CHARACTERIZING COMPUTER INFRASTRUCTURE USING MACHINE LEARNING TECHNIQUES

Application No. 18/099,586

• Filed: Jan 20, 2023
• Examiner: GORMLEY, AARON PATRICK
• Art Unit: 2148
• Tech Center: 2100 — Computer Architecture & Software
• Assignee: DELL PRODUCTS, L.P.
• OA Round: 2 (Final)

Office Action

DETAILED ACTION
This action is in response to the application filed 01/20/2023. Claims 1-20 are pending and have been examined.

Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .

Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.

Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed inventions are directed to non-statutory subject matter without significantly more.

Claim 1
Step 1: The claim recites “A computer-implemented method”, and is therefore directed to the statutory category of process
Step 2A Prong 1: The claim recites the following judicial exception(s)
generating, using the at least one machine learning model, at least one additional label for at least one additional computer infrastructure element using information corresponding to one or more user interactions with the at least one additional computer infrastructure element and configuration information associated with the at least one additional computer infrastructure element: This can be performed as a mental process. One can merely imagine a label for the infrastructure element related to user interactions and configuration info related to it.
Step 2A Prong 2: The judicial exception(s) are not integrated into a practical application through the following additional element(s)
obtaining at least one machine learning model, wherein the at least one machine learning model is trained, using a set of training data, to generate one or more labels for one or more of a plurality of computer infrastructure elements, and wherein the training data is based at least in part on information corresponding to one or more user interactions with at least a portion of the plurality of computer infrastructure elements and configuration information associated with at least a portion of the plurality of computer infrastructure elements: This constitutes mere data reception and is insignificant extra-solution activity (MPEP 2106.05(g)).
generating, using the at least one machine learning model, at least one additional label for at least one additional computer infrastructure element using information corresponding to one or more user interactions with the at least one additional computer infrastructure element and configuration information associated with the at least one additional computer infrastructure element: This is mere instruction to apply a judicial exception with a generic data structure (MPEP 2106.05(f)).
performing one or more automated actions related to the at least one additional computer infrastructure element based at least in part on the at least one additional label, wherein the at least one additional computer infrastructure element comprises at least one software infrastructure element, and wherein at least one of the one or more automated actions comprises at least one maintenance operation for the at least one software infrastructure element: This is mere instruction to perform a generic maintenance operation based on a judicial exception in a generic manner (MPEP 2106.05(f)).
wherein the method is performed by at least one processing device comprising a processor coupled to a memory: This is mere instruction to execute the judicial exception with generic computing hardware (MPEP 2106.05(f)).
Step 2B: The following additional element(s) of the claim, taken alone or in combination, do not amount to significantly more than the recited judicial exception(s)
obtaining at least one machine learning model, wherein the at least one machine learning model is trained, using a set of training data, to generate one or more labels for one or more of a plurality of computer infrastructure elements, and wherein the training data is based at least in part on information corresponding to one or more user interactions with at least a portion of the plurality of computer infrastructure elements and configuration information associated with at least a portion of the plurality of computer infrastructure elements: This is an instance of retrieving data from memory, a limitation known to be well-understood, routine, and conventional (MPEP 2106.05(d) II. iv.)
generating, using the at least one machine learning model, at least one additional label for at least one additional computer infrastructure element using information corresponding to one or more user interactions with the at least one additional computer infrastructure element and configuration information associated with the at least one additional computer infrastructure element: This is mere instruction to apply a judicial exception with a generic data structure (MPEP 2106.05(f)).
performing one or more automated actions related to the at least one additional computer infrastructure element based at least in part on the at least one additional label, wherein the at least one additional computer infrastructure element comprises at least one software infrastructure element, and wherein at least one of the one or more automated actions comprises at least one maintenance operation for the at least one software infrastructure element: This is mere instruction to perform a generic maintenance operation based on a judicial exception in a generic manner (MPEP 2106.05(f)).
wherein the method is performed by at least one processing device comprising a processor coupled to a memory: This is mere instruction to execute the judicial exception with generic computing hardware (MPEP 2106.05(f)).

Claim 2
Step 1: The claim recites a process, as in claim 1
Step 2A Prong 1: The claim recites no further judicial exception(s)
Step 2A Prong 2: The judicial exception(s) are not integrated into a practical application through the further additional element(s)
wherein the set of training data comprises a set of existing labels associated with one or more of the plurality of computer infrastructure elements, and wherein the at least one additional label is different than each of the existing labels: Obtaining at least one machine learning model is still mere data reception and is insignificant extra-solution activity (MPEP 2106.05(g)).
Step 2B: The further additional element(s) of the claim, taken alone or in combination, do not amount to significantly more than the recited judicial exception(s)
wherein the set of training data comprises a set of existing labels associated with one or more of the plurality of computer infrastructure elements, and wherein the at least one additional label is different than each of the existing labels: Obtaining at least one machine learning model is still an instance of retrieving data from memory, a limitation known to be well-understood, routine, and conventional (MPEP 2106.05(d) II. iv.)

Claim 3
Step 1: The claim recites a process, as in claim 1
Step 2A Prong 1: The claim recites no further judicial exception(s)
Step 2A Prong 2: The judicial exception(s) are not integrated into a practical application through the further additional element(s)
wherein the machine learning model is trained at least in part by: generating a set of words by transforming at least one of: (i) one or more portions of the information corresponding to the one or more user interactions into a natural language format and (ii) one or more portions of the configuration information into a natural language format; and processing the set of words to generate a corresponding set of embeddings, wherein each embedding encodes one or more features of a given word in the set of words: Obtaining at least one machine learning model is still mere data reception and is insignificant extra-solution activity (MPEP 2106.05(g)).
Step 2B: The further additional element(s) of the claim, taken alone or in combination, do not amount to significantly more than the recited judicial exception(s)
wherein the machine learning model is trained at least in part by: generating a set of words by transforming at least one of: (i) one or more portions of the information corresponding to the one or more user interactions into a natural language format and (ii) one or more portions of the configuration information into a natural language format; and processing the set of words to generate a corresponding set of embeddings, wherein each embedding encodes one or more features of a given word in the set of words: Obtaining at least one machine learning model is still an instance of retrieving data from memory, a limitation known to be well-understood, routine, and conventional (MPEP 2106.05(d) II. iv.)

Claim 4
Step 1: The claim recites a process, as in claim 1
Step 2A Prong 1: The claim recites no further judicial exception(s)
Step 2A Prong 2: The judicial exception(s) are not integrated into a practical application through the further additional element(s)
wherein the at least one machine learning model comprises at least one of: a transformer-based model, a long short-term memory model, and a recurrent neural network model: Generating a label with the machine learning model is still mere instruction to apply a judicial exception with a generic data structure (MPEP 2106.05(f)).
Step 2B: The further additional element(s) of the claim, taken alone or in combination, do not amount to significantly more than the recited judicial exception(s)
wherein the at least one machine learning model comprises at least one of: a transformer-based model, a long short-term memory model, and a recurrent neural network model: Generating a label with the machine learning model is still mere instruction to apply a judicial exception with a generic data structure (MPEP 2106.05(f)).

Claim 5
Step 1: The claim recites a process, as in claim 1
Step 2A Prong 1: The claim recites the following further judicial exception(s)
assigning the at least one additional label to the at least one additional computer infrastructure element in response to one or more inputs provided by the user: This can be performed as a mental process. One can merely decide on a label for the infrastructure element in response to user inputs.
Step 2A Prong 2: The judicial exception(s) are not integrated into a practical application through the further additional element(s)
outputting the at least one additional label to a user: This is mere output of data and amounts to insignificant extra-solution activity (MPEP 2106.05(g)).
Step 2B: The further additional element(s) of the claim, taken alone or in combination, do not amount to significantly more than the recited judicial exception(s)
outputting the at least one additional label to a user: This is an instance of storing data in memory, a limitation known to be well-understood, routine, and conventional (MPEP 2106.05(d) II. iv.).

Claim 6
Step 1: The claim recites a process, as in claim 5
Step 2A Prong 1: The claim recites the following further judicial exception(s)
wherein the one or more inputs comprise one or more edits to the additional label, and wherein the assigning comprises: updating the at least one additional label based on the one or more edits; and assigning the updated at least one additional label to the at least one additional computer infrastructure element: Assigning the label to the infrastructure element can still be performed as a mental process. One can merely decide a label based on the user edits and mentally assign it to the element.
Step 2A Prong 2: The judicial exception(s) are not integrated into a practical application through the additional element(s)
Step 2B: The additional element(s) of the claim, taken alone or in combination, do not amount to significantly more than the recited judicial exception(s)

Claim 7
Step 1: The claim recites a process, as in claim 5
Step 2A Prong 1: The claim recites no further judicial exception(s)
Step 2A Prong 2: The judicial exception(s) are not integrated into a practical application through the further additional element(s)
wherein the outputting is performed in response to detecting, within a particular time period, at least one of: a threshold number of interactions with the additional computer infrastructure element by the user; a threshold number of times the user interacted with the additional computer infrastructure element; and a threshold number of actions performed by the user related to the additional computer infrastructure element: Outputting the at least one additional label is still mere data output, and is insignificant extra-solution activity (MPEP 2106.05(g)).
Step 2B: The further additional element(s) of the claim, taken alone or in combination, do not amount to significantly more than the recited judicial exception(s)
wherein the outputting is performed in response to detecting, within a particular time period, at least one of: a threshold number of interactions with the additional computer infrastructure element by the user; a threshold number of times the user interacted with the additional computer infrastructure element; and a threshold number of actions performed by the user related to the additional computer infrastructure element: Outputting the at least one additional label is an instance of storing data in memory, a limitation known to be well-understood, routine, and conventional (MPEP 2106.05(d) II. iv.).

