KING.COM LIMITED

Patent Trials and Appeals Board

Jul 28, 2020

2019006593 (P.T.A.B. Jul. 28, 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.
14/546,248 11/18/2014 Simon WOODWARD 365569.00074 9357
78905 7590 07/28/2020
Saul Ewing Arnstein & Lehr LLP (Philadelphia)
Attn: Patent Docket Clerk
Centre Square West
1500 Market Street, 38th Floor
Philadelphia, PA 19102-2186
EXAMINER
PAULINO, LENIN
ART UNIT PAPER NUMBER
2193
NOTIFICATION DATE DELIVERY MODE
07/28/2020 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):
patents@saul.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
____________________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________________

Ex parte SIMON WOODWARD
____________________

Appeal 2019-006593
Application 14/546,248
Technology Center 2100
____________________

Before ROBERT E. NAPPI, JOHNNY A. KUMAR, and
JAMES W. DEJMEK, Administrative Patent Judges.

DEJMEK, Administrative Patent Judge.

DECISION ON APPEAL
Appellant1 appeals under 35 U.S.C. § 134(a) from a Non-Final
Rejection of claims 11–13, 17, 25–27, and 30–38. Appellant has canceled
claims 1–10, 14–16, 18–24, 28, and 29. See Amdt 2–7 (filed May 7, 2018).
We have jurisdiction over the remaining pending claims under 35 U.S.C.
§ 6(b). See Ex parte Lemoine, 46 USPQ2d 1420, 1423 (BPAI 1994)
(precedential).
We reverse.

1 Throughout this Decision, we use the word “Appellant” to refer to
“applicant” as defined in 37 C.F.R. § 1.42 (2018). Appellant identifies
King.com Limited as the real party in interest. Appeal Br. 2.
Appeal 2019-006593
Application 14/546,248

2
STATEMENT OF THE CASE
Introduction
Appellant’s disclosed and claimed invention generally relates to the
testing of computer executable applications and the retrieval of test event
data for analysis. Spec. ¶¶ 1, 6–7, 16–17. Appellant describes a system for
testing different options (i.e., an A/B test) of a computer executable
application delivered to a population of users. Spec. ¶¶ 6, 15. The
differences may reflect variations of a particular feature to determine an
overall preference among the user population. Spec. ¶ 15. Appellant
describes the creation of a per-user-per-test identifier that serves (after
processing, including a hash function) as an index to (i) determine which
variant of the computer executable the user is given; and (ii) associate event
data with the user’s experience using the application for analysis purposes.
Spec. ¶¶ 6–7, 16–17. According to the Specification:
Advantages of the above methods lie in efficiently
allocating a user to a test group in a manner that is
computationally efficient and two-way, avoiding absolute look-
up of users and their associated groups, especially once a test
option has been assigned, since the test option may be returned
by simply running the same hashing and indexing procedure at
the analysis server rather than searching a database for the User
ID and associated groups.
Spec. ¶ 88; see also Spec. ¶ 60. Of note, the performing of the hash
operation on the per-user-per-test identifier is done at the time a testing
variant is allocated to a user and again at the time event data is being
retrieved for analysis. See Spec. ¶¶ 77–78, 91–93, Figs. 5, 7.
Claim 11 is illustrative of the subject matter on appeal and is
reproduced below with the disputed limitations emphasized in italics:
Appeal 2019-006593
Application 14/546,248

3
11. A computer device for analysing event data associated
with an application under test, the device comprising:
an event database storing a data structure storing at least
one user identifier associated with a user, application data and
event data associated with the user, said event data generated by
the user executing the application modified by one of multiple
test options; and
at least one processor configured in response to an
analysis query comprising a user identifier to:
request a test identifier and associated application event
data associated with that user,
combine a user identifier associated with that user, and
the test identifier, to form a per-user-per-test identifier,
perform a hash on the per-user-per-test identifier to
generate a hash of the per-user-per-test identifier,
process the hash to generate an index value,
compare said index value with a distribution of numbers
divided into multiple ranges to determine within which range the
index value falls, each range being associated with one of the text
options,
indicate one of the test options associated with the range
into which the index value falls, and
associate the indicated test option with the associated
event data for analysis.

The Examiner’s Rejections
1. Claims 11, 12, 27, 31–33, and 37 stand rejected under
35 U.S.C. § 103 as being unpatentable over Lyon et al. (US 2013/0030868
A1; Jan. 31, 2013) (“Lyon”); Sahni et al. (US 2011/0231821 A1; Sept. 22,
2011) (“Sahni”); and Santini et al. (US 2014/0329585 A1; Nov. 6, 2014)
(“Santini”). Non-Final Act. 4–9.
Appeal 2019-006593
Application 14/546,248

4
2. Claims 13, 17, 25, 26, 30, and 34 stand rejected under
35 U.S.C. § 103 as being unpatentable over Lyon, Sahni, Santini, and Ahuja
et al. (US 2013/0246334 A1; Sept. 19, 2013) (“Ahuja”). Non-Final Act. 10–
12.
3. Claims 35 and 38 stand rejected under 35 U.S.C. § 103 as being
unpatentable over Lyon, Sahni, Santini, and Bain, III (US 8,209,549 B1;
June 26, 2012) (“Bain”). Non-Final Act. 12–14.
4. Claim 36 stand rejected under 35 U.S.C. § 103 as being
unpatentable over Lyon, Sahni, and Bain. Non-Final Act. 15–17.

