Status of the Application
This Office Action is in response to Application Serial 17/795, 571. In response to the Examiner’s action November 10, 2025, Applicant submitted amendments and arguments dated December 31, 2025. Applicant amended claims 1, 8, and 14. Claims 1-20 are pending.
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 . In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
Information Disclosure Statement
Applicant did not submit an information disclosure statement for consideration by the examiner.
Response to Amendments
Claims 1-20 is/are pending in this application. The claim(s) 1,8, and 14 are amended.
Regarding the 35 U.S.C. 101 rejection. The claims 1-20 is/are rejected under 35 U.S.C. 101, see below.
Regarding the pending 35 U.S.C. 103 rejection. The Applicant’s amendments to claim 1-20 are not persuasive. Please see the prior art rejection below.
Response to Arguments
Applicant’s arguments filed on December 31, 2025, have been fully considered but they are not persuasive and/or are moot in view of the revised rejections. Applicant’s arguments will be addressed herein below.
Claim Rejection Under 35 U.S.C. 101
On pages 13-15 of the Applicant’s 35 U.S.C. 101 arguments, Applicant traverses:
Applicant respectfully asserts that the amended claim 1 includes additional elements that integrate the alleged abstract idea into a practical application. For example, representative claim 1 as a whole integrates into a practical application
because it is directed to improvements in the technical field of process mining and automation of techniques for improving operational processes based on computer-generated event data. Paragraph 16 of the as-filed specification recites:
[a]n aspect of the inventive arrangements disclosed is the leveraging of process mining combined with artificial intelligence (AI) to quantify the potential for improving a process and unlocking the potential by identifying transformations that improve the operations of the process. . . . A machine-generated process value debt quantifies the potential for improving the process, and a machine-generated process transformation propensity generates specific transformations recommendations to capture the improvement potential.
Independent claims 8 and 14, as amended, include features similar to independent claims 1. Thus, independent claims 1, 8, and 14 and the claims that depend thereon, integrate the alleged abstract idea into a practical application and are patent-eligible under 35 U.S.C. 101. According, Applicant respectfully requests that the Examiner reconsiders and withdraw the rejection of claims 1-20 under 35 U.S.C. 101.
Examiner respectfully disagrees with the Applicant’s 35 U.S.C. 101 arguments. The Applicant’s claims are analyzed in light of MPEP 2103-2106. Claims 1-20 in view of the claims limitations, are related to the abstract idea of recommending a process improvement based on tracked activities which is a mental concept. Furthermore, the claims recite “determining, for each of the KPIs, a KPI gap and a KPI impact score, wherein each KPI gap is based on a difference between an observed KPI value and a baselined KPI value, and wherein each KPI impact score is based on a plurality of scaled impact values corresponding to a plurality of predetermined performance metrics for the process” and “inputting one or more discrepancies between the process and the process model to a subtract/optimize/divide/add (SODA) model trained “ which are mathematical concepts. The claims recite mental concepts and mathematical concepts and therefore recite abstract concepts. The claims are directed to a judicial exception under Step 2A prong one.
This judicial exception are not integrated into a practical application under the second prong of Step 2A. In particular, the claims recite the additional elements beyond the recited abstract idea of, “A system, comprising: a processor configured to initiate operations, the operations including:”, “by a first machine learning model”, “by process mining event data,” “to computer-tracked activities associated with a process”, “through supervised learning” in claim 8 (and similarly in claims 1 and 14). However, when viewed as an ordered combination, and pursuant to the broadest reasonable interpretation, each of the additional elements are computing elements recite adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer or merely uses a computer as a tool to perform an abstract idea – see MPEP 2106.05 (f).
Regarding an improvement, the claims broadly recite a computer tracked process and transforming the process by adding and subtracting activities. The Applicant is encouraged to clarify the process transformation that is rooted in technology. Here, the process transformation and KPI performance scaling is evaluated as a business process that can be drawn as a flow chart using pen and paper. The process transformation can subtract steps or add steps using pen and paper. Applicant is encouraged to clarify an embodiment that is disclosed in the instant specification. Therefore, the claims also fails to recite any improvements to technology or technical field, improvements to the functioning of the computer itself.
Regarding the Applicant’s argument that [a]n aspect of the inventive arrangements disclosed is the leveraging of process mining combined with artificial intelligence (AI) to quantify the potential for improving a process and unlocking the potential by identifying transformations that improve the operations of the process. . . . A machine-generated process value debt quantifies the potential for improving the process, and a machine-generated process transformation propensity generates specific transformations recommendations to capture the improvement potential. Examiner respectfully disagrees.