Claim 8
Step 1: The claim recites a process, as in claim 1
Step 2A Prong 1: The claim recites no further judicial exception(s)
Step 2A Prong 2: The judicial exception(s) are not integrated into a practical application through the further additional element(s)
wherein the at least one machine learning model is retrained in response to at least one of a change to at least one label that is currently assigned to a given one of the computer infrastructure elements and a new label being assigned to at least one of the plurality of computer infrastructure elements: Obtaining at least one machine learning model is still mere data reception and is insignificant extra-solution activity (MPEP 2106.05(g)).
Step 2B: The further additional element(s) of the claim, taken alone or in combination, do not amount to significantly more than the recited judicial exception(s)
wherein the at least one machine learning model is retrained in response to at least one of a change to at least one label that is currently assigned to a given one of the computer infrastructure elements and a new label being assigned to at least one of the plurality of computer infrastructure elements: Obtaining at least one machine learning model is still an instance of retrieving data from memory, a limitation known to be well-understood, routine, and conventional (MPEP 2106.05(d) II. iv.)

Claim 9
Step 1: The claim recites a process, as in claim 1
Step 2A Prong 1: The claim recites no further judicial exception(s)
Step 2A Prong 2: The judicial exception(s) are not integrated into a practical application through the further additional element(s)
wherein the at least one maintenance operation for the at least one software infrastructure element comprises at least one of:
an update operation of the at least one software infrastructure element: This is a routine operation which fails to place meaningful limitations on the invention, and is thus insignificant extra-solution activity (MPEP 2106.05(g)).
a restore operation of the at least one additional computer infrastructure element: This is a routine operation which fails to place meaningful limitations on the invention, and is thus insignificant extra-solution activity (MPEP 2106.05(g)).
performing a reboot operation of the at least one software infrastructure element: This is a routine operation which fails to place meaningful limitations on the invention, and is thus insignificant extra-solution activity (MPEP 2106.05(g)).
Step 2B: The further additional element(s) of the claim, taken alone or in combination, do not amount to significantly more than the recited judicial exception(s)
an update operation of the at least one software infrastructure element: This is an instance of updating a computer device, a conventional technique in the art, as noted by Garel et al. (COMPUTER BASED SYSTEM FOR CONFIGURING, MANUFACTURING, TESTING, DIAGNOSING, AND RESETTING TARGET UNIT EQUIPMENT AND METHODS OF USE THEREOF, filed 6/21/2022, US 20230028513 A1): “conventional maintenance efforts utilized to maintain, reconfigure and/or update one or more computer devices require a user to manually enter commands, via a computer input device (e.g., keyboard, mouse, pointer, etc.), and manually verify that the computer device was reconfigured or updated in the manner intended” (Garel, [0005])
a restore operation of the at least one software infrastructure element: This is an instance of restoring a computer device, an operation widely known in the art, as noted by Cheng et al. (File System Based Offline Disk Management, published 9/27/2007, US 20070226436 A1): “Disk imaging is widely used to backup and restore data on computer storage media” (Cheng, [0001]).
performing a reboot operation of the at least one software infrastructure element: This is an instance of rebooting a computer device, a popular technique used for computer systems, as noted by Wei et al. (SYSTEM AND METHOD FOR DETECTING A WORK STATUS OF A COMPUTER SYSTEM, published 9/25/2008, US 20080235546 A1): “Presently, remote monitor and reboot methods are popularly adopted to recover halted computer systems” (Wei, [0005]).

Claim 10
Step 1: The claim recites a process, as in claim 1
Step 2A Prong 1: The claim recites no further judicial exception(s)
Step 2A Prong 2: The judicial exception(s) are not integrated into a practical application through the further additional element(s)
wherein the plurality of computer infrastructure elements corresponds to at least one datacenter and comprises: a hardware infrastructure element deployed at the at least one datacenter: This is mere instruction to apply the judicial exception(s) with generic computing hardware (MPEP 2106.05(f)).
Step 2B: The further additional element(s) of the claim, taken alone or in combination, do not amount to significantly more than the recited judicial exception(s)
wherein the plurality of computer infrastructure elements corresponds to at least one datacenter and comprises: a hardware infrastructure element deployed at the at least one datacenter: This is mere instruction to apply the judicial exception(s) with generic computing hardware (MPEP 2106.05(f)).

Claim 11
Step 1: The claim recites a process, as in claim 1
Step 2A Prong 1: The claim recites no further judicial exception(s)
Step 2A Prong 2: The judicial exception(s) are not integrated into a practical application through the further additional element(s)
providing a dashboard related to the plurality of computer infrastructure elements, wherein the dashboard is configured to at least one of:
display computer infrastructure element information corresponding to one or more of the plurality of computer infrastructure elements based at least in part on one or more labels generated using the at least one machine learning model: This is mere instruction to display information based on the generated labels in a generic manner (MPEP 2106.05(f)).
initiate one or more tasks corresponding to one or more of the plurality of computer infrastructure elements based at least in part on one or more labels generated using the at least one machine learning model: This is mere instruction to initiate a task based on the generated labels in a generic manner (MPEP 2106.05(f)).
Step 2B: The further additional element(s) of the claim, taken alone or in combination, do not amount to significantly more than the recited judicial exception(s)
providing a dashboard related to the plurality of computer infrastructure elements, wherein the dashboard is configured to at least one of:
display computer infrastructure element information corresponding to one or more of the plurality of computer infrastructure elements based at least in part on one or more labels generated using the at least one machine learning model: This is mere instruction to display information based on the generated labels in a generic manner (MPEP 2106.05(f)).
initiate one or more tasks corresponding to one or more of the plurality of computer infrastructure elements based at least in part on one or more labels generated using the at least one machine learning model: This is mere instruction to initiate a task based on the generated labels in a generic manner (MPEP 2106.05(f)).

Claim 12
Step 1: The claim recites a process, as in claim 1
Step 2A Prong 1: The claim recites no further judicial exception(s)
Step 2A Prong 2: The judicial exception(s) are not integrated into a practical application through the further additional element(s)
wherein the information corresponding to the one or more user interactions comprises at least one of: a type of interaction with a given one of the plurality of computer infrastructure elements; a number of interactions with a given one of the plurality of computer infrastructure elements; an amount of time interacting with a given one of the plurality of computer infrastructure elements; and one or more preferences associated with at least one user performing the one or more user interactions: Obtaining at least one machine learning model still amounts to mere reception of data, and is insignificant extra-solution activity (MPEP 2106.05(g)).
Step 2B: The further additional element(s) of the claim, taken alone or in combination, do not amount to significantly more than the recited judicial exception(s)
wherein the information corresponding to the one or more user interactions comprises at least one of: a type of interaction with a given one of the plurality of computer infrastructure elements; a number of interactions with a given one of the plurality of computer infrastructure elements; an amount of time interacting with a given one of the plurality of computer infrastructure elements; and one or more preferences associated with at least one user performing the one or more user interactions: Obtaining at least one machine learning model is still an instance of retrieving information from memory, a limitation known to be well-understood, routine, and conventional (MPEP 2106.05(d) II. iv.).

Claim 13
Step 1: The claim recites a process, as in claim 1
Step 2A Prong 1: The claim recites no further judicial exception(s)
Step 2A Prong 2: The judicial exception(s) are not integrated into a practical application through the further additional element(s)
wherein the configuration information associated with a given one of the plurality of computer infrastructure elements comprises at least one of: an identifier for the given computer infrastructure element; a type of the given computer infrastructure element; a type of deployment of the given computer infrastructure element; a geographical location of the given computer infrastructure element; and at least one existing label assigned to the given computer infrastructure element: Obtaining at least one machine learning model still amounts to mere reception of data, and is insignificant extra-solution activity (MPEP 2106.05(g)).
Step 2B: The further additional element(s) of the claim, taken alone or in combination, do not amount to significantly more than the recited judicial exception(s)
wherein the configuration information associated with a given one of the plurality of computer infrastructure elements comprises at least one of: an identifier for the given computer infrastructure element; a type of the given computer infrastructure element; a type of deployment of the given computer infrastructure element; a geographical location of the given computer infrastructure element; and at least one existing label assigned to the given computer infrastructure element: Obtaining at least one machine learning model is still an instance of retrieving information from memory, a limitation known to be well-understood, routine, and conventional (MPEP 2106.05(d) II. iv.).

Claim 14
Step 1: The claim recites “A non-transitory processor-readable storage medium”, and is therefore directed to the statutory category of article of manufacture
Step 2A Prong 1: The claim recites the following judicial exception(s)
to generate, using the at least one machine learning model, at least one additional label for at least one additional computer infrastructure element using information corresponding to one or more user interactions with the at least one additional computer infrastructure element and configuration information associated with the at least one additional computer infrastructure element: This can be performed as a mental process. One can merely imagine a label for the infrastructure element related to user interactions and configuration info related to it.
Step 2A Prong 2: The judicial exception(s) are not integrated into a practical application through the following additional element(s)
A non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code when executed by at least one processing device causes the at least one processing device: This is mere instruction to execute the judicial exception with generic computing hardware (MPEP 2106.05(f)).
to obtain at least one machine learning model, wherein the at least one machine learning model is trained, using a set of training data, to generate one or more labels for one or more of a plurality of computer infrastructure elements, and wherein the training data is based at least in part on information corresponding to one or more user interactions with at least a portion of the plurality of computer infrastructure elements and configuration information associated with at least a portion of the plurality of computer infrastructure elements: This constitutes mere data reception and is insignificant extra-solution activity (MPEP 2106.05(g)).
to generate, using the at least one machine learning model, at least one additional label for at least one additional computer infrastructure element using information corresponding to one or more user interactions with the at least one additional computer infrastructure element and configuration information associated with the at least one additional computer infrastructure element: This is mere instruction to apply a judicial exception with a generic data structure (MPEP 2106.05(f)).
to perform one or more automated actions related to the at least one additional computer infrastructure element based at least in part on the at least one additional label, wherein the at least one additional computer infrastructure element comprises at least one software component, and wherein at least one of the one or more automated actions comprises a maintenance operation for the at least one software component: This is mere instruction to perform a generic maintenance operation based on a judicial exception in a generic manner (MPEP 2106.05(f)).
Step 2B: The following additional element(s) of the claim, taken alone or in combination, do not amount to significantly more than the recited judicial exception(s)
A non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code when executed by at least one processing device causes the at least one processing device: This is mere instruction to execute the judicial exception with generic computing hardware (MPEP 2106.05(f)).
to obtain at least one machine learning model, wherein the at least one machine learning model is trained, using a set of training data, to generate one or more labels for one or more of a plurality of computer infrastructure elements, and wherein the training data is based at least in part on information corresponding to one or more user interactions with at least a portion of the plurality of computer infrastructure elements and configuration information associated with at least a portion of the plurality of computer infrastructure elements: This is an instance of retrieving data from memory, a limitation known to be well-understood, routine, and conventional (MPEP 2106.05(d) II. iv.)
to generate, using the at least one machine learning model, at least one additional label for at least one additional computer infrastructure element using information corresponding to one or more user interactions with the at least one additional computer infrastructure element and configuration information associated with the at least one additional computer infrastructure element: This is mere instruction to apply a judicial exception with a generic data structure (MPEP 2106.05(f)).
to perform one or more automated actions related to the at least one additional computer infrastructure element based at least in part on the at least one additional label, wherein the at least one additional computer infrastructure element comprises at least one software component, and wherein at least one of the one or more automated actions comprises a maintenance operation for the at least one software component: This is mere instruction to perform a generic maintenance operation based on a judicial exception in a generic manner (MPEP 2106.05(f)).