ANALYSIS2
Independent claim 11 (as well as independent claims 12 and 36)
relates to analyzing the event data associated with an application under test.
For example, claim 11 is directed to a computer device for analyzing event
data associated with an application under test; claim 12 recites a method
“executed by an analytics server” to request event data generated from
executing the modified application; and claim 36 describes an analytics
server “configured to execute the same sequence of steps . . . for analysing
event data.” In each of the independent claims, a user identifier and a test
identifier are combined to form a per-user-per-test identifier. A hash is

2 Throughout this Decision, we have considered the Appeal Brief, filed
March 25, 2019 (“Appeal Br.”); the Reply Brief, filed September 5, 2019
(“Reply Br.”); the Examiner’s Answer, mailed July 5, 2019 (“Ans.”); and
the Non-Final Office Action, mailed September 24, 2018 (“Non-Final
Act.”), from which this Appeal is taken.
Appeal 2019-006593
Application 14/546,248

5
performed on the per-user-per-test identifier, is further processed, and is
used to associate the test option with the event data for analysis.
Appellant asserts the cited prior art references, as relied upon by the
Examiner do not teach or suggest generating3 a hash of a per-user-per-test
identifier to retrieve event data associated with an application under test—
i.e., at the time of analysis. Appeal Br. 7–13; Reply Br. 2–4. In particular,
Appellant argues Lyon does not generate a hash on a per-user-per-test
identifier in response to an analysis query. Appeal Br. 9. Rather, Appellant
argues Lyon’s use of a hash function is done only at the test allocation stage.
Appeal Br. 9. Appellant additionally asserts that when Lyon seeks to
analyze the event data associated with the execution of an application under
test, Lyon suggests a database function associating users with their allocated
test options. Appeal Br. 8 (citing Lyon ¶ 64).
Moreover, Appellant asserts the additional references do not cure the
deficiency of Lyon. In particular, Appellant argues Sahni is silent regarding
how any analysis is performed, Santini “does not disclose any analysis of
test data at all,” and Bain does not disclose analyzing event data associated
with test options. Appeal Br. 9–13.

3 We note that in the Appeal Brief, Appellant describes the step as
“regenerating” the hash value. See Appeal Br. 7. The Examiner correctly
notes (and Appellant acknowledges) the claims do not actually recite
“regenerating” the hash value, but rather “perform[ing] a hash on the per-
user-per-test identifier to generate a hash.” Ans. 3–4; see also Reply Br. 2–
3. Appellant explains, and we agree, the initial word choice of
“regenerating” was an unfortunate selection and used because in the overall
system, the hash is generated twice—first as part of test selection and second
as part of test analysis. See Reply Br. 2–3.
Appeal 2019-006593
Application 14/546,248

6
We are mindful that the test for obviousness is whether the
combination of references, taken as a whole, would have suggested the
patentee’s invention to a person having ordinary skill in the art. In re Merck
& Co., 800 F.2d 1091, 1097 (Fed. Cir. 1986).
As an initial matter, it is unclear as to what the Examiner is relying on
Bain to teach in the rejection of independent claim 36.4 Nonetheless, the
Examiner explains that neither Santini nor Bain were relied upon to teach
the generation of a hash of a per-user-per-test identifier. Ans. 4. In addition,
the Examiner explains that Sahni was relied on only to teach unification of a
user identifier and a test identifier to create a single ID (i.e., the per-user-per-
test identifier). Ans. 3. Thus, we must determine whether Lyon teaches
generating a hash on an identifier (as taught by Sahni) to associate a test
option with event data for analysis. Here, we agree with Appellant that
Lyon’s disclosure of using a hashed value of an identifier relates to the
allocation of the test option to a user (i.e., Lyon’s enrollment bucket). See,
e.g., Lyon ¶¶ 44, 45, 48, 54. Moreover, the Examiner has not provided
sufficient persuasive evidence or technical reasoning that an ordinarily
skilled artisan would have generated (again) a hash of the identifier to
associate a test option with the event data of the user for analysis.
Thus, even when considered in combination, we agree with Appellant
that, as relied on by the Examiner, the cited portions of the references do not
teach or suggest performing a hash function on a per-user-per-test identifier

4 Although the Examiner references Bain in the statement of rejection of
claim 36, Bain is not relied upon in the body of the rejection. See Non-Final
Act. 15–17.
Appeal 2019-006593
Application 14/546,248

7
to associate an indicated test option with the associated event data for
analysis.
On the record before us, and for the reasons discussed supra, we do
not sustain the Examiner’s rejections of independent claims 11, 12, and 36.
For similar reasons we do not sustain the Examiner’s rejections of claims 13,
17, 25–27, 30–35, 37, and 38, which depend directly or indirectly therefrom.

CONCLUSION
We reverse the Examiner’s decision rejecting claims 11–13, 17, 25–
27, and 30–38 under 35 U.S.C. § 103.

DECISION SUMMARY
Claims
Rejected
35 U.S.C. § Reference(s)/Basis Affirmed Reversed
11, 12, 27,
31–33, 37
103 Lyon, Sahni,
Santini
11, 12,
27, 31–
33, 37
13, 17, 25,
26, 30, 34
103 Lyon, Sahni,
Santini, Ahuja
13, 17,
25, 26,
30, 34
35, 38 103 Lyon, Sahni,
Santini, Bain
35, 38
36 103 Lyon, Sahni, Bain 36
Overall
Outcome
11–13,
17, 25–
27, 30–38

REVERSED