Examiner points Applicant to Subject Matter Eligibility Guidance Example 47- Anomaly Detection. Example 47 provides an example of integration into a practical application. As disclosed in the prior pertinent art, supervised learning is a type of machine learning. Particularly, supervised learning scales data. Scaling data is a mathematical operation similar to the Applicant’s nomenclature SODA. A model that subtract/ optimize/ divide/ add (SODA) is simply completing a mathematical concept.
Additionally, Applicant is encouraged to clarify ”, “by process mining event data, ”, and “to computer-tracked activities associated with a process”.
The Applicant’s arguments are not persuasive. The claims are rejected under 35 U.S.C 101.
Rejection under 35 U.S.C. 103 based on WETZSTEIN and AL-ANGOUDI
On pages 15-18, of the Applicant’s 35 U.S.C. 103 rejection, Applicant respectfully submits that WETZSTEIN and AL-ANGOUDI do not disclose each and every feature recited in amended claim 1.
Applicant traverses the cited art does not disclose the PVD exceeding a predetermined threshold and the amendments to claim 1.
Applicant traverses Ganeshmani is directed to a machine learning model that identifies a process anomaly.
Although potentially of different scope than claim 1, independent claims 8 and 14, as amended, recite similar features. Therefore, independent claims 1, 8, and 14, and the claims that depend thereon, are patentable over the cited sections of the applied references, whether taken alone or in any reasonable combination. Accordingly, Applicant respectfully requests that the Examiner reconsider and withdraw the rejection of claims 1-20 under 35 U.S.C. § 103 based on WETZSTEIN and AL-ANGOUDI.
Examiner respectfully disagrees with Applicant’s 35 U.S.C. 103 arguments. The Applicant’s amendments to claims necessitate grounds for a new rejection. See prior art rejection below.
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 1-7 are process.
Claims 8-13 is machine.
Claims 14-20 is manufacture.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The claim 8 (and similarly claim 1 and claim 14) recite, “… mapping, … , key performance indicators (KPIs) to a discovered process model, wherein the discovered process model is created … retrieved from one or more event logs generated in response … ;
determining, for each of the KPIs, a KPI gap and a KPI impact score, wherein each KPI gap is based on a difference between an observed KPI value and a baselined KPI value, and wherein each KPI impact score is based on a plurality of scaled impact values corresponding to a plurality of predetermined performance metrics for the process; generating, for each of the KPIs, a KPI-level enhancement potential based on the KPI gap and KPI impact score of each KPI; generating a process value debt (PVD) for the process based on an average of the KPI-level enhancement potential of each of the KPIs; responsive to the PVD exceeding a predetermined threshold, inputting one or more discrepancies between the process and the process model to a subtract/optimize/divide/add (SODA) model trained … using process conformance data associated with the event data and a reference model; classifying, by the SODA model, the one or more discrepancies into one or more process transformations likely to reduce the PVD, each process transformation comprising one or more of a subtract recommendation associated with removing a first step from the process, an optimize recommendation associated with modifying a second step of the process, a divide recommendation associated with splitting a third step of the process into multiple steps, or an add recommendation associated with adding a fourth step to the process; determining a process transformation propensity (PTP) score for each of the one or more process transformations, wherein the PTP score for each of the one or more process transformations is correlated with a decrease in the PVD; and outputting a process transformation recommendations recommending at least one of the one or more process transformations selected based on the PTP score of each of the one or more process transformations.” Claims 1-20 in view of the claims limitations, are related to the abstract idea of recommending a process improvement based on tracked activities which is a mental concept. Furthermore, the claims recite “determining, for each of the KPIs, a KPI gap and a KPI impact score, wherein each KPI gap is based on a difference between an observed KPI value and a baselined KPI value, and wherein each KPI impact score is based on a plurality of scaled impact values corresponding to a plurality of predetermined performance metrics for the process” and “inputting one or more discrepancies between the process and the process model to a subtract/optimize/divide/add (SODA) model trained “ which are mathematical concepts. The claims recite mental concepts and mathematical concepts and therefore recite abstract concepts. The claims are directed to a judicial exception under Step 2A prong one.
