DETAILED ACTION
Notice of Pre-AIA or AIA Status
[1] The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Continued Examination Under 37 CFR 1.114
[2] 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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 15 June 2026 has been entered.
Notice to Applicant
[3] This communication is in response to the Amendment and the Request for Continued Examination (RCE) filed 15 June 2026. Claims 5-7 and 15-17 have been cancelled. Claims 1, 12, and 20 have been amended. Claims 23-25 have been added. Claims 1-4, 8-14, and 18-25 are pending.
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 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.
[4] Claim(s) (now claims 1-4, 8-14, and 18-25) is/are rejected under 35 U.S.C. 103 as being unpatentable over Tripathy et al. (United States Patent Application Publication No. 2025/0139417 hereinafter ‘Tripathy’) in view of Roper, JR et al. (United States Patent Application Publication No. 2025/0217114 hereinafter ‘Roper’) and further in view of Kulkarni et al. (United States Patent Application Publication No. 2024/0134682 hereinafter ‘Kulkarni’).
With respect to (currently amended) claim 1, Tripathy discloses a method comprising: receiving a specification of a desired workflow (Tripathy et al.; paragraphs [0022] [0027] [0053] [0072]; See at least text/natural language problem/workflow description. See further mapping natural language description to groups of workflow templates); and generating a template for the desired workflow via a workflow predictive model (Tripathy et al.; paragraphs [0037] [0053]; See at least machine learning model maps description to template), wherein generating the template includes: identifying one of the one or more workflow clusters corresponding to the specification of the desired workflow (Tripathy et al.; paragraphs [0029] [0076]-[0082]; See at least mapping to template types in template repository); and using a portion of the workflow predictive model trained using the identified one workflow pattern cluster to generate the template for the desired workflow (Tripathy et al.; paragraphs [0029] [0037]; See at least machine-learning model trained to learning mappings to map description statement to types of templates in the template repository).
Claim 1 has been amended to further include “…transforming a plurality of existing workflows into corresponding workflow embeddings; transforming text descriptions in a natural language format associated with the plurality of existing workflows into corresponding description embeddings;”
With respect to generation of workflow and description embeddings and workflow pattern clusters, Tripathy discloses workflow patterns for a user and further references and applies a mapping process to groups of templates to match a pattern of activities/tasks to a template. Tripathy generally indicates that word embeddings can be applied to calculate similarities. While Tripathy discloses workflow patterns and groups of workflows by type, Tripathy fails to apply a clustering process to form a group of templates by pattern.
However, as evidenced by Roper, it is well-known in the art to generate a joint embedding space that includes digital tasks and digital workflow embeddings to create and match workflow templates for recommending a template for a specified workflow input (Roper et al.; paragraphs [0044] [0105] [0106] [0242] [0253] [0277]; See at least template embeddings and clustering of templates based on historical usage patterns. See further generation of embeddings for digital tasks and workflow templates to match embeddings to recommend a best-fit template. See further embedding generator). As further evidenced by Roper, it is well-known in the art to apply usage patterns to workflow templates to generate clusters of workflow templates based on usage patterns and to further apply template embeddings for future mapping to workflow descriptions (Roper et al.; paragraphs [0261] [0262]; See at least template embeddings and clustering of templates based on historical usage patterns).
Claim 1 has been amended to further specify that the previously recited “clustering” is performed “…clustering, using a clustering machine learning model, a plurality of workflow patterns into one or more workflow pattern clusters based at least in part on the corresponding workflow embeddings and the corresponding description embeddings, wherein each of the plurality of workflow patterns was based on one or more of the plurality of existing workflows…”
While Roper teaches a vector space including workflow template embeddings, i.e., workflow embeddings, and digital task embeddings, i.e., description embeddings, for the purpose of matching tasks/descriptions to templates, neither Tripathy nor Roper teaches a clustering algorithm.
However, as evidenced by Kulkarni, the use of clustering machine learning models to identify workflow pattern clusters to generate a workflow template is well-known in the art (Kulkarni et al.; paragraphs [0043]-[0047] [0052]; See at least application of clustering algorithm to generate clusters of workflows and see use of machine learning mode to classify/cluster workflows based on pattern).
