DETAILED ACTION
The following NON-FINAL Office Action is in response to Applicant’s Remarks filed on 04/10/2026.
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 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.
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 04/10/2026 has been entered.
Status of Claims
Claims 1-20 were previously pending and subject to a final Office Action mailed 03/24/2026. Claims 1, 8, and 15 were amended. Claims 1-20 are currently pending and are subject to the non-final Office Action below.
Response to Arguments
35 USC § 112
Applicant has amended Claims 1, 8, and 15 to recite “without prescribing granular steps” as described in paragraph 11 of the specification. Accordingly, the 35 U.S.C. 112(b) rejections of Claims 1-20 have been rendered moot and thus, have been withdrawn.
35 USC § 101
Applicant’s arguments, see pages 7-11, filed 04/10/2026, with respect to the 35 U.S.C. 101 rejections of Claims 1-20 have been fully considered and are not persuasive.
Regarding Applicant’s arguments in section A, Examiner respectfully disagrees.
Examiner understands that Applicant’s invention is directed to converting a CMMN file to a BPMN file (para. 42 of specification). However, “converting” does not require a computational result as Applicant argues where elements of a process model data structure (discretionary tasks, sentry connections, milestones) are transformed into elements of a different type of process model data structure (sub-processes, sequences, events).
Paragraph 34 of Applicant’s specification describes several insights and paragraph 35 states “The insights can then be used to optimize the CMMN by creating several recommended ways to optimize the CMMN, such as removing the step of customer satisfaction, specifying an explicit mandatory task with century condition to involve the manager for big contracts, and always involve person A if the ticket is related to runtime module”.
Paragraph 50 of Applicant’s specification “the recommendation may be to involve person 1 because the system identifies that a ticket is related to a runtime module based on historical learning from observing running CMMN processes, and person 1 is involved in the runtime module”.
Given broadest reasonable interpretation, “converting” is directed to business relations where a user or an organization is receiving recommendations on how to optimize their business process. For example, changing a discretionary task such as “a user’s decision to involve person A in the process” into a non-interrupting event-based sub-processes where “the user must always involve person A if the ticket is related to a runtime module”. Such an interpretation would also work for converting sentry connections into sequences and converting achieved milestones into events.
Further, the interpretation is supported by paragraph 12 of Applicant’s specification, where tasks are individual work items or activities that need to be performed, sentries are dynamic conditions/rules that act as a criterion to activate a task, and milestones are indications of initiation/completion of tasks.
Applicant’s arguments would be more convincing if the claim did not merely recite the end result where one element is converted into another. There is no “set-up” or detail to the conversion process. Discretionary tasks, sentry connections, and milestones do not read as technical elements but merely data elements that make up a process.
Examiner recommends Applicant further clarify/provide the conversion process. For example, providing context as to what the initial data structure includes and further, how the data structure is modified and converted.
Regarding Applicant’s arguments in section B specifically Applicant argues that the process model is a computer-implemented data structure and not the business process it represents. Examiner respectfully disagrees. The claim limitation recites “accessing an unstructured process model defining outcomes to be achieved without prescribing granular steps prior to execution of the unstructured process model”.
Given broadest reasonable interpretation “accessing” may amount to a business process as described in paragraph 33 of Applicant’s specification where an unstructured process model may involve ticket handling and include stages, tasks, and milestones – “Case: Ticket Handling
Stage 1: Analysis
Task 1: Upon the arrival of a ticket, it is assigned to L1 support for initial analysis.
Milestone: Closure milestone is initiated to track the progress.
Stage 2: Resolution
Task 2: L1 support provides a solution to address the ticket.
Milestone: Closure milestone is achieved to signify the ticket's successful resolution.
Stage 3: Optional Actions (Discretionary Tasks)
Task 3: L1 support may choose to review previous tickets for reference.
Task 4: L1 support may consult experts for guidance.
Task 5: L1 support may contact the customer for clarification.
Task 6: After resolving the ticket, L1 support may follow up with the customer to assess satisfaction.”
Thus, the unstructured process model is not a “computer-implemented data structure” but rather a business process which is input into a machine learning model and a heuristic rules engine to derive recommendations to optimize the business process.
See above for discussion regarding the discretionary tasks, sentry connections, milestones, etc.