Claims 15-17
Step 1: Claims 15-17 recite an article of manufacture, as in claim 14.
Step 2A Prong 1: Claims 15-17 recite the same judicial exception(s) as claims 2-4, respectively.
Step 2A Prong 2: The judicial exception(s) are not integrated into a practical application through any additional elements. The analysis of claims 15-17 at this step mirrors that of claims 2-4, respectively, with the exception that claims 15-17 are directed to “A non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code when executed by at least one processing device causes the at least one processing device”, performing operations mirroring those of claims 2-4. This is a mere instruction to apply the exceptions using generic computer equipment (MPEP 2106.05(f)).
Step 2B: The additional element(s) of the claim, taken alone or in combination, do not amount to significantly more than the recited judicial exception(s). The analysis of claims 15-17 at this step mirrors that of claims 2-4, with the exception that claims 15-17 are directed to “A non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code when executed by at least one processing device causes the at least one processing device”, performing operations mirroring those of claims 2-4. This is mere instruction to apply the exceptions using generic computer equipment (MPEP 2106.05(f)).

Claim 18
Step 1: The claim recites “An apparatus”, and is therefore directed to the statutory category of machine
Step 2A Prong 1: The claim recites the following judicial exception(s)
to generate, using the at least one machine learning model, at least one additional label for at least one additional computer infrastructure element using information corresponding to one or more user interactions with the at least one additional computer infrastructure element and configuration information associated with the at least one additional computer infrastructure element
Step 2A Prong 2: The judicial exception(s) are not integrated into a practical application through the following additional element(s)
An apparatus comprising: at least one processing device comprising a processor coupled to a memory; the at least one processing device being configured: This is mere instruction to execute the judicial exception with generic computer hardware (MPEP 2106.05(f)).
to obtain at least one machine learning model, wherein the at least one machine learning model is trained, using a set of training data, to generate one or more labels for one or more of a plurality of computer infrastructure elements, and wherein the training data is based at least in part on information corresponding to one or more user interactions with at least a portion of the plurality of computer infrastructure elements and configuration information associated with at least a portion of the plurality of computer infrastructure elements: This constitutes mere data reception and is insignificant extra-solution activity (MPEP 2106.05(g)).
to generate, using the at least one machine learning model, at least one additional label for at least one additional computer infrastructure element using information corresponding to one or more user interactions with the at least one additional computer infrastructure element and configuration information associated with the at least one additional computer infrastructure element: This is mere instruction to apply a judicial exception with a generic data structure (MPEP 2106.05(f)).
to perform one or more automated actions related to the at least one additional computer infrastructure element based at least in part on the at least one additional label, wherein the at least one additional computer infrastructure element comprises at least one software component, and wherein at least one of the one or more automated actions comprises a maintenance operation for the at least one software component: This is mere instruction to perform a generic maintenance operation based on a judicial exception in a generic manner (MPEP 2106.05(f)).
Step 2B: The following additional element(s) of the claim, taken alone or in combination, do not amount to significantly more than the recited judicial exception(s)
An apparatus comprising: at least one processing device comprising a processor coupled to a memory; the at least one processing device being configured: This is mere instruction to execute the judicial exception with generic computer hardware (MPEP 2106.05(f)).
to obtain at least one machine learning model, wherein the at least one machine learning model is trained, using a set of training data, to generate one or more labels for one or more of a plurality of computer infrastructure elements, and wherein the training data is based at least in part on information corresponding to one or more user interactions with at least a portion of the plurality of computer infrastructure elements and configuration information associated with at least a portion of the plurality of computer infrastructure elements: This is an instance of retrieving data from memory, a limitation known to be well-understood, routine, and conventional (MPEP 2106.05(d) II. iv.)
to generate, using the at least one machine learning model, at least one additional label for at least one additional computer infrastructure element using information corresponding to one or more user interactions with the at least one additional computer infrastructure element and configuration information associated with the at least one additional computer infrastructure element: This is mere instruction to apply a judicial exception with a generic data structure (MPEP 2106.05(f)).
to perform one or more automated actions related to the at least one additional computer infrastructure element based at least in part on the at least one additional label, wherein the at least one additional computer infrastructure element comprises at least one software component, and wherein at least one of the one or more automated actions comprises a maintenance operation for the at least one software component: This is mere instruction to perform a generic maintenance operation based on a judicial exception in a generic manner (MPEP 2106.05(f)).

Claims 19-20
Step 1: Claims 19-20 recite a machine, as in claim 18.
Step 2A Prong 1: Claims 19-20 recite the same judicial exception(s) as claims 2-3, respectively.
Step 2A Prong 2: The judicial exception(s) are not integrated into a practical application through any additional elements. The analysis of claims 19-20 at this step mirrors that of claims 2-3, respectively, with the exception that claims 19-20 are directed to “An apparatus comprising: at least one processing device comprising a processor coupled to a memory; the at least one processing device being configured”, performing operations mirroring those of claims 2-3. This is a mere instruction to apply the exceptions using generic computer equipment (MPEP 2106.05(f)).
Step 2B: The additional element(s) of the claim, taken alone or in combination, do not amount to significantly more than the recited judicial exception(s). The analysis of claims 19-20 at this step mirrors that of claims 2-3, with the exception that claims 19-20 are directed to “An apparatus comprising: at least one processing device comprising a processor coupled to a memory; the at least one processing device being configured”, performing operations mirroring those of claims 2-3. This is mere instruction to apply the exceptions using generic computer equipment (MPEP 2106.05(f)).

Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.

The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.

Claims 1-6, 9, and 12-20 are rejected under 35 U.S.C. 103 as being unpatentable over Iyer et al. (Mixed Initiative Approach for Reliable Tagging of Maintenance Records with Machine Learning, published 2022, ANNUAL CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2022), hereafter referred to as Iyer, in view of Muppidi et al. (CLASSIFICATION AND POLICY MANAGEMENT FOR SOFTWARE COMPONENTS, published 2/7/2012, US 8,112,370 B2), hereafter referred to as Muppidi.

Regarding claim 1, Iyer discloses [a] computer-implemented method comprising:
obtaining at least one machine learning model, wherein the at least one machine learning model is trained, using a set of training data, to generate one or more labels for one or more of a plurality of computer infrastructure elements:
“Free-form text-based maintenance and service records related to industrial assets (infrastructure elements) capture the observations and actions of service engineers and are a crucial resource for assessing system-level asset health. To facilitate tracking of historical asset health issues, these records are categorized using tags (labels) from a predefined taxonomy” (Iyer, page 1, left column, paragraph 1).
“a supervised learning approach can be implemented to automate the tagging (label[ing]) process, using Deep Learning based language models like BERT (machine learning model)” (Iyer, page 2, right column, paragraph 1)
“Figure 1 shows the overall workflow based on our approach. For the baseline model, historical semi-structured records, along with tags that were assigned to them, are provided as training data to a BERT-based classification model for supervised learning of tags (labels)” (Iyer, page 3, left column, paragraph 1)
“Aircraft maintenance and service records are critical for maintaining airworthiness of an aircraft; they carry details related to the repair performed on the aircraft” (Iyer, page 3, left column, paragraph 2). These records contain information about hardware, falling within the examples given of “computer infrastructure element[s]” in page 3, paragraph 4 of the instant specification.
…wherein the training data is based at least in part on information corresponding to one or more user interactions with at least a portion of the plurality of computer infrastructure elements and configuration information associated with at least a portion of the plurality of computer infrastructure elements:
“These records capture the observations and actions of service engineers (user[s]) and are a crucial resource for assessing system-level health of an asset and for inferring reliability issues arising from those. Typically, in these records, free text is used to describe observed issues and relevant corrective actions that were performed in response - Figure 2 shows snapshots of 2 such examples of service records” (Iyer, page 3, left column, paragraph 2)