This judicial exception are not integrated into a practical application under the second prong of Step 2A. In particular, the claims recite the additional elements beyond the recited abstract idea of, “A system, comprising: a processor configured to initiate operations, the operations including:”, “by a first machine learning model”, “by process mining event data, ”, “to computer-tracked activities associated with a process”, “through supervised learning” in claim 8 (and similarly in claims 1 and 14). However, when viewed as an ordered combination, and pursuant to the broadest reasonable interpretation, each of the additional elements are computing elements recite adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer or merely uses a computer as a tool to perform an abstract idea – see MPEP 2106.05 (f).
Regarding an improvement, the claims broadly recite a computer tracked process and transforming the process by adding and subtracting activities. The Applicant is encouraged to clarify the process transformation that is rooted in technology. Here, the process transformation and KPI performance scaling is evaluated as a business process that can be drawn as a flow chart using pen and paper. The process transformation can subtract steps or add steps using pen and paper. Applicant is encouraged to clarify an embodiment that is disclosed in the instant specification. Therefore, the claims also fails to recite any improvements to another technology or technical field, improvements to the functioning of the computer itself.
At step 2B, it is MPEP 2106.05 (d) – Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information).
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the additional elements when considered both individually and as an ordered combination do not amount to significantly more than the abstract idea. See MPEP 2106.05(f) - mere instructions to Apply an Exception.
The dependent claim 2-7, 9-13 and 12-20 do not recite additional elements beyond what is recited in the independent claims. Dependent claims 2-7 further narrow the abstract idea of independent claim 1. Dependent claims 9-13 further narrow the abstract idea of independent claim 8. Dependent claims 15-20 further narrow the abstract idea of independent claim 14. The claims 1-20 are not patent eligible.
Moreover, aside from the aforementioned additional elements, the remaining elements of dependent claims 2- 7, 9-13, 15-20 do not transform the recited abstract idea into a patent eligible invention because these claims merely recite further limitations that provide no more than simply narrowing the recited abstract idea.
Since there are no limitations in these claims that transform the exception into a patent eligible application such that these claims amount to significantly more than the exception itself, claims 1-20 are rejected under 35 U.S.C. 101 as being directed to non-statutory subject matter.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
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.
Claim(s) 1-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Zaslavsky (US 11,625,626 B2) and Al- Anqoudi (2021, Using Machine Learning in Business Process Re-Engineering).
Regarding Claim 1, (Currently Amended) [ and similarly claim 8 and 14]
A computer-implemented method, comprising: mapping, by a first machine learning model, key performance indicators (KPIs) to a process model, wherein the process model is discovered by process mining event data retrieved from one or more event logs generated in response to computer-tracked activities associated with a process;
Zaslavsky generates performance improvement recommendations for machine learning models., Zaslavsky [abstract].
Zaslavsky teaches organizations employ processes and/or systems that are dependent on machine learning models (e.g., in a supervised or an unsupervised learning system). Anomaly detection systems, for example, may employ machine learning models to detect anomalous activity within an organization. A given transaction of an organization, for example, may be classified as a suspicious transaction based on a risk score assigned by the fraud detection system. User behavior models may be used by the fraud detection system to determine, for example, how closely current user behavior associated with the given transaction aligns with the user behavior expected by the user behavior models. , Zaslavsky [column 1 lines 13- 31], [Figure 1], [column 4 lines 15-18].
Zaslavsky teaches wn in FIG. 4, an exemplary features usage category encompasses KPIs for multi-channel, threat intelligence, raw data reports and transaction monitoring. Finally, the exemplary table 400 of FIG. 4 comprises a data quality category that encompasses KPIs for nullity check and distributions., and thus, Zaslavsky teaches computer-tracked activities, and data., Zaslavsky [column 8 lines 28-33].
determining, for each of the KPIs, a KPI gap and a KPI impact score, wherein each KPI gap is based on a difference between an observed KPI value and a baselined KPI value, and wherein each KPI impact score is based on a plurality of scaled impact values corresponding to a plurality of predetermined performance metrics for the process;
Zaslavsky teaches with fraud detection systems, for example, one performance metric is the fraud detection rate. If the fraud detection performance is low or not optimal, there is currently no way for a customer to understand from this metric how to improve the performance of the product. Performance metrics are often referred to as KPIs. Zaslavsky [column 2 lines 65 - 68], [column 2 lines 1-4].
Zaslavsky teaches the exemplary performance improvement recommendation process 200 evaluates a predefined set of KPIs on each customer implementation separately, as discussed further below in conjunction with FIGS. 3-5. The considered KPIs may be defined, for example, by one or more subject matter experts (SME), such as those employed by the provider of the model modification recommender 180 and/or the provider of the evaluated machine learning models. (baseline) An overall score is calculated, in some embodiments, by summing the weighted category-based KPI scores, as discussed further below in conjunction with FIGS. 4 and 5. Zaslavsky [column 6 lines 33 - 56].