Regarding Applicant’s arguments in section C, Examiner respectfully disagrees as “Converting” is part of the abstract idea and amounts to converting an unstructured process model (a sequence of outcomes without granular steps) using the recommendations to derive a structured process model.
Examiner understands that the result of Applicant’s inventive concept is that the BPMN file is suited for execution by process engines that require prescriptive sequence-based models. However, such technical elements are not clearly recited within the claim.
Regarding Applicant’s ML model and heuristic rules engine argument, Examiner respectfully disagrees. Even though the limitations specify inputs, operations, and outputs of each component, the components are still limiting the claims to the field of process modelling/mining/enhancement where analysis of the unstructured process model and context data is performed by the first machine learning model and heuristic rules engine.
Applicant’s arguments would be more convincing if the ML model or heuristic rules engine performed more specified technical features instead of generic analysis features limited to machine learning models or heuristic rules engines. For example, any machine learning model would take inputs, perform operations on the inputted data like training to output a specific result, and then, outputting the results.
Regarding Applicant’s arguments in section D, Examiner respectfully disagrees as the claimed invention “dual-path analysis”, “optimization”, and “conversion” amounts to organizing human activity and mathematical concepts along with mere instructions to apply the exception using generic computer components and “ML model and heuristic rules engine” amount to field of use.
Accordingly, the 35 U.S.C. 101 rejection is maintained.
35 USC § 103
Applicant’s arguments, see pages 12-16, filed 04/10/2026, with respect to the 35 U.S.C. 103 rejections of Claims 1-20 have been fully considered and are persuasive. The 35 U.S.C. 103 rejection has been withdrawn.
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 invention is directed to an abstract idea without significantly more.
Step 1
Claims 1-7 are directed to a system (i.e., a machine), Claims 8-14 are directed to a method (i.e., a process), and Claims 15-20 are directed to a system (i.e., a machine). Therefore, the claims all fall within one of the four statutory categories of invention.
Step 2A Prong 1
Independent Claim 1, 8, and 15 recites:
accessing an unstructured process model defining outcomes to be achieved without prescribing granular steps, prior to execution of the unstructured process model;
accessing context data regarding the unstructured process model, the context data including data gathered during past executions of the unstructured process model;
passing the unstructured process model and the context data …to output one or more recommendations on how to optimize an input unstructured process model, based on the context data, … output a first set of one or more recommendations on how to optimize the unstructured process model;
passing the unstructured process model and the context data … to apply rules defining recommended process patterns and behaviors to generate a second set of one or more recommendations on how to optimize the unstructured process model or use a process stored … based on comparison of attributes in the unstructured process model and context data with attributes stored…;
optimizing the unstructured data model by applying one or more recommendations in the first and/or second sets; and
converting the optimized unstructured data model into a structured data model by at least one of: converting discretionary tasks into non-interrupting event-based sub-processes, changing sentry connections into sequences, or changing achieved milestones into events.
The limitations stated above are processes that under broadest reasonable interpretation covers “certain methods of organizing human activity” (“commercial interactions” or “managing personal behavior or relations or interactions between people”). Specifically, business relations where a user or an organization is receiving recommendations on how to optimize their business process in light of Applicant’s specification paragraph 34-35 “The context data and runtime execution logs can then be used as input to an AI/ML analysis to provide one or more insights about the case. Examples of such insights may include: 1) L1 never resolved issues with “runtime module” and always involves “Person A” for consultation. 2) Even after finding product related queries in previous ticket-always manager is involved for consultation if the customer's contract is above 100 million. 3) Customer satisfaction was asked only in 2 cases, etc. These insights can then be used to optimize the CMMN by creating several recommended ways to optimize the CMMN, such as removing the step of “customer satisfaction”, specifying an explicit mandatory task with century condition to involve the manager for big contracts, and always involve “Person A” if the ticket is related to “runtime module””.
Further, specification paragraph 32 states “An optimization, informed by the data analysis, AI insights, and heuristic assessments, allows for improved results. Structured recommendations emerged, including the removal of redundant or occasionally used activities, reassignment of resources, and the introduction of efficiency-improving measures. Resource allocation optimization was achieved, ensuring that resources were efficiently matched to the demands of each case or activity. Efficiency improvements includes the streamlining of complex decision-making, automation of repetitive tasks, and optimization of activity sequencing. The optimized process aligned with industry best practices, regulatory requirements, and organizational standards. The optimized unstructured process can be seamlessly transitioned into a structured model such as BPMN, facilitating efficient and well-defined implementation”.