PNG
media_image1.png
200
400
media_image1.png
Greyscale
“Two examples of maintenance/service records with red boxes indicating the ATA Chapters (tag) assigned to each” (Iyer, page 4, left column, Figure 2). These records show user interactions with infrastructure elements (investigation, cleaning, replacement) and configuration information associated with infrastructure elements (leaking gasket, gear not extending, broken shaft, part time since overhaul).
generating, using the at least one machine learning model, at least one additional label for at least one additional computer infrastructure element using information corresponding to one or more user interactions with the at least one additional computer infrastructure element and configuration information associated with the at least one additional computer infrastructure element: “Figure 1 shows the overall workflow based on our approach. For the baseline model, historical semi-structured records (records corresponding to at least one additional computer infrastructure element), along with tags that were assigned to them, are provided as training data to a BERT-based classification model for supervised learning of tags (at least one additional label)” (Iyer, page 3, left column, paragraph 1). As discussed for the previous limitation, user interaction information and configuration information can be found in the records.
performing one or more automated actions related to the at least one additional computer infrastructure element based at least in part on the at least one additional label, wherein the at least one additional computer infrastructure element comprises at least one software infrastructure element, and wherein at least one of the one or more automated actions comprises at least one maintenance operation for the at least one software infrastructure element: “the less reliable tag (additional label) assignments from the classifier are routed to the human expert for tag classification … Since the fraction of tag assignments that get routed via full automation are a good measure of efficiency of the mixed initiative system, we track this number as a metric and we call it coverage.” (Iyer, page 6, left column, paragraph 1)
While Iyer fails to disclose the further limitations of the claim, Muppidi discloses a method, wherein:
performing one or more automated actions related to the at least one additional computer infrastructure element based at least in part on the at least one additional label, wherein the at least one additional computer infrastructure element comprises at least one software infrastructure element, and wherein at least one of the one or more automated actions comprises at least one maintenance operation for the at least one software infrastructure element:
“The illustrative embodiments provide a method, system, and computer usable program product for classification and policy management, for software components (at least one software infrastructure element). A metadata associated with a component is identified. A mapping determination is made whether the metadata maps to a classification in a set of classifications. If the mapping determination is true, the component is assigned to the classification (at least one additional label) and a policy associated with the classification is associated with the component.”(Muppidi, column 2, paragraph 3)
“In addition, the policy that is applicable to the classification is identified or defined. The identified or defined policy is associated the classification (additional label). Furthermore, because of the policy being associated with the classification, associating (one or more automated actions) the policy with the component may occur automatically from assigning the component to the classification” (Muppidi, column 2, paragraph 4)
“The illustrative embodiments provide a method, system, and computer usable program product for classification and policy management, for software components” (Muppidi, column 2, paragraph 3)
“Process 900 begins by receiving a policy update (step 902) (maintenance operation). In one embodiment, a policy update may include changes to policies or policy templates already associated with classifications. In another embodiment, a policy update may include newly added policies or policy templates” (Muppidi, column 11, paragraph 4). As made clear by claim 9, a software update operation is a maintenance operation of software.
the method is performed by at least one processing device comprising a processor coupled to a memory: “Furthermore, the invention can take the form of a computer program product accessible from a computer-usable or computer-readable medium (memory) providing program code for use by or in connection with a computer or any instruction execution system” (Muppidi, column 12, paragraph 4); “A data processing system suitable for storing and/or executing program code will include at least one processor coupled directly or indirectly to memory elements through a system bus. The memory elements can include local memory 50 employed during actual execution of the program code, bulk storage, and cache memories, which provide temporary storage of at least some program code in order to reduce the number of times code must be retrieved from bulk storage during execution” (Muppidi, column 12, paragraph 8)
Iyer and Muppidi relate to automatic classification of infrastructure components and are analogous to the claimed invention. Iyer teaches a method of automatically tagging and managing infrastructure records. The claimed invention improves upon Iyer’s method by performing software updates based on tags. Muppidi teaches a method of updating software policy based on tags. A person of ordinary skill in the art would have recognized that performing software policy updates based on element tags would lead to the predictable result of automated selective maintenance on software infrastructure, and would improve the efficiency of the known device by automating different updates across different groups of software infrastructure with different requirements (MPEP 2143 I. (D) Applying a known technique to a known device (method, or product) ready for improvement to yield predictable results).
Additionally, the claimed invention improves upon this method by storing it in the form of instructions on computer hardware. Iyer teaches hardware able to run methods of automatically tagging infrastructure information, applicable to Iyer. A person of ordinary skill in the art would have recognized that storing Iyer’s method as computer instructions on Muppidi’s hardware would lead to the predictable result of the method being executable by a computing system, and would improve the known device by allowing it to be performed with real data (MPEP 2143 I. (D) Applying a known technique to a known device (method, or product) ready for improvement to yield predictable results).

Regarding claim 2, the rejection of claim 1 in view of Iyer and Muppidi is incorporated. Iyer further discloses a method, wherein the set of training data comprises a set of existing labels associated with one or more of the plurality of computer infrastructure elements, and wherein the at least one additional label is different than each of the existing labels: “For the baseline model, historical semi-structured records, along with tags (existing labels) that were assigned to them, are provided as training data to a BERT-based classification model for supervised learning of tags (additional label[s])” (Iyer, page 3, left column, paragraph 1). Tags assigned before the machine learning process are ‘existing labels’, while tags generated by the machine learning model are ‘additional labels’.

Regarding claim 3, the rejection of claim 1 in view of Iyer and Muppidi is incorporated. Iyer further discloses a method, wherein the machine learning model is trained at least in part by:
generating a set of words by transforming at least one of: (i) one or more portions of the information corresponding to the one or more user interactions into a natural language format and (ii) one or more portions of the configuration information into a natural language format:
“For BM1, the service records are first tokenized (natural language format), using the tokenizer provided with the BERT distribution, and fed to the encoder of the pre-trained BERT model” (Iyer, page 6, right column, paragraph 2)
processing the set of words to generate a corresponding set of embeddings, wherein each embedding encodes one or more features of a given word in the set of words.

PNG
media_image2.png
200
400
media_image2.png
Greyscale
“The BERT-tiny architecture with elements of the epistemic framework to generate reliability assessments for tag predictions. Boxes a, b and c indicate the primary elements of the architecture - a is the pretrained BERT model, b includes additions to perform classifications from the embedding of the pretrained model, and c shows extensions that make use of the overall architecture to generate epistemic reliability assessments for individual tag predictions by the model” (Iyer, page 4, right column, Figure 3). As seen in figure 3, representative embeddings corresponding to each token are generated by the BERT-tiny model. Thus, each embedding corresponds to the word(s) from the corresponding token.

Regarding claim 4, the rejection of claim 1 in view of Iyer and Muppidi is incorporated. Iyer further discloses a method, wherein the at least one machine learning model comprises at least one of: a transformer-based model, a long short-term memory model, and a recurrent neural network model:
“We leverage the BERT pre-trained model as made available by Google (Devlin, Chang, Lee, & Toutanova, 2018). BERT makes use of an encoder-decoder architecture that contains Transformers” (Iyer, page 3, right column, paragraph 2)
“As a first step of our approach, we develop and show outcomes from the application of one such model for the tag assignment task. Typically, an unsupervised language model is trained using a large corpus of data and then fine-tuned on the downstream task. Multiple instances of such language models exist in literature including ELMo (Embeddings from Language Models)” (Iyer, page 3, right column, paragraph 1). ELMo models are LSTM recurrent networks, as made clear by Peters et al. (Deep contextualized word representations, published 2018, arXiv:1802.05365v2): “We use vectors derived from a bidirectional LSTM that is trained with a coupled language model (LM) objective on a large text corpus. For this reason, we call them ELMo (Embeddings from Language Models) representations.” (Peters, page 1, left column, paragraph 3).

Regarding claim 5, the rejection of claim 1 in view of Iyer and Muppidi is incorporated. Iyer further discloses a method, comprising:
outputting the at least one additional label to a user: “the less reliable tag assignments (additional label[s]) from the classifier are routed to the human expert (user) for tag classification” (Iyer, page 6, left column, paragraph 1)
assigning the at least one additional label to the at least one additional computer infrastructure element in response to one or more inputs provided by the user: “our approach also identifies service records where the potential for a model, even with high statistical accuracy, to assign an incorrect tag is high, thereby warranting the need for intervention by a human expert to resolve the case and assign the right tag” (Iyer, page 3, left column, paragraph 2). The tag is updated by a human expert and assigned to the record corresponding to computer infrastructure element.

Regarding claim 6, the rejection of claim 5 in view of Iyer and Muppidi is incorporated. Iyer further discloses a method,
wherein the one or more inputs comprise one or more edits to the additional label, and wherein the assigning comprises: updating the at least one additional label based on the one or more edits; and assigning the updated at least one additional label to the at least one additional computer infrastructure element: “our approach also identifies service records where the potential for a model, even with high statistical accuracy, to assign an incorrect tag is high, thereby warranting the need for intervention by a human expert to resolve the case and assign the right tag” (Iyer, page 3, left column, paragraph 2). The tag is updated by a human expert and assigned to the record corresponding to computer infrastructure element.

Regarding claim 9, the rejection of claim 1 in view of Iyer and Muppidi is incorporated. Muppidi further discloses a method, wherein the at least one maintenance operation for the at least one software infrastructure element comprises at least one of:
an update operation of the at least one software infrastructure element:
“the illustrative embodiments provide a method, system, and computer usable program product for classification (tagging) and policy management for software components (software infrastructure element[s]). The illustrative embodiments may be used in conjunction with any application or any data processing system that may use components, including but not limited to services or web service. The illustrative embodiments are described using services as an example of components to which the illustrative embodiments are applicable. Description of illustrative embodiments using services, however, is only used as an example and is not intended to be limiting on the illustrative embodiments.” (Muppidi, [0032])
“Process 900 determines a classification (label) to which the policy update applies (step 904). Process 900 determines if any services (software infrastructure element[s]) belonging to the classification of step 904 exist or are 30 executing within the boundary of process 900 (step 906). For example, process 900 may be limited to applying policy updates to a particular data processing system. In such an example, process 900 may determine if a service classified under the classification of step 904 exists in that data process. If process 900 determines that a service belonging to the classification of step 904 exists or is executing within the boundary of process 900 ("Yes" path of step 906), process 900 enforces the policy update for that service (step 908). Process 900 ends thereafter” (Muppidi, column 11, paragraphs 5-6)
a restore operation of the at least one software infrastructure element
performing a reboot operation of the at least one software infrastructure element
Muppidi relates to automatic classification of infrastructure components and is analogous to the claimed invention. Iyer teaches a method of automatically tagging and managing infrastructure records. The claimed invention improves upon Iyer’s method by performing software updates based on tags. Muppidi teaches a method of updating software policy based on tags. A person of ordinary skill in the art would have recognized that performing software policy updates based on element tags would lead to the predictable result of automated selective maintenance on software infrastructure, and would improve the efficiency of the known device by automating different updates across different groups of software infrastructure with different requirements (MPEP 2143 I. (D) Applying a known technique to a known device (method, or product) ready for improvement to yield predictable results).