PNG
media_image1.png
110
316
media_image1.png
Greyscale
Zaslavsky teaches as shown in FIG. 3, the weight learning module 160 of FIG. 1 collects calculated KPIs and performance data for a plurality of customers 310-A through 310-O (for example, via the data collection module (or agent) 145. The weight learning module 160 then employs the supervised machine learning techniques to learn an appropriate weight for each KPI (and/or KPI category) based on the performance and experience of multiple customers 310-A through 310-O of the given vendor. As noted above, each weight is based, at least in part, on the expected improvement for a modification of at least one factor related to each KPI. Zaslavsky [column 4 lines 65 - 68], [column 2 lines 1-4].
PNG
media_image2.png
539
792
media_image2.png
Greyscale
(Weighted learning is scaling. The defined KPI by one of more SME recommenders is the baseline.)
generating, for each of the KPIs, a KPI-level enhancement potential based on the KPI gap and KPI impact score of each KPI;
Zaslavsky teaches category-based KPI scores, that are provided to customers, along with one or more recommendations to improve the performance of an anomaly detection model, or another machine learning model. Zaslavsky [column 3 lines 22-26] ,[Figure 4], [Figure 5].
Zaslvasky teaches multiple KPIs (or other performance metrics) are grouped by category, in some embodiments, and a different performance score is computed for each category. The category scores are determined by the KPI weight and KPI score for a given KPI, as defined by the category score equation in FIG. 2. Zaslvasky [column 8 lines 33-39], [Figure 2].
generating a process value debt (PVD) for the process based on an average of the KPI- level enhancement potential of each of the KPIs;
Zaslavasky teaches techniques for generating performance improvement recommendations for machine learning models (for example, in an online fraud detection market) will help customers to understand how to improve system performance by mapping defined steps (based on the predefined KPIs) that a customer can take to improve the performance. In addition, the suggested steps can be prioritized in some embodiments by a potential performance impact. Zaslvasky [column 10 lines 42-46], [Figure 2].
Zaslavasky generates performance improvement recommendations for machine learning models, and thus, Zaslavasky quantifies the specific transformation recommendations to capture, improvement potential. Therefore, Zaslavasky teaches Process value debt, as discloses in the instant specification [016].
responsive to the PVD exceeding a predetermined threshold, inputting one or more discrepancies between the process and the process model to a subtract/optimize/divide/add (SODA) model trained supervised learning process conformance data associated with the event data and a reference model;
Zaslavasky teaches KPI definition 170-1 and scoring and model modification recommender 180., Zaslavasky [Figure 1]. Zaslavasky teaches the considered KPIs may be defined, for example, by one or more subject matter experts (SME), such as those employed by the provider of the model modification recommender 180 and/or the provider of the evaluated machine learning models., Zaslavasky [column 6 lines 36-40].
Zaslavasky teaches low case management category, the reasoning indicates that the score is caused by a low case marking percentage (e.g., lower than 30%), and the suggested action is that customers with a higher case marking percentage (e.g., 30%-50%) benefited from an additional 17% improvement in fraud detection, and thus, Zaslavasky teaches thresholds., Zaslavasky [column 9 lines 14-30].
classifying, SODA, the one or more discrepancies into one or more process transformations likely to reduce the PVD, each process transformation comprising one or more of a subtract recommendation associated with removing a first step from the process, an optimize recommendation associated with modifying a second step of the process, a divide recommendation associated with splitting a third step of the process into multiple steps, or an add recommendation associated with adding a fourth step to the process;
See Zaslavasky [Figure 2 and the associated text].
Zaslavasky teaches for example, the predefined remedial steps and/or mitigation steps to address the detected predefined anomalies may comprise the transmission of an alert or alarm to the user device and/or user for important or suspicious events; isolating, removing, quarantining, limiting permissions, analyzing, and deactivating one or more of the user devices and/or one or more files, accounts or aspects of the user devices or the user. Zaslavasky [column 10 Lines 25-40], [Figure 2].
The weight learning module 160 then employs the supervised machine learning techniques to learn an appropriate weight for each KPI (and/or KPI category). Zaslavsky [column 4 lines 65 - 68], [column 2 lines 1-4].
