s finding with respect to claim 1, we
find no error with the Examiner’s reliance on Handley to teach and suggest
the step of outputting “a distance sum” and calculating “a clustering
significance verification index coefficient for each record by using the
distance sum for each record” (claim 7).
Appellants repeat the above arguments with respect to claims 8—10
(App. Br. 16—18). On this record, we also affirm the rejection of claims 7—
10 over Dean and Handley.
V. CONCLUSION AND DECISION
We affirm the Examiner’s rejections of claims 1—10 under 35 U.S.C.
§ 103(a).
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a)(l)(iv).
AFFIRMED
8