Regarding claim 12, the rejection of claim 1 in view of Iyer and Muppidi is incorporated. Iyer further discloses a method, wherein the information corresponding to the one or more user interactions comprises at least one of: a type of interaction with a given one of the plurality of computer infrastructure elements; a number of interactions with a given one of the plurality of computer infrastructure elements; an amount of time interacting with a given one of the plurality of computer infrastructure elements; one or more preferences associated with at least one user performing the one or more user interactions:
PNG
media_image1.png
200
400
media_image1.png
Greyscale
(Iyer, page 4, left column, Figure 2). These records comprise an investigation of an element (type of interaction), two reports (number of interactions), and users choosing to replace parts or disassemble elements (one or more preferences).

Regarding claim 13, the rejection of claim 1 in view of Iyer and Muppidi is incorporated. Iyer further discloses a method, wherein the configuration information associated with a given one of the plurality of computer infrastructure elements comprises at least one of: an identifier for the given computer infrastructure element; a type of the given computer infrastructure element; a type of deployment of the given computer infrastructure element; a geographical location of the given computer infrastructure element; and at least one existing label assigned to the given computer infrastructure element:
PNG
media_image1.png
200
400
media_image1.png
Greyscale
(Iyer, page 4, left column, Figure 2). These records specify the names of the element types (identifier / type), deploying an element to perform a leak check (type of deployment), and existing ATA labels (at least one existing label).

Regarding claim 14, Iyer discloses a method:
to obtain at least one machine learning model, wherein the at least one machine learning model is trained, using a set of training data, to generate one or more labels for one or more of a plurality of computer infrastructure elements:
“Free-form text-based maintenance and service records related to industrial assets (infrastructure elements) capture the observations and actions of service engineers and are a crucial resource for assessing system-level asset health. To facilitate tracking of historical asset health issues, these records are categorized using tags (labels) from a predefined taxonomy” (Iyer, page 1, left column, paragraph 1).
“a supervised learning approach can be implemented to automate the tagging (label[ing]) process, using Deep Learning based language models like BERT (machine learning model)” (Iyer, page 2, right column, paragraph 1)
“Figure 1 shows the overall workflow based on our approach. For the baseline model, historical semi-structured records, along with tags that were assigned to them, are provided as training data to a BERT-based classification model for supervised learning of tags (labels)” (Iyer, page 3, left column, paragraph 1)
“Aircraft maintenance and service records are critical for maintaining airworthiness of an aircraft; they carry details related to the repair performed on the aircraft” (Iyer, page 3, left column, paragraph 2). These records contain information about hardware, falling within the examples given of “computer infrastructure element[s]” in page 3, paragraph 4 of the instant specification.
…wherein the training data is based at least in part on information corresponding to one or more user interactions with at least a portion of the plurality of computer infrastructure elements and configuration information associated with at least a portion of the plurality of computer infrastructure elements:
“These records capture the observations and actions of service engineers (user[s]) and are a crucial resource for assessing system-level health of an asset and for inferring reliability issues arising from those. Typically, in these records, free text is used to describe observed issues and relevant corrective actions that were performed in response - Figure 2 shows snapshots of 2 such examples of service records” (Iyer, page 3, left column, paragraph 2)

PNG
media_image1.png
200
400
media_image1.png
Greyscale
“Two examples of maintenance/service records with red boxes indicating the ATA Chapters (tag) assigned to each” (Iyer, page 4, left column, Figure 2). These records show user interactions with infrastructure elements (investigation, cleaning, replacement) and configuration information associated with infrastructure elements (leaking gasket, gear not extending, broken shaft, part time since overhaul).
to generate, using the at least one machine learning model, at least one additional label for at least one additional computer infrastructure element using information corresponding to one or more user interactions with the at least one additional computer infrastructure element and configuration information associated with the at least one additional computer infrastructure element: “Figure 1 shows the overall workflow based on our approach. For the baseline model, historical semi-structured records (records corresponding to at least one additional computer infrastructure element), along with tags that were assigned to them, are provided as training data to a BERT-based classification model for supervised learning of tags (at least one additional label)” (Iyer, page 3, left column, paragraph 1). As discussed for the previous limitation, user interaction information and configuration information can be found in the records.
to perform one or more automated actions related to the at least one additional computer infrastructure element based at least in part on the at least one additional label, wherein the at least one additional computer infrastructure element comprises at least one software component, and wherein at least one of the one or more automated actions comprises a maintenance operation for the at least one software component: “the less reliable tag (additional label) assignments from the classifier are routed to the human expert for tag classification … Since the fraction of tag assignments that get routed via full automation are a good measure of efficiency of the mixed initiative system, we track this number as a metric and we call it coverage.” (Iyer, page 6, left column, paragraph 1)
While Iyer fails to disclose the further limitations of the claim, Muppidi discloses [a] non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code when executed by at least one processing device causes the at least one processing device: “Furthermore, the invention can take the form of a computer program product accessible from a computer-usable or computer-readable medium (memory) providing program code for use by or in connection with a computer or any instruction execution system” (Muppidi, column 12, paragraph 4); “Examples of a computer-readable medium include a semiconductor or solid state memory, magnetic tape, a removable computer diskette (non-transitory), a random access memory (RAM), a readonly memory (ROM), a rigid magnetic disk, and an optical disk. Current examples of optical disks include compact diskread only memory (CD-ROM), compact disk-read/write (CD-R/W) and DVD” (Muppidi, column 12, paragraph 6); “A data processing system suitable for storing and/or executing program code will include at least one processor coupled directly or indirectly to memory elements through a system bus. The memory elements can include local memory 50 employed during actual execution of the program code, bulk storage, and cache memories, which provide temporary storage of at least some program code in order to reduce the number of times code must be retrieved from bulk storage during execution” (Muppidi, column 12, paragraph 8)
Muppidi’s storage medium contains instructions to perform one or more automated actions related to the at least one additional computer infrastructure element based at least in part on the at least one additional label, wherein the at least one additional computer infrastructure element comprises at least one software component, and wherein at least one of the one or more automated actions comprises a maintenance operation for the at least one software component:
“The illustrative embodiments provide a method, system, and computer usable program product for classification and policy management, for software components (at least one software infrastructure element). A metadata associated with a component is identified. A mapping determination is made whether the metadata maps to a classification in a set of classifications. If the mapping determination is true, the component is assigned to the classification (at least one additional label) and a policy associated with the classification is associated with the component.”(Muppidi, column 2, paragraph 3)
“In addition, the policy that is applicable to the classification is identified or defined. The identified or defined policy is associated the classification (additional label). Furthermore, because of the policy being associated with the classification, associating (one or more automated actions) the policy with the component may occur automatically from assigning the component to the classification” (Muppidi, column 2, paragraph 4)
“The illustrative embodiments provide a method, system, and computer usable program product for classification and policy management, for software components” (Muppidi, column 2, paragraph 3)
“Process 900 begins by receiving a policy update (step 902) (maintenance operation). In one embodiment, a policy update may include changes to policies or policy templates already associated with classifications. In another embodiment, a policy update may include newly added policies or policy templates” (Muppidi, column 11, paragraph 4). As made clear by claim 9, a software update operation is a maintenance operation of software.
Iyer and Muppidi relate to automatic classification of infrastructure components and are analogous to the claimed invention. Iyer teaches a method of automatically tagging infrastructure records. The claimed invention improves upon this method by storing it in the form of instructions on computer hardware. Iyer teaches hardware able to run methods of automatically tagging infrastructure information, applicable to Iyer. A person of ordinary skill in the art would have recognized that storing Iyer’s method as computer instructions on Muppidi’s hardware would lead to the predictable result of the method being executable by a computing system, and would improve the known device by allowing it to be performed with real data (MPEP 2143 I. (D) Applying a known technique to a known device (method, or product) ready for improvement to yield predictable results).
Additionally, the claimed invention improves upon Iyer’s method by performing software updates based on tags. Muppidi teaches a method of updating software policy based on tags. A person of ordinary skill in the art would have recognized that performing software policy updates based on element tags would lead to the predictable result of automated selective maintenance on software infrastructure, and would improve the efficiency of the known device by automating different updates across different groups of software infrastructure with different requirements (MPEP 2143 I. (D) Applying a known technique to a known device (method, or product) ready for improvement to yield predictable results).

The analysis of claims 15-17 mirrors that of claims 2-4, with the exception that claims 15-17 are directed to additional generic computer hardware which executes the methods of claims 2-4. This generic hardware is taught by Muppidi, as discussed regarding claim 14. Thus, claims 15-17 are rejected under the same rationales used for claims 2-4, respectively.

Regarding claim 18, Iyer discloses an apparatus, able to:
to obtain at least one machine learning model, wherein the at least one machine learning model is trained, using a set of training data, to generate one or more labels for one or more of a plurality of computer infrastructure elements:
“Free-form text-based maintenance and service records related to industrial assets (infrastructure elements) capture the observations and actions of service engineers and are a crucial resource for assessing system-level asset health. To facilitate tracking of historical asset health issues, these records are categorized using tags (labels) from a predefined taxonomy” (Iyer, page 1, left column, paragraph 1).
“a supervised learning approach can be implemented to automate the tagging (label[ing]) process, using Deep Learning based language models like BERT (machine learning model)” (Iyer, page 2, right column, paragraph 1)
“Figure 1 shows the overall workflow based on our approach. For the baseline model, historical semi-structured records, along with tags that were assigned to them, are provided as training data to a BERT-based classification model for supervised learning of tags (labels)” (Iyer, page 3, left column, paragraph 1)
“Aircraft maintenance and service records are critical for maintaining airworthiness of an aircraft; they carry details related to the repair performed on the aircraft” (Iyer, page 3, left column, paragraph 2). These records contain information about hardware, falling within the examples given of “computer infrastructure element[s]” in page 3, paragraph 4 of the instant specification.
…wherein the training data is based at least in part on information corresponding to one or more user interactions with at least a portion of the plurality of computer infrastructure elements and configuration information associated with at least a portion of the plurality of computer infrastructure elements:
“These records capture the observations and actions of service engineers (user[s]) and are a crucial resource for assessing system-level health of an asset and for inferring reliability issues arising from those. Typically, in these records, free text is used to describe observed issues and relevant corrective actions that were performed in response - Figure 2 shows snapshots of 2 such examples of service records” (Iyer, page 3, left column, paragraph 2)