It would have been obvious to one of ordinary skill in the art at the time the invention was made to have modified the workflow type-based repositories and description to template matching by similarity scoring of Tripathy by further including well-known template matching by template and description embeddings and clustering workflow templates by usage patterns as taught by Roper. The instant invention is directed to a system and method of generating a workflow template from natural language descriptions. As Tripathy disclose the use of workflow type-based repositories and description to template matching by similarity scoring in the context of a system and method for generating a workflow template from natural language descriptions and Roper similarly discloses the utility template matching by template and description embeddings and clustering workflow templates by usage patterns in the context of a system and method for generating a workflow template from natural language descriptions, the teachings are reasonably considered to have been derived from analogous references and applied in the manner disclosed by the respective references. Accordingly, one of ordinary skill in the art would have been motivated to make the noted combination/modification as rationalized by the simple substitution of one known element (e.g., similarity scoring to match templates and workflow type repository) for another (e.g., embeddings to perform similarity matching and template clustering based on usage patterns) to obtain the predictable result of continually improving workflow generation by accurately matching available templates to a general description of a workflow need, thereby reducing lengthy development times and minimizing the need for human operators having specialized knowledge (Tripathy; paragraph [0002]).
Regarding the combination that includes Kulkarni, it would have been obvious to one of ordinary skill in the art at the time the invention was made to have modified the workflow type-based repositories and description to template matching by similarity scoring of Tripathy by further including well-known application of clustering algorithm to generate clusters of workflows and see use of machine learning mode to classify/cluster workflows based on pattern as taught by Kulkarni. The instant invention is directed to a system and method of generating a workflow template from natural language descriptions. As Tripathy disclose the use of workflow type-based repositories and description to template matching by similarity scoring in the context of a system and method for generating a workflow template from natural language descriptions and Kulkarni similarly discloses the utility of application of clustering algorithm to generate clusters of workflows and see use of machine learning models to classify/cluster workflows based on pattern in the context of a system and method for generating a workflow template from natural language descriptions, the teachings are reasonably considered to have been derived from analogous references and applied in the manner disclosed by the respective references. Accordingly, one of ordinary skill in the art would have been motivated to make the noted combination/modification as rationalized by the simple substitution of one known element (e.g., similarity scoring to match templates and workflow type repository) for another (e.g.,) to obtain the predictable result of continually improving workflow generation by accurately matching available templates to a general description of a workflow need, thereby reducing lengthy development times and minimizing the need for human operators having specialized knowledge.
With respect to claim 2, Tripathy discloses a method further comprising: training the workflow predictive model using the one or more workflow clusters (Tripathy et al.; paragraphs [0029] [0037]; See at least machine-learning model trained to learning mappings to map description statement to types of templates in the template repository).
With respect to claim 3, Tripathy discloses a method wherein the desired workflow comprises one or more workflow patterns (Tripathy et al.; paragraphs [0029] [0030]; See at least mapping to workflow template categories).
With respect to claim 4, Tripathy discloses a method wherein one of the plurality of workflow patterns comprises one or more of the following: a plurality of tasks arranged in a sequence, a plurality of tasks connected via a branch decision, or a plurality of tasks triggered by a particular trigger (Tripathy et al.; paragraphs [0039] [0047]-[0048] [0056]; See at least sequences and triggers and conditions to initiate tasks).
Claims 5-7 are cancelled.
With respect to claim 8, Tripathy discloses a method further comprising: transforming the specification of the desired workflow into a version of the specification provided to the workflow predictive model, comprising by: correcting one or more errors in the specification of the desired workflow; standardizing or normalizing the specification of the desired workflow; and transforming the specification of the desired workflow into a corresponding description embedding (Tripathy et al.; paragraphs [0022] [0025]; See at least user review and correction of workflow).
With respect to claim 9, Tripathy discloses a method further comprising: providing an editable version of the template for the desired workflow (Tripathy et al.; paragraphs [0022] [0025] [0059]-[0060]; See at least user review and correction of workflow. See further edit interface).
With respect to claim 10, Tripathy discloses a method wherein the editable version of the template for the desired workflow comprises one or more placeholder editable tasks (Tripathy et al.; paragraphs [0022] [0025] [0059]-[0060]; See at least user review and correction of workflow. See further edit interface).
