International Business Machines Corporation

Patent Trials and Appeals Board

Aug 13, 2021

2021001043 (P.T.A.B. Aug. 13, 2021)

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.
14/979,095 12/22/2015 Sinem Guven YOR920151297US1 9847
48063 7590 08/13/2021
RYAN, MASON & LEWIS, LLP
48 South Service Road
Suite 100
Melville, NY 11747
EXAMINER
CHOY, PAN G
ART UNIT PAPER NUMBER
3624
NOTIFICATION DATE DELIVERY MODE
08/13/2021 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):
nyoffice@rml-law.com
wel@rml-law.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
____________________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________________

Ex parte SINEM GUVEN, KARIN MURTHY,
and AMITKUMAR M. PARADKAR
__________________

Appeal 2021-001043
Application 14/979,095
Technology Center 3600
____________________

Before JAMES P. CALVE, CYNTHIA L. MURPHY, and AMEE A. SHAH,
Administrative Patent Judges.

CALVE, Administrative Patent Judge.

DECISION ON APPEAL

STATEMENT OF THE CASE
Pursuant to 35 U.S.C. § 134(a), Appellant1 appeals from the decision
of the Examiner to reject claims 1–12, 15–17, and 19–23, which are all of
the pending claims.2 We have jurisdiction under 35 U.S.C. § 6(b).
We REVERSE.

1 “Appellant” refers to “applicant” as defined in 37 C.F.R. § 1.42. Appellant
identifies International Business Machines Corporation as the real party in
interest. Appeal Br. 1.
2 Claims 13, 14, and 18 are cancelled. See Final Act. 2.
Appeal 2021-001043
Application 14/979,095

2
CLAIMED SUBJECT MATTER
Claims 1, 19, and 20 are independent. Representative claim 1 recites:
1. A method comprising:
obtaining, from a service management database, one or
more change tickets and one or more incident tickets relating to
an information technology infrastructure;
extracting, from the change tickets and the incident
tickets, a plurality of dimensions;
generating a first probability distribution of changes that
led to incidents for each of the plurality of dimensions;
generating a second probability distribution of changes
that did not lead to incidents for each of the plurality of
dimensions;
determining, for each of the plurality of dimensions, a
difference between the first probability distribution and the
second probability distribution;
selecting a subset of the plurality of dimensions as one or
more dimensions indicating potential causality between one or
more changes and one or more incidents based at least in part
on the determined differences between the first probability
distributions and the second probability distributions;
identifying configuration items of the information
technology infrastructure associated with the one or more
changes and the one or more incidents;
defining one or more linkages between the one or more
changes and the one or more incidents;
storing, in the service management database, the defined
linkages;
building a risk assessment model using the defined
linkages;
monitoring subsequent changes to configuration items in
the information technology infrastructure using the risk
assessment model; and
altering application of a given subsequent change to
modify operation of a given configuration item in the
information technology infrastructure responsive to said
monitoring to reduce a probability of the given subsequent
Appeal 2021-001043
Application 14/979,095

3
change leading to an incident in the information technology
infrastructure;
wherein the method is performed by at least one
processing device coupled to the information technology
infrastructure over at least one network.
Appeal Br. 31 (Claims App.).
REJECTIONS3
Claims 1–9, 12, 15–17, and 19–23 are rejected under 35 U.S.C. § 103
as unpatentable over Anand,4 Bobak,5 and Ide.6
Claims 10 and 11 are rejected under 35 U.S.C. § 103 as unpatentable
over Anand, Bobak, Ide, and Paik.7
ANALYSIS
Claims 1–9, 12, 15–17, 19–23
Rejected over Anand, Bobak, and Ide
Regarding independent claims 1, 19, and 20, the Examiner finds that
Anand obtains change and incident tickets for information technology (IT)
infrastructure, extracts plural dimensions, identifies configuration items of
the IT infrastructure for the changes and incidents, defines linkages between
the changes and incidents, and stores the defined linkages in a service
management database, and the method is performed by a processing device
coupled to the IT infrastructure over a network. Final Act. 9–10, 17–18, 20.

3 The Examiner withdrew rejections of claim 21 under 35 U.S.C. § 112(a)
for lack of a written description and claims 1–12, 15–17, and 19–23 under
35 U.S.C. § 101 as directed to an abstract idea without significantly more.
Ans. 3; see Final Act. 4–8.
4 US 2014/0129536 A1, published May 8, 2014.
5 US 2009/0171730 A1, published July 2, 2009.
6 US 2008/0010330 A1, published January 10, 2008.
7 US 6,076,088, issued June 13, 2000.
Appeal 2021-001043
Application 14/979,095