PNG
media_image1.png
200
400
media_image1.png
Greyscale
“Two examples of maintenance/service records with red boxes indicating the ATA Chapters (tag) assigned to each” (Iyer, page 4, left column, Figure 2). These records show user interactions with infrastructure elements (investigation, cleaning, replacement) and configuration information associated with infrastructure elements (leaking gasket, gear not extending, broken shaft, part time since overhaul).
to generate, using the at least one machine learning model, at least one additional label for at least one additional computer infrastructure element using information corresponding to one or more user interactions with the at least one additional computer infrastructure element and configuration information associated with the at least one additional computer infrastructure element: “Figure 1 shows the overall workflow based on our approach. For the baseline model, historical semi-structured records (records corresponding to at least one additional computer infrastructure element), along with tags that were assigned to them, are provided as training data to a BERT-based classification model for supervised learning of tags (at least one additional label)” (Iyer, page 3, left column, paragraph 1). As discussed for the previous limitation, user interaction information and configuration information can be found in the records.
to perform one or more automated actions related to the at least one additional computer infrastructure element based at least in part on the at least one additional label, wherein the at least one additional computer infrastructure element comprises at least one software component, and wherein at least one of the one or more automated actions comprises a maintenance operation for the at least one software component: “the less reliable tag (additional label) assignments from the classifier are routed to the human expert for tag classification … Since the fraction of tag assignments that get routed via full automation are a good measure of efficiency of the mixed initiative system, we track this number as a metric and we call it coverage.” (Iyer, page 6, left column, paragraph 1)
While Iyer fails to disclose the further limitations of the claim, Muppidi discloses [a]n apparatus comprising: at least one processing device comprising a processor coupled to a memory; the at least one processing device being configured: “Furthermore, the invention can take the form of a computer program product accessible from a computer-usable or computer-readable medium (memory) providing program code for use by or in connection with a computer or any instruction execution system” (Muppidi, column 12, paragraph 4); “A data processing system suitable for storing and/or executing program code will include at least one processor coupled directly or indirectly to memory elements through a system bus. The memory elements can include local memory 50 employed during actual execution of the program code, bulk storage, and cache memories, which provide temporary storage of at least some program code in order to reduce the number of times code must be retrieved from bulk storage during execution” (Muppidi, column 12, paragraph 8)
Muppidi’s apparatus is able to performing one or more automated actions related to the at least one additional computer infrastructure element based at least in part on the at least one additional label, wherein the at least one additional computer infrastructure element comprises at least one software infrastructure element, and wherein at least one of the one or more automated actions comprises at least one maintenance operation for the at least one software infrastructure element:
“The illustrative embodiments provide a method, system, and computer usable program product for classification and policy management, for software components (at least one software infrastructure element). A metadata associated with a component is identified. A mapping determination is made whether the metadata maps to a classification in a set of classifications. If the mapping determination is true, the component is assigned to the classification (at least one additional label) and a policy associated with the classification is associated with the component.”(Muppidi, column 2, paragraph 3)
“In addition, the policy that is applicable to the classification is identified or defined. The identified or defined policy is associated the classification (additional label). Furthermore, because of the policy being associated with the classification, associating (one or more automated actions) the policy with the component may occur automatically from assigning the component to the classification” (Muppidi, column 2, paragraph 4)
“The illustrative embodiments provide a method, system, and computer usable program product for classification and policy management, for software components” (Muppidi, column 2, paragraph 3)
“Process 900 begins by receiving a policy update (step 902) (maintenance operation). In one embodiment, a policy update may include changes to policies or policy templates already associated with classifications. In another embodiment, a policy update may include newly added policies or policy templates” (Muppidi, column 11, paragraph 4). As made clear by claim 9, a software update operation is a maintenance operation of software.
Iyer and Muppidi relate to automatic classification of infrastructure components and are analogous to the claimed invention. Iyer teaches a method of automatically tagging infrastructure records. The claimed invention improves upon this method by storing it in the form of instructions on computer hardware. Iyer teaches hardware able to run methods of automatically tagging infrastructure information, applicable to Iyer. A person of ordinary skill in the art would have recognized that storing Iyer’s method as computer instructions on Muppidi’s hardware would lead to the predictable result of the method being executable by a computing system, and would improve the known device by allowing it to be performed with real data (MPEP 2143 I. (D) Applying a known technique to a known device (method, or product) ready for improvement to yield predictable results).
Additionally, the claimed invention improves upon Iyer’s method by performing software updates based on tags. Muppidi teaches a method of updating software policy based on tags. A person of ordinary skill in the art would have recognized that performing software policy updates based on element tags would lead to the predictable result of automated selective maintenance on software infrastructure, and would improve the efficiency of the known device by automating different updates across different groups of software infrastructure with different requirements (MPEP 2143 I. (D) Applying a known technique to a known device (method, or product) ready for improvement to yield predictable results).

The analysis of claims 19-20 mirrors that of claims 2-3, with the exception that claims 19-20 are directed to additional generic computer hardware which executes the methods of claims 2-3. This generic hardware is taught by Muppidi, as discussed regarding claim 18. Thus, claims 19-20 are rejected under the same rationales used for claims 2-3, respectively.

Claim 7 is rejected under 35 U.S.C. 103 as being unpatentable over Iyer et al. (Mixed Initiative Approach for Reliable Tagging of Maintenance Records with Machine Learning, published 2022, ANNUAL CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2022), hereafter referred to as Iyer, in view of Muppidi et al. (CLASSIFICATION AND POLICY MANAGEMENT FOR SOFTWARE COMPONENTS, published 2/7/2012, US 8,112,370 B2), hereafter referred to as Muppidi, and further in view of Wark (Method And System For Identifying Data And Users Of Interest From Patterns Of User Interaction With Existing Data, published 1/9/2014, US 20140012870 A1).

Regarding claim 7, the rejection of claim 5 in view of Iyer and Muppidi is incorporated. Wark, in combination with Iyer, discloses a method wherein the outputting is performed in response to detecting, within a particular time period, at least one of: a threshold number of interactions with the additional computer infrastructure element by the user; a threshold number of times the user interacted with the additional computer infrastructure element; and a threshold number of actions performed by the user related to the additional computer infrastructure element: “expert analysts or users may be identifiable, through using the recommendation system to identify users' who have had a certain amount of interaction (threshold number of interactions / actions) with certain types of data elements. One example where this may be useful is in the financial industry, which encompasses thousands of analysts, and where it may be difficult to locate which analyst has the desired familiarity (interaction) with certain data sets. Likewise, similar scenarios arise in the military, research labs, and other industries having numerous analysts. In scientific research, if a particular user often performs a similar synthesis or analysis on certain types of data elements, such as looking for particular markers among mouse data, that information may be used to identify that user as one familiar with that type of marker.” (Wark, [0040]). Output in Iyer’s method is performed in response to the assignment of experts (decisions are output to experts for label review). With Wark’s method, those experts can be selected based on a threshold of (inter)actions.
Wark relates to identifying users relevant to a particular interaction type and is analogous to the claimed invention. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Iyer and Muppidi to automatically determine expert users based on a threshold number of interactions with relevant data types, as disclosed by Wark. Wark’s method enables quick and automatic determination of experts, even in fields where there may be thousands of similar users that would be hard to categorize manually. See Wark, [0040].

Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Iyer et al. (Mixed Initiative Approach for Reliable Tagging of Maintenance Records with Machine Learning, published 2022, ANNUAL CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2022), hereafter referred to as Iyer, in view of Muppidi et al. (CLASSIFICATION AND POLICY MANAGEMENT FOR SOFTWARE COMPONENTS, published 2/7/2012, US 8,112,370 B2), hereafter referred to as Muppidi, and further in view of Saetia et al. (Data-driven Approach to Equipment Taxonomy Classification, published 2019, ANNUAL CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2019), hereafter referred to as Saetia.

Regarding claim 8, the rejection of claim 1 in view of Iyer and Muppidi is incorporated. Saetia, in combination with Iyer, discloses a method, wherein the at least one machine learning model is retrained in response to at least one of a change to at least one label that is currently assigned to a given one of the computer infrastructure elements and a new label being assigned to at least one of the plurality of computer infrastructure elements: “To measure performance of the compatibility check process (Figure 3), 600 unlabeled descriptions with the lowest compatibility scores were reviewed and labeled (new label[s]) by an SME. The classification model (machine learning model) was retrained with the additional 600 newly labeled samples. The number of equipment descriptions auto-labeled by the compatibility and prediction score criteria rose to 33%.” (Saetia, page 10, right column, paragraph 1). When applied to Iyer, this method can be used to label unlabeled records associated with computer infrastructure elements and retrain the model with them.
Saetia relates to using machine learning to automatically tag infrastructure and is analogous to the claimed invention. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Iyer and Muppidi to label unlabeled data and use it to retrain the model, as disclosed by Saetia. Doing so would enable the use of otherwise unusable unlabeled data in the training set, which may contain words and infrastructure not present in the labeled data. When Saetia incorporated this retraining, the model’s automatic labeling ability increased significantly. See Saetia, page 10, left column, paragraph 2 to page 10, right column, paragraph 1.

Claims 10-11 are rejected under 35 U.S.C. 103 as being unpatentable over Iyer et al. (Mixed Initiative Approach for Reliable Tagging of Maintenance Records with Machine Learning, published 2022, ANNUAL CONFERENCE OF THE PROGNOSTICS AND HEALTH MANAGEMENT SOCIETY 2022), hereafter referred to as Iyer, in view of Muppidi et al. (CLASSIFICATION AND POLICY MANAGEMENT FOR SOFTWARE COMPONENTS, published 2/7/2012, US 8,112,370 B2), hereafter referred to as Muppidi, and further in view of Stenström et al. (Natural language processing of maintenance records data, published 2015, International Journal of COMADEM - April 2015), hereafter referred to as Stenstrom.