With respect to claim 11, Tripathy discloses a method wherein the editable version of the template for the desired workflow comprises one or more graphical user interface (GUI) elements for adding additional workflow components (Tripathy et al.; paragraphs [0022] [0025] [0059]-[0060]; See at least user review and correction of workflow. See further edit interface).
With respect to claims 21 and 22, Tripathy discloses a method further comprising:
executing the desired workflow using the template in response to creation or update of a record associated with an application (Tripathy et al.; paragraphs [0022] [0025] [0059]-[0060]; See at least user review and correction of workflow. See further edit interface. The corrections are reasonably a form of update of a record).
With respect to (new) claim 23, Tripathy discloses a method further comprising:
receiving a subsequent specification including a requirement for modifying the desired workflow; and generating a modified version of the desired workflow based on the subsequent specification (Tripathy et al.; paragraphs [0022] [0025] [0059]-[0060]; See at least user review and correction of workflow).
With respect to (new) claim 24, Tripathy discloses a method wherein the desired workflow includes placeholders that have missing or blank parameters (Tripathy et al.; paragraphs [0022] [0025] [0059]-[0060]; See at least user review and correction of workflow. See further edit interface).
With respect to (new) claim 25, Tripathy discloses a method wherein the workflow predictive machine learning model comprises a plurality of model portions, and wherein training the workflow predictive machine learning model comprises training each of the model portions exclusively on workflows belonging to a respective one of the workflow pattern clusters (Tripathy et al.; paragraphs [0029] [0037]; See at least machine-learning model trained to learning mappings to map description statement to types of templates in the template repository).
Claims 12-14, 16-20, and 22 as presented by amendment substantially repeat the subject matter addressed above with respect to claims 1-4, 6-11, and 21 as directed to the enabling system and computer-readable medium storing computer-executable instructions. With respect to these elements, Tripathy discloses enabling the disclosed method employing analogous systems and executable instructions. Accordingly, claims 12-14, 16-20, and 22 are rejected under the applied teachings, conclusions obviousness, and rationale to modify as discussed above with respect to claims 1-4, 6-11, and 21.
Response to Remarks/Amendment
[5] Applicant's remarks filed 15 June 2026 have been fully considered and are addressed as follows:
[i] Applicant’s remarks directed to previous rejection(s) of claim(s) 1-4, 7-8, 12-13, 16-17, and 20 (now claims 1-4, 8-14, and 18-25) under 35 U.S.C. 103(a) as being unpatentable as set forth in the previous Office Action mailed 8 April 2026 have been fully considered and are not pursuasive.
In particular, the joint embedding space of Roper includes includes digital tasks and digital workflow embeddings to create and match workflow templates for recommending a template for a specified workflow input (Roper et al.; paragraphs [0044] [0105] [0106] [0242] [0253] [0277]; See at least template embeddings and clustering of templates based on historical usage patterns. See further generation of embeddings for digital tasks and workflow templates to match embeddings to recommend a best-fit template. See further embedding generator generates or “transforms” digital tasks, descriptions into embeddings and generates workflow template embeddings). As further evidenced by Roper, it is well-known in the art to apply usage patterns to workflow templates to generate clusters of workflow templates based on usage patterns and to further apply template embeddings for future mapping to workflow descriptions (Roper et al.; paragraphs [0261] [0262]; See at least template embeddings and clustering of templates based on historical usage patterns).
Conclusion
[6] The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Cited NON-PATENT Literature:
Shidaganti et al., Robotic Process Automation with AI and OCR to Improve Business Process: Review, 2021-08-04, 2021 Second International Conference on Electronics and Sustainable Communication Systems (ICESC) (2021, Page(s): 1612-1618): Relevant Teachings: Shidaganti provides a review systems/methods that provide process automation workflows using AI to enhance existing process automation tools. The publication establishes that at least using AI-assisted extraction of text descriptions to generate process automation tasks is common practice in the art.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ROBERT D RINES whose telephone number is (571)272-5585. The examiner can normally be reached M-F 9am - 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, Beth V Boswell can be reached at 571-272-6737. 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.
/ROBERT D RINES/Primary Examiner, Art Unit 3625
Prosecution Insights