Accordingly, when given the broadest reasonable interpretation in light of specification paragraph 32, “optimizing” recites mathematical calculations. Thus, the claim will be considered as falling within the “mathematical concepts” grouping as well.
Accordingly, the claims recite an abstract idea.
Step 2A Prong 2
This judicial exception is not integrated into a practical application. The independent claims also recite at least one hardware processor, a computer-readable medium, a first machine learning model, a heuristic rules engine, a repository, and a repository of processes.
The additional elements of at least one hardware processor, a computer-readable medium, a first machine learning model, a heuristic rules engine, a repository, and a repository of processes are all recited at a high-level of generality (generic computer/functions) such that when viewed as a whole/ordered combination, it amounts to no more than mere instructions to apply the judicial exception using generic computer components. See MPEP 2106.05(f).
Further, a first machine learning model “trained to output one or more recommendations” and a heuristic rules engine which “apply rules… to generate a second set of one or more recommendations” limits the abstract idea to a particular field of use. Specifically, limiting the claims to the field of process modelling/mining/enhancement where analysis of the unstructured process model and context data is performed by the first machine learning model and heuristic rules engine.
Thus, the claim as a whole, looking at additional elements individually and in combination, does not integrate the judicial exception into a practical application as the additional elements are mere instructions to apply the judicial exception using generic computer components and field of use which does not impose meaningful limits on practicing the abstract idea. The claims are directed to an abstract idea.
Step 2B
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements of at least one hardware processor, a computer-readable medium, a first machine learning model, a heuristic rules engine, a repository, and a repository of processes to perform the steps/functions recited above amounts to no more than mere instructions to apply the exception using a generic computer. Mere instructions to apply the exception using a generic computer component cannot provide an inventive concept.
Again, a first machine learning model “trained to output one or more recommendations” and a heuristic rules engine which “apply rules….to generate a second set of one or more recommendations” limits the abstract idea to a particular field of use. Specifically, limiting the claims to the field of process modelling/mining/enhancement where analysis of the unstructured process model and context data is performed by the first machine learning model and heuristic rules engine.
None of the steps/functions of Claim 1, Claim 8, and Claim 15 when evaluated individually or as an ordered combination amount to significantly more than the abstract idea. The additional elements are merely used to perform the limitations directed to the abstract idea and amount to no more than mere instructions to apply the exception using a generic computer or field of use, thus, the analysis does not change when considered as an ordered combination. Further, the additional elements do not meaningfully limit the claim. Accordingly, Claim 1, Claim 8, and Claim 15 are ineligible.
Dependent Claims 2, 9, and 16 further specify transforming the context data using “data normalization”. When considered as an additional element, data normalization amounts to extra-solution activity. Specifically, “selecting a particular data source or type of data to be manipulated”. See MPEP2106.05(g).
Applicant’s specification para. 16 “The collected data undergoes rigorous pre-processing, ensuring its quality and consistency. Data cleaning procedures systematically eliminate errors and inaccuracies, guaranteeing the reliability and accuracy of the dataset. Data normalization is then applied to standardize units of measurement, making it possible to compare and analyze different data elements effectively” demonstrates that data normalization is well-known, routine, and conventional as it need not be described in detail.
Dependent Claims 3, 10, and 17 recite “extracting”, dependent claims 4, 11, and 18 recite “calculate”, dependent Claims 5, 12, and 19 recite “group”, dependent claims 6, 13, and 20 recite “reduce” which are limitations that further narrow the abstract idea of organizing human activity. Further, feature extraction, Naïve Bayes classifier (used to calculate a probability), K-means clustering, and PCA are directed to mathematical calculations or relationships and thus, fall into the abstract idea grouping of “mathematical concepts” as well.
The feature extraction process, wherein the first machine learning model is a Naïve Bayes classifier, wherein the first machine learning model is a K-means clustering model, and Principal Component Analysis, when viewed as additional elements, amounts to limits the abstract idea to a particular field of use. Specifically, limiting the claims to the field of process modelling/mining/enhancement where analysis and processing of the context data is limited to execution by the feature extraction process, Naïve Bayes classifier, K-means clustering model, and/or PCA.