Regarding claim 10, the rejection of claim 1 in view of Iyer and Muppidi is incorporated. While Iyer and Muppidi fail to disclose the further limitations of the claim, Stenstrom discloses a method,
wherein the plurality of computer infrastructure elements corresponds to at least one datacenter: “The data used in this study was provided by Trafikverket (Swedish Transport Admini-stration)” (Stenstrom, page 3, right column, paragraph 5). Both the location managed by Trafikverket storing the data and the location that received the data for this study can be considered datacenters.
and further comprises: a hardware infrastructure element deployed at the at least one datacenter: “The text entry fields of the 10 958 records is found to contain 69 382 words in total” (Stenstrom, page 4, right column, paragraph 4); “Another type is “freeze”, which occurred 144 times. Freeze is referring to computer freeze/hang” (Stenstrom, page 5, left column, paragraph 4). A computer freeze is a problem with the software infrastructure being run on the hardware infrastructure of the computer.
Stenstrom relates to using machine learning to analyze infrastructure and is analogous to the claimed invention. Iyer teaches a method of automatically tagging infrastructure records. Stenstrom teaches a method of retrieving software / hardware computer infrastructure records from a data center. It would have been obvious to one of ordinary skill in the art to combine these methods by retrieving software / hardware computer infrastructure from a data center for automatic tagging by Iyer’s method. This would achieve the predictable result of tagged software / hardware computer infrastructure records, with Iyer and Stenstrom’s methods performing the same together as they did separately. (MPEP 2143 I. (A) Combining prior art elements according to known methods to yield predictable results).

Regarding claim 11, the rejection of claim 1 in view of Iyer and Muppidi is incorporated. Stenstrom, in combination with Iyer, discloses a method of providing a dashboard related to the plurality of computer infrastructure elements, wherein the dashboard is configured to at least one of:
display computer infrastructure element information corresponding to one or more of the plurality of computer infrastructure elements based at least in part on one or more labels generated using the at least one machine learning model: “The data of preventive and corrective maintenance work is commonly called maintenance records (computer infrastructure element information), reports or work orders, and follows a set template and procedure for registration and closure, through a graphical user interface (GUI) (dashboard). Maintenance records contain a number of fields/boxes (labels), such as: record identification number; asset information regarding system, subsystem and components; maintenance activity; failure cause; and remedy. However, the content depends if it is corrective or preventive maintenance records. The records fields within a GUI comprise of drop-down lists, list boxes, check boxes and text entry fields” (Stenstrom, page 2, left column, paragraph 1). Stenstrom’s method in combination with Iyer can be used to display record information, including tags automatically applied with Iyer’s method.
initiate one or more tasks corresponding to one or more of the plurality of computer infrastructure elements based at least in part on one or more labels generated using the at least one machine learning model
Stenstrom relates to using machine learning to analyze infrastructure and is analogous to the claimed invention. Iyer teaches a method of automatically tagging infrastructure records. The claimed invention improves upon this method by displaying automatically tagged records on a dashboard. Stenstrom teaches a method of displaying record information on a dashboard GUI, applicable to the combination of Iyer and Muppidi. A person of ordinary skill in the art would have recognized that presenting records tagged with Iyer’s method on a GUI dashboard would lead to the predictable result of displaying automatically generated tags, and would improve the known device by making tags generated through Iyer’s method readable / accessible by users (MPEP 2143 I. (D) Applying a known technique to a known device (method, or product) ready for improvement to yield predictable results).

Response to Arguments
The following responses address arguments and remarks made in the instant remarks dated 01/28/2026.

Miscellaneous
The Examiner notes that an additional piece of pertinent prior art not relied upon has been added to the conclusion, namely Castel et al. (COMPUTER SYSTEM AND METHOD FOR MAINTENANCE MANAGEMENT INCLUDING COLLABORATION ACROSS CLIENTS, published 2014, US 2014/0164603 A1) discloses a method of automatically recommending and performing computer infrastructure maintenance based on automatically generated tags.

101 Rejections
On page 11 of the instant remarks, the Applicant argues that the amended claims do not recite mental processes:
“The Office Action characterizes the claimed "generating" and "performing" steps as mental
processes, arguing that the generating could correspond to merely imagining a label or that the
performing step is merely an instruction to perform a generic action. (Office Action, pages 17-18).
Applicant respectfully disagrees. For example, amended claim 1 recites performing at least one
maintenance operation for the at least one software infrastructure element. An automated
maintenance operation on a software infrastructure element ( e.g., executing an update, a patch, or
a reboot) is technical action that changes the state of a computer system, and the action is based
on labels generated by at least one machine learning model. Such features are not observations,
evaluations or opinions, and cannot practically be performed in the human mind.

In this regard, Applicant points to the August 4, 2025 USPTO Memorandum Reminders
on Evaluating Subject Matter Eligibility of Claims Under 35 USC. §JOI (hereinafter "USPTO
Memo"), which states that a claim does not recite a mental process when it contains limitations
that "cannot practically be performed in the human mind." (USPTO Memo, page 2). A human
cannot mentally interface with a computer system to obtain such digital interaction and
configuration data, nor can they mentally execute a software maintenance operation (e.g., a patch,
update, or reboot) on a software component. Accordingly, the claims do not fall within the mental
process grouping and are eligible under Step 2A, Prong One.”
In regards to the Applicant’s arguments above, the Examiner respectfully disagrees that claim 1, as amended, recites no mental processes. As stated in MPEP 2106.04(a)(2)(III), The courts do not distinguish between mental processes that are performed entirely in the human mind and mental processes that require a human to use a physical aid (e.g., pen and paper or a slide rule) to perform the claim limitation. See, e.g., Benson, 409 U.S. at 67, 65, 175 USPQ at 674-75, 674 … Nor do the courts distinguish between claims that recite mental processes performed by humans and claims that recite mental processes performed on a computer. As the Federal Circuit has explained, "[c]ourts have examined claims that required the use of a computer and still found that the underlying, patent-ineligible invention could be performed via pen and paper or in a person’s mind." Versata Dev. Group v. SAP Am., Inc., 793 F.3d 1306, 1335, 115 USPQ2d 1681, 1702 (Fed. Cir. 2015). See also Intellectual Ventures I LLC v. Symantec Corp., 838 F.3d 1307, 1318, 120 USPQ2d 1353, 1360 (Fed. Cir. 2016) (‘‘[W]ith the exception of generic computer-implemented steps, there is nothing in the claims themselves that foreclose them from being performed by a human, mentally or with pen and paper.’’); Mortgage Grader, Inc. v. First Choice Loan Servs. Inc., 811 F.3d 1314, 1324, 117 USPQ2d 1693, 1699 (Fed. Cir. 2016) (holding that computer- implemented method for "anonymous loan shopping" was an abstract idea because it could be "performed by humans without a computer").
Claim 1 recites limitations amounting to mental processes performed on generic data structures. Generic data structures amount to generic computer components and are insufficient to render a mentally performable task non-abstract. For example, claim 1 recites the limitation “generating, using the at least one machine learning model, at least one additional label for at least one additional computer infrastructure element using information corresponding to one or more user interactions with the at least one additional computer infrastructure element and configuration information associated with the at least one additional computer infrastructure element”, reciting a mental process of generating at least one additional label, performed by a machine learning model, a generic data structure insufficient to render the limitation non-abstract.
The Examiner asserts that claim 1, as amended, recites mental processes, and maintains its rejection on the basis of the Alice/Mayo tests performed (See 101 rejections).

On page 12 of the instant remarks, the Applicant argues that the automated actions performed by the claimed invention integrate any recited judicial exceptions into a practical application:
“Regarding Step 2A, Prong Two, the Office Action characterizes the "automated actions"
as a "mere instruction to perform a generic action related to some information." Applicant
respectfully disagrees. For example, the USPTO Memo specifically warns examiners "not to
oversimplify claim limitations" when applying the "apply it" consideration. Additionally, claim 1
is amended to further clarify that the automated action is not generic, but a specific maintenance
operation performed on a software infrastructure element, thereby integrating any alleged abstract
idea into a practical application of, e.g., automated self-maintenance of software infrastructure.

Even if the amended claims recite an abstract idea (which Applicant contests), the claims
are patent eligible under Step 2A, Prong One because they integrate any such abstract idea into a
practical application ( e.g., automated maintenance and improvement of computer infrastructure).

The Office Action asserts that the automated action is a "mere instruction to perform a
generic action." (Office Action, page 18). This analysis overlooks the specific technical
improvement recited in the amended claims. For example, the present specification describes a
technical problem where conventional infrastructure management relies on manual grouping,
which is time-consuming and leads to inconsistencies, thereby negatively impacting security,
usability, efficiency and/or availability. (see, e.g., page 1, lines 9-14). The specification also
describes a technical solution to this problem, which includes using machine learning to
characterize elements and perform automated actions ( e.g., initiating an update, performing a
restore or performing a reboot as recited in forth in dependent claim 9).

These features are reflected in amended claim 1, which recites performing a maintenance
operation on a software infrastructure element based at least in part on the at least one additional
label. This is a specific improvement to the technical operation of the infrastructure itself By
automatically executing maintenance operations, the claims can help mitigate the technical
problems associated with monitoring and maintaining computer infrastructure described in present
specification (see page 2, line 1-7).

The Appeals Review Panel (ARP) recently addressed the eligibility of machine learning
claims in Ex parte Desjardins, Appeal 2024-000567 (Sept. 26, 2025) (hereinafter "Desjardins").
In Desjardins, the ARP vacated a § 101 rejection, holding that claims directed to training a machine
learning model were eligible because they reflected the specification's disclosed improvements,
specifically that the system could "use less of their storage capacity" and have "reduced system
complexity". The ARP found that adjusting parameters to protect performance on previous tasks
constituted "an improvement to how the machine learning model itself operates," (see page 9).
The ARP also warned against evaluating claims at a "high level of generality" and equating
machine learning with unpatentable algorithms, noting that such an approach ignores the specific
technological context (see Desjardins, page 9).

Similar to how the claim in Desjardins improved the operation of the model, the amended
claims improve the operation of the infrastructure being managed. For example, the amended
claims do not merely generate a label, but the claims use the label to trigger a specific maintenance
operation on a software element. This ensures, for example, that the software remains secure,
updated, or functional without manual intervention. Under the logic of Desjardins, this is a specific
improvement to computer functionality and not an abstract idea.