Within claim 1, Zaslavasky discloses process transformation and scaling based on KPI, and thus, Zaslavasky discloses the one or more discrepancies into one or more process transformations likely to reduce the PVD, each process transformation comprising one or more of a subtract recommendation associated with removing a first step from the process. Claim 1 a "Markush" claim recites a list of alternatively useable members. In re Harnisch, 631 F.2d 716, 719-20, 206 USPQ 300, 303 (CCPA 1980); Ex parte Markush, 1925 Dec. Comm'r Pat. 126, 127 (1924). The listing of specified alternatives within a Markush claim is referred to as a Markush group or a Markush grouping. Abbott Labs v. Baxter Pharmaceutical Products, Inc., 334 F.3d 1274, 1280-81, 67 USPQ2d 1191, 1196 (Fed. Cir. 2003) (citing to several sources that describe Markush groups)- See MPEP 706.03.
… determining a process transformation propensity (PTP) score for each of the one or more process transformations, wherein the PTP score for each of the one or more process transformations is correlated with a decrease in the PVD;
See above. Zaslavasky [column 10 Lines 25-40], [Figure 2].
and outputting a process transformation recommendation recommending at least one of the one or more process transformations selected based on the PTP score of each of the one or more process transformations,
PNG
media_image3.png
663
890
media_image3.png
Greyscale
Al- Anqoudi further teaches:
… SODA through supervised learning using … classifying, by the SODA model, the one or more discrepancies ….
Al- Anqoudi teaches the proposed model looks at each activity
in the process separately. The machine-learning model takes the featured data, as inputs, and examines them against a supervised machine-learning approach. The model then labels the input and recommends an action against each activity. After examining all process activities, the outcome is a re-engineered version of the original process., Machine learning inputs include Activity performance against its KPIs: poor, good, and excellent. This attribute can be more accurate when the KPIs data are available. Al- Anqoudi [Figure 3], [p.13].
PNG
media_image4.png
407
1008
media_image4.png
Greyscale
Zaslavasky teaches generating performance improvement recommendations for machine learning models. Al-Anqoudi teaches business process re-engineering to manage progress towards achieving targets and objectives. It would have been obvious to combine before the effective filing date, evaluating performance metrics for multiple implementations of a machine learning model, as taught by Zaslavasky, with utilis[ing] machine learning for business process management in general and business process re-engineering, as taught by Al-Anqoudi because re-engineering a business process will sometimes transform every aspect of an organisation, including organisation structures, values, and reward systems [15], maximising the impact of the change on the organisation culture., Al-Anqoudi [p.2]
Although not relied on Examiner points the Applicant to prior art that describes supervised learning used to adjust changes in a process.
Wolf (2008, SODA: An Optimizing Scheduler for Large-Scale Stream-Based Distributed Computer System) where SODA wants scheduler must be able to shift resource allocation dynamically in response to changes to resource availability, job arrivals and departures, incoming data rates.
Zhai (US 2022/0189612 A1) teaching the system can modify the supervised learning technique after the pre-training to no longer use weight decay. Some supervised learning techniques, after each training step in which a gradient is computed, subtract, from the current values of the network parameters, an update that is determined by applying an optimize.
Wikipedia - Supervised Learning https:// en.wikipedia.org/wiki/Supervised_learning (retrieved May 16, 2026) discloses Supervised learning uses a training set to teach models to yield the desired output. The training dataset includes inputs and correct outputs, which allow the model to learn over time. The algorithm measures its accuracy through the loss function, adjusting until the error has been sufficiently minimized. This process involves several key steps: Data Collection and Preprocessing: Gathering labeled data and cleaning it to handle missing values and scale features.
Regarding Claim 2, (Previously Presented) [and similarly claim 9 and claim 15]
The computer-implemented method of claim 1, further comprising: determining a second PVD in response to modifying the process in accordance with the process transformation recommendation; generating a second process transformation recommendation in response to determining that the second PVD is greater than the predetermined threshold; and outputting the second process transformation recommendation.
See above. Zaslavasky [column 10 Lines 25-40], [Figure 2].
Regarding Claim 3, (Original) [and similarly claim 10 and claim 16
The computer-implemented method of The computer-implemented method of wherein the process transformation recommendation includes at least one of a recommendation to eliminate a process step identified as a redundant step of the process, a recommendation to restructure a process step based on an optimization determination, a recommendation to split an existing process step into two or more steps, or a recommendation to introduce a new step into the process.
See above. Zaslavasky [column 10 Lines 25-40], [Figure 2].