Dependent Claims 7 and 14 further specify suggesting the one or more recommendations to a user via a user interface prior to the optimizing. “Suggesting” is part of the abstract idea of organizing human activity in light of specification paragraph 50 where recommendations are displayed to the user allowing the user to choose to modify the model or not. The user interface is recited at a high-level of generality (generic computer/functions) such that when viewed as a whole/ordered combination, it amounts to no more than mere instructions to apply the judicial exception using generic computer components.
Again, such limitations are further directed towards mathematical concepts and/or organizing human activity as a user or an organization is receiving recommendations on how to optimize their business process and the additional elements are extra-solution activity, mere instructions to apply the judicial exception or field of use.
Thus, taken alone and when viewed in combination, nothing in dependent claims 2-7, 9-14, and 16-20 adds additional elements that are sufficient to amount to significantly more than the judicial exception. Claims 1-20 are ineligible.
Closest Prior Art
Examiner noting that Claims 1-20 are rejected under 35 U.S.C. 101.
The following is a statement of reasons for the indication of closest prior art: current prior art alone or in combination fail to disclose every element of the independent claims, specifically the limitations of “accessing” and “converting”.
The following are the closest prior art:
Wilk et al. (US2025/0200485) teaches a module may utilize linguistic analysis techniques to analyze process event logs and provide one or more recommendations to the user using the output of the machine learning model so that the user may select which of the recommendations to implement. Further, the module uses the feedback to enable the RPA bot to perform additional tasks/activities within the process which may suffer from inefficiencies or bottlenecks and the RPA bot may perform activities which were previously not automated. Wilk does not teach “accessing an unstructured process model defining outcomes to be achieved without prescribing granular steps” as Wilks is implementing recommendations for an existing process.
Hooks et al. (US Patent No. 12,118,490) teaches monitoring multiple instances of the workflow, scoring the workflow with one or more heuristic rules and determining one or more workflow patterns based on the scoring, and providing one or more insights about the workflow. Further, the scoring is based on heuristic algorithms or rules based on considerations of what types of workflow choices are likely to be productive or efficient choices. Hooks does not teach “accessing” or “converting”.
Mueller et al. (US2024/0303571) teaches a risk manager formulates a control (a textural description or workflow) and a machine learning model automatically converts the control into a business process model and notation (BPMN) workflow. Further, the control follows business rules.
Dianov et al. (US2020/0342382) teaches mapping and converting Gantt tasks, subtasks, milestones into BPMN process model.
Hull et al. (US2019/0087755) models a business process using BPMN, CMMN, and/or DMN where unstructured input is used to generate a specification of business processes which are then used to generate business process models.
Killisperger et al. (US2009/0235225) teaches a reference process (workflow), a high-level instantiation (first adaptation of the structure of the process), detailed instantiation (sub processes), and implementation of instantiation process.
El Hattami et al. (US2023/0376838) in para. 15 teaches discovering steps of actions that have been taken to resolve problems in situations where a formal workflow does not yet exist and a machine learning framework that automatically extracts workflows from dialogues.
Porto Guedes et al. (US2023/0117225) teaches creating an updated workflow by identifying a problem and determining solutions to implement.
Sridhara et al. (US2021/0304139) teaches receiving an undeveloped process flow diagram and converting it into a feature vector and a machine learning model converts the feature vector into a predicted process flow diagram element.
Gandhi et al. (US2024/0419641) teaches based on a set of rules, providing potential modifications to unstructured data as a recommendation to a user
Buhler (US Patent No. 11,610,137) teaches model embodiments including BPMN, CMMN, and DMN which may be combined to model the behavior of the agent process
Gupta et al. (US2025/0373516) teaches heuristic models and machine learning models processing structured input data in order to output policy recommendations.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Lisa Ma whose telephone number is (571)272-2495. The examiner can normally be reached Monday to Thursday 7 AM - 5 PM.
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, Shannon Campbell can be reached at (571)272-5587. 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.
/L.M./Examiner, Art Unit 3628
/SHANNON S CAMPBELL/Supervisory Patent Examiner, Art Unit 3628
Prosecution Insights