Additionally, the maintenance operation recited in amended claim 1 is not a generic action
in the context of §101. The maintenance operation is a specific application of the labels generated
by the machine learning model to solve the problem of infrastructure availability and security. The
USPTO Memo reminds Examiners that the "apply it" consideration should not be used to
"oversimplify claim limitations," (see page 4). Characterizing a specific software maintenance
operation as generic ignores the specific technological context of the claim.”
In response to the Applicant’s assertion that the claimed invention is practically integrated and / or amounts to significantly more through improvement to technology, the Examiner respectfully disagrees. The Examiner notes that practical integration through improvement only comes through improvements to the functioning of a technology or technical field, as noted by MPEP 2106.04(d)(I): Limitations the courts have found indicative that an additional element (or combination of elements) may have integrated the exception into a practical application include: … An improvement in the functioning of a computer, or an improvement to other technology or technical field, as discussed in MPEP §§ 2106.04(d)(1) and 2106.05(a). Such technical improvements can similarly indicate a limitation amounts to significantly more than the claim’s recited judicial exceptions, as noted in MPEP 2106.05(I)(A): Limitations that the courts have found to qualify as "significantly more" when recited in a claim with a judicial exception include: i. Improvements to the functioning of a computer, e.g., a modification of conventional Internet hyperlink protocol to dynamically produce a dual-source hybrid webpage, as discussed in DDR Holdings, LLC v. Hotels.com, L.P., 773 F.3d 1245, 1258-59, 113 USPQ2d 1097, 1106-07 (Fed. Cir. 2014) (see MPEP § 2106.05(a)).
Improving a manual human-performable process by automating it does not constitute improvement to a technology or technical field, and is insufficient to show an improvement in computer-functionality, as noted in MPEP 2106.05(a)(I): Examples that the courts have indicated may not be sufficient to show an improvement in computer-functionality: … iii. Mere automation of manual processes, such as using a generic computer to process an application for financing a purchase, Credit Acceptance Corp. v. Westlake Services, 859 F.3d 1044, 1055, 123 USPQ2d 1100, 1108-09 (Fed. Cir. 2017) or speeding up a loan-application process by enabling borrowers to avoid physically going to or calling each lender and filling out a loan application, LendingTree, LLC v. Zillow, Inc., 656 Fed. App'x 991, 996-97 (Fed. Cir. 2016) (non-precedential).
The problem described by the Applicant, inconsistencies and high time-consumption of human infrastructure tagging, is that of a human-performable process and does not constitute a problem in a technology or technical field. Thus, improvements to this problem represented by the claimed invention are insufficient to practically integrate recited judicial exceptions or amount to significantly more than them.
Additionally, regarding the assertion that the automated actions and associated maintenance operation(s) recited in the first claim amount to more than mere instruction to apply, the Examiner respectfully disagrees. Limitations that recite the effects of applying a judicial exception at a high level of generality, to the point where the method to achieve those effects is either unclear or very broad, amount to mere instruction to apply the judicial exception, as noted in MPEP 2106.05(f)(3): a claim that generically recites an effect of the judicial exception or claims every mode of accomplishing that effect, amounts to a claim that is merely adding the words "apply it" to the judicial exception. See Internet Patents Corporation v. Active Network, Inc., 790 F.3d 1343, 1348, 115 USPQ2d 1414, 1418 (Fed. Cir. 2015) (The recitation of maintaining the state of data in an online form without restriction on how the state is maintained and with no description of the mechanism for maintaining the state describes "the effect or result dissociated from any method by which maintaining the state is accomplished" and does not provide a meaningful limitation because it merely states that the abstract idea should be applied to achieve a desired result).
Regarding “performing one or more automated actions related to the at least one additional computer infrastructure element based at least in part on the at least one additional label, wherein the at least one additional computer infrastructure element comprises at least one software infrastructure element, and wherein at least one of the one or more automated actions comprises at least one maintenance operation for the at least one software infrastructure element”, as recited in amended claim 1, the relationship between the additional label (generated through a recited mental process) and the automated actions is completely unspecified. Stating that the computer infrastructure element comprises a “software infrastructure element” or that the automated actions comprise at least one maintenance operation for said software infrastructure element, does not clarify this relationship, or the method by which the automated actions are performed in response to the label. Thus, this limitation merely claims intended effects (performing one or more automated actions) of applying a judicial exception (the at least one additional label), and amounts to mere instruction to apply.
No rejections are withdrawn on these grounds. See the 101 rejections section for more detail.

On pages 13 of the instant remarks, the Applicant argues that dependent claims are eligible due to their dependence on eligible independent claims:
“The dependent claims are also eligible under § 101 at least due to their respective
dependence from independent claims 1, 14 and 18.”
Regarding the Applicant’s arguments above, the Examiner respectfully disagrees. The arguments against the patentability of amended claim 1 under 35 U.S.C. 101 discussed above in previous responses are applicable to the substantially similar independent claims 14 and 18.
Thus, no independent claims are found to be patentable over 101, and no rejections are withdrawn on these grounds.

103 Rejections
On pages 10-11 of the instant remarks, the Applicant argues that the cited references don’t disclose all limitations of the amended independent claims:
“Applicant respectfully submits that the cited references do not teach or suggest all of the features
of the independent claims.

In rejecting former claim 1, the Office Action asserts that Iyer discloses:
performing one or more automated actions related to the at least one
additional computer infrastructure element based at least in part on the at least one
additional label: "the less reliable tag (additional label) assignments from the
classifier are routed to the human expert for tag classification ... Since the fraction
of tag assignments that get routed via full automation are a good measure of
efficiency of the mixed initiative system, we track this number as a metric and we
call it coverage." (Iyer, page 6, left column, paragraph 1 ).

Iyer merely describes a system for assigning tags to text-based maintenance and service
records to facilitate historical tracking of aircraft asset health (see, e.g., page 3, section 2.1). The
automation cited by the Office Action is merely the assignment of the tag itself to the record (or
routing the record to a human for manual tagging). Iyer' s process concludes once the tag is saved
to a database for future retrieval (see, e.g., Figure 1 ). Iyer' s process concludes once the tag is saved
to a database for future retrieval. Iyer does not describe or suggest performing any subsequent
action on the asset (e.g., aircraft part) itself based on that tag. The "action" in Iyer is merely
administrative ( e.g., to track historical asset and health issues).

Amended claim 1 recites at least one additional computer infrastructure element comprises
at least one software infrastructure element, and that at least one of the one or more automated
actions comprises a maintenance operation for the at least one software infrastructure element. Iyer
does not teach or suggest such features because Iyer operates on the historical maintenance and
service text records describing physical aircraft parts (e.g., landing gear, engine gaskets). Iyer does
not perform any action on a functional software infrastructure element (e.g., a virtual machine,
application, or software container) as recited by amended claim 1. Additionally, Iyer does not teach
or suggest performing a maintenance operation ( e.g., updating, restoring, or rebooting, as further
detailed in dependent claim 9) for a software infrastructure element.

Because Iyer is limited to tagging static text records and fails to teach active maintenance
of software infrastructure, it cannot render the amended claims obvious. Therefore, Applicant
respectfully requests withdrawal of the § 103 rejection.

Accordingly, Applicant submits that the cited references do not teach or suggest at least
the aforementioned features of the amended independent claims, and for at least this reason, the
§ 103 rejection should be withdrawn.”
Regarding the Applicant’s arguments surrounding claim 1’s amendments above, the Examiner agrees that Iyer does not disclose performing maintenance operations on software infrastructure. However, this deficiency is remedied by Muppidi. Muppidi’s system automatically labels software infrastructure elements and assigns policies to labeled elements (Muppidi, column 2, paragraphs 3-4).
Regarding amended claim 9, while Iyer doesn’t disclose updating a software element as a maintenance operation, this deficiency is remedied by Muppidi, which discloses selectively updating groups of software infrastructure policies based on assigned labels (Muppidi, column 11, paragraph 4).
Thus, both claim 1 and claim 9, as amended, are found to be obvious over Iyer in view of Muppidi. See the 103 rejections section for more detail. No rejections are withdrawn on these grounds.

On page 11 of the instant remarks, the Applicant argues that dependent claims should be allowed based on their dependence on allowable independent claims:
“Though the dependent claims contain their own allowable subject matter, these claims
should at least be allowable due to their respective dependence on the allowable independent
claims. However, to expedite prosecution at this time, no further comment will be made.”
As noted above, independent claim 1 is not found to be allowable over the prior art. Substantially similar independent claims 14 and 18 are found to be obvious over the prior art under similar rationales. Thus, no dependent claims are allowed purely through dependence on the independents.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Castel et al. (COMPUTER SYSTEM AND METHOD FOR MAINTENANCE MANAGEMENT INCLUDING COLLABORATION ACROSS CLIENTS, published 2014, US 2014/0164603 A1) discloses a method of automatically recommending and performing computer infrastructure maintenance based on automatically generated tags.
Lowenmark et al. (Processing of Condition Monitoring Annotations with BERT and Technical Language Substitution: A Case Study, published 2022, Proceedings of the 7th European Conference of the Prognostics and Health Management Society 2022 pp. 306-314) discloses retraining a machine learning fault report analysis system.
Ozturk et al. (Analysis and relevance of service reports to extend predictive maintenance of large-scale plants, published 2022, 55th CIRP Conference on Manufacturing Systems pp. 1551 – 1558) discloses a method of analyzing software and hardware infrastructure maintenance data.

Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.

Any inquiry concerning this communication or earlier communications from the examiner should be directed to Aaron P Gormley whose telephone number is [phone]. The examiner can normally be reached Monday - Friday 12:00 PM - 8:00 PM EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Michelle T Bechtold can be reached at [phone]. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.

/AG/Examiner, Art Unit 2148
/MICHELLE T BECHTOLD/Supervisory Patent Examiner, Art Unit 2148

Prosecution Timeline

• Jan 20, 2023: Application Filed
• Oct 17, 2025: Non-Final Rejection — §101, §103
• Jan 12, 2026: Interview Requested
• Jan 27, 2026: Examiner Interview Summary
• Jan 27, 2026: Applicant Interview (Telephonic)
• Jan 28, 2026: Response Filed
• Mar 05, 2026: Final Rejection — §101, §103 (current)