Al- Anqoudi [Figure 3], [p.15]-[16] [Figure 5] teaches Activity 1 – Activity 4., see above.
Zaslavasky teaches generating performance improvement recommendations for machine learning models. Al-Anqoudi teaches business process re-engineering to manage progress towards achieving targets and objectives. It would have been obvious to combine before the effective filing date, evaluating performance metrics for multiple implementations of a machine learning model, as taught by Zaslavasky, with utilis[ing] machine learning for business process management in general and business process re-engineering, as taught by Al-Anqoudi because re-engineering a business process will sometimes transform every aspect of an organisation, including organisation structures, values, and reward systems [15], maximising the impact of the change on the organisation culture., Al-Anqoudi [p.2]
Regarding Claim 4, (Original) [and similarly claim 11 and claim 17]
The computer-implemented method of claim 1, further comprising: using process volumetrics from the process mining to determine a total number of steps of the process; determining a process transformation type of the process transformation recommendation; generating a process transformation recommendation percentage based on the total number of steps and process transformation type; and generating the PTP score based on the process transformation recommendation percentage.
See above. Zaslavasky [column 10 Lines 25-40], [Figure 2]. Zaslavasky [column 9 lines 14-30].
Al- Anqoudi [Figure 3], [p.13].
Zaslavasky teaches generating performance improvement recommendations for machine learning models. Al-Anqoudi teaches business process re-engineering to manage progress towards achieving targets and objectives. It would have been obvious to combine before the effective filing date, evaluating performance metrics for multiple implementations of a machine learning model, as taught by Zaslavasky, with utilis[ing] machine learning for business process management in general and business process re-engineering, as taught by Al-Anqoudi because re-engineering a business process will sometimes transform every aspect of an organisation, including organisation structures, values, and reward systems [15], maximising the impact of the change on the organisation culture., Al-Anqoudi [p.2]
Regarding Claim 5, (Original) [and similarly claim 12 and claim 18]
The computer-implemented method of claim 1, wherein the mapping maps each of the KPIs to at least one of a plurality of categories, each of the plurality of categories indicating a predetermined process activity type.
See above. Zaslavasky [column 10 Lines 25-40], [Figure 2]. Zaslavasky [column 9 lines 14-30].
Regarding Claim 6, (Original) [and similarly claim 19]
The computer-implemented method of claim 5, wherein the plurality of categories includes an efficiency category, an experience category, and a compliance category.
See above. Zaslavasky [column 10 Lines 25-40], [Figure 2]. Zaslavasky [column 9 lines 14-30].
Regarding Claim 7, (Previously Presented) [and similarly claim 13 and claim 20]
The computer-implemented method of The computer-implemented method of wherein the process is a computer-implemented process, and at least one of the one or more process transformations within the process transformation recommendation is performed on the computer-implemented process.
See above. Zaslavasky [column 10 Lines 25-40], [Figure 2]. Zaslavasky [column 9 lines 14-30].
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Marquez-Chamorro (2018, Predictive Monitoring of Business Processes: A Survey) teaches business process management (BPM) which consists in the extraction of information from the events logs of a business process and machine learning employed in BP monitoring.
Makhija (US 2023/0137639 A1) uses AI data to data model a pack-ship operation.
Wolf (2008, SODA: An Optimizing Scheduler for Large-Scale Stream-Based Distributed Computer System) where SODA wants scheduler must be able to shift resource allocation dynamically in response to changes to resource availability, job arrivals and departures, incoming data rates.
Zhai (US 2022/0189612 A1) teaching the system can modify the supervised learning technique after the pre-training to no longer use weight decay. Some supervised learning techniques, after each training step in which a gradient is computed, subtract, from the current values of the network parameters, an update that is determined by applying an optimize.
Wikipedia - Supervised Learning https:// en.wikipedia.org/wiki/Supervised_learning (retrieved May 16, 2026) discloses Supervised learning uses a training set to teach models to yield the desired output. The training dataset includes inputs and correct outputs, which allow the model to learn over time. The algorithm measures its accuracy through the loss function, adjusting until the error has been sufficiently minimized. This process involves several key steps: Data Collection and Preprocessing: Gathering labeled data and cleaning it to handle missing values and scale features.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to THEA LABOGIN whose telephone number is (571)272-9149. The examiner can normally be reached Monday -Friday, 8am-5pm.
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, Patricia Munson can be reached on 571-270- 5396. 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.
/THEA LABOGIN/Examiner, Art Unit 3624
Prosecution Insights