4
The Examiner finds that Bobak builds a risk assessment model using
the defined linkages, monitors subsequent changes to configuration items in
the IT infrastructure using the risk assessment model, and alters application
of a given subsequent change to modify operation of a given configuration
item in the IT infrastructure responsive to said monitoring to reduce a
probability of a given subsequent change leading to an incident in the IT
infrastructure. Id. at 10.
The Examiner finds that Ide generates a first probability distribution
of changes that led to incidents for each of the plurality of dimensions and
generates a second probability distribution of changes that did not lead to
incidents for each of the plurality of dimensions, determines for each of the
plurality of dimensions a difference between the first and second probability
distributions, and selects a subset of the plurality of dimensions as one or
more dimensions indicating potential causality between one or more changes
and one or more incidents based on the determined differences between the
first and second probability distributions. Id. at 10–11.
Appellant argues that Ide does not teach generating a first probability
distribution of changes that led to incidents for each of plural dimensions
and generating a second probability distribution of changes that did not lead
to incidents for each of plural dimensions as recited in claims 1, 19, and 20.
Appeal Br. 16. Appellant argues that Ide’s first time series data is currently
obtained data and the second time series data set is data in a normal status so
that neither data set is analogous to changes that led to incidents for each of
a plurality of dimensions or changes that did not lead to incidents for each of
the plurality of dimensions, and Ide does not select a subset of dimensions as
indicating causality between changes and incidents as claimed. Id. at 17.
Appeal 2021-001043
Application 14/979,095

5
In the Answer, the Examiner interprets the phrases “led to incidents”
and “not lead to incidents” as nonfunctional descriptive material labels that
do not distinguish the invention over the prior art for patentability. Ans. 6.
Appellant replies that the first and second probability distributions for
changes that led and did not lead to incidents are functional because they are
used to determine for each of a plurality of dimensions a difference between
the first and second probability distributions, select a subset of the plurality
of dimensions as indicating potential causality between the changes and the
incidents, and build a risk assessment model using the selected subset of
dimensions to monitor changes to configuration items in an information
technology (IT) infrastructure. Reply Br. 3. We agree.
The claimed methods, computer program products, and apparatuses
identify linkages between changes and incidents in IT systems. Spec. Title,
¶¶ 1, 17, 57. Incidents are service disruptions or outages of IT systems that
may bring an entire company to a standstill, and 80% of incidents that result
in outages are caused by changes to configuration items in IT infrastructures.
Id. ¶¶ 17, 22. The claims generate probability distributions of changes that
led to incidents and changes that did not lead to incidents for a plurality of
dimensions so dimensions can be identified that indicate potential causality
between changes and incidents (are statistically significant). Id. ¶¶ 57, 76.
Dimensions include time elapsed between a change and an incident, whether
a change or incident is associated with a configuration item (e.g., a particular
server), or whether a change and incident are of the same type or group. Id.
¶¶ 58–64. A risk assessment model is built from the defined linkages and
used to monitor and modify changes to the IT infrastructure to avoid similar
incidents in the IT infrastructure from the proposed changes. Id. ¶¶ 81–87.
Appeal 2021-001043
Application 14/979,095

6
As a result, probability distributions of changes that led to incidents
and changes that did not lead to incidents for each dimension are more than
labels or nonfunctional descriptive material. They are functional attributes.
They are used to determine differences between first and second probability
distributions of changes for each dimension so dimensions that indicate a
potential cause between changes and incidents can be selected, and linkages
can be defined between the changes and incidents. The linkages also are
used to build a risk assessment model that monitors subsequent changes to
configuration items of the IT infrastructure so the application of the changes
can be altered to reduce a probability that the subsequent changes will lead
to an incident in the IT infrastructure. Appeal Br. 31, 34–36 (Claims App.).
Therefore, generating probability distributions of changes that led to or did
not lead to incidents for plural dimensions is related functionally to the
claimed method, program, and apparatus. Reply Br. 3.
In contrast, Ide generates and compares probability distributions to
detect anomalies in dynamic systems such as production lines. Ide ¶ 3. Ide
generates probability distributions for first and second time series data sets
of observed values. Id. ¶¶ 12, 13. Then, Ide calculates a score based on the
absolute value of the difference between the cumulative probabilities of the
first and second generated probability distributions. Id. ¶ 14. The difference
between the probability distributions can indicate if an anomaly exists when
one data set reflects a “normal” status. Id. ¶¶ 43–47. Ide analyzes changes
in values of a dynamic system (e.g., a production line) at different times to
detect an anomaly (i.e., an incident). Id. Ide thus differs from the claimed
method, program, and apparatus that start with anomalies (incidents) and
link changes to the incidents to generate a risk assessment model.
Appeal 2021-001043
Application 14/979,095

7
Accordingly, we do not sustain the rejection of claims 1, 19, and 20 or
claims 2–9, 12, 15–17, and 21–23 depending therefrom.
Claims 10 and 11
Rejected Over Anand, Bobak, Ide, and Paik
The Examiner’s reliance on Paik to teach features of claims 10 and 11
does not cure the deficiencies noted above as to claim 1 from which claims
10 and 11 depend. See Final Act. 23–24. Accordingly, we do not sustain
the rejection of these claims.
CONCLUSION
The rejections of claims 1–12, 15–17, and 19–23 are reversed.
In summary:
Claim(s)
Rejected
35
U.S.C. §
Reference(s)/
Basis
Affirmed Reversed
1–9, 12,
15–17, 19–23
103 Anand, Bobak, Ide 1–9, 12,
15–17, 19–23
10, 11 103 Anand, Bobak, Ide,
Paik
10, 11
Overall
Outcome
1–12, 15–17,
19–23

REVERSED