Prosecution Insights
Last updated: November 04, 2026
Application No. 18/241,803

KNOWLEDGE BASE FOR AUTOMATED GENERAL KNOWLEDGE WORKER

Final Rejection §103
Filed
Sep 01, 2023
Examiner
GRUSZKA, DANIEL PATRICK
Art Unit
2121
Tech Center
2100 — Computer Architecture & Software
Assignee
Aico Inc.
OA Round
2 (Final)
33%
Grant Probability
At Risk
3-4
OA Rounds
1y 0m
Est. Remaining
99%
With Interview

Examiner Intelligence

Grants only 33% of cases
33%
Career Allowance Rate
2 granted / 6 resolved
-21.7% vs TC avg
Strong +67% interview lift
Without
With
+66.7%
Interview Lift
resolved cases with interview
Typical timeline
4y 2m
Avg Prosecution
24 currently pending
Career history
54
Total Applications
across all art units

Statute-Specific Performance

§101
38.6%
-1.4% vs TC avg
§103
47.3%
+7.3% vs TC avg
§102
7.6%
-32.4% vs TC avg
§112
5.4%
-34.6% vs TC avg
Black line = Tech Center average estimate • Based on career data from 6 resolved cases

Office Action

§103
Notice of Pre-AIA or AIA Status This Final communication is in response to Application No. 18/241,803 filed on 09/01/2023. The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA . Response to Amendment The amendments filed 8/4/2026 have been entered. They provide amendments to claims 1, 14, 20 and cancel claims 2-3 and 15. The amendments to the drawings overcome the objections. Response to Arguments Applicant’s arguments with respect to 35 U.S.C 103 filed 8/4/2026 (arguments B-D on pages 16-20 of applicant’s arguments) have been fully considered but they are not persuasive. Applicant argues that Pachet does not disclose the claimed positive and negative mapping examples because Pachet’s examples are used to train a classifier, whereas the claimed examples are included in text based prompts provided to a natural language processor and that there is no reason to combine Pachet with Kumar and Qin (arguments B and D of applicant’s arguments). The examiner respectfully disagrees. Pachet shows that positive and negative examples were used in helping train machine learning models. Qin teaches providing contextual information in an augmented text based prompt as input into an LLM (Qin [0019]). The combination would have suggested providing positive and negative examples as contextual information into an LLM. A person of ordinary skill in the art would have wanted to do this to provide classification examples to the language model to help train the models and thus improve accuracy. Applicant argues the combination of Kumar and Qin does not produce a text-based prompt that directs a model to output a mapping of a stored plan of action to nodes of a classification hierarchy. The examiner respectfully disagrees. Kumar teaches “mapping the plan of action to a version of a classification hierarchy for the organization to a natural language processor” as show in paragraphs [0038]-[0039]. Kumar just does not teach text based prompting. Qin however does teach text based prompting. It would have been obvious to implement Kumar’s mapping operation using Qin’s text based prompting. Applicant’s arguments (argument A pages 14-16 of applicant’s arguments) with respect to no prior art reference teaches a hierarchy for an organization with nodes that represent organizational units or roles of the organization have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument. Thus, the 103 rejection is maintained. 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. Claims 1, 4-14, and 16-20 are rejected under 35 U.S.C. 103 as being unpatentable over Kumar (US 2021/0089627 A1) in view of Qin (US 2024/0346256 A1), Homsi (US 7,373,310 B1) and Pachet (US 2010/0299328 A1). Regarding claim 1, Kumar teaches: One or more non-transitory computer-readable media storing computer-executable instructions that upon execution cause one or more processors to perform acts comprising: ([0076] “The computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device.”)) storing a plan of action for performing a knowledge worker task for an organization in a knowledge base, the plan of action comprising a plurality of programmatic objects; ([0043] “The processing unit (112) consumes the selected structure to an AI schema for use and implementation in the chatbot platform to support real-time communication flow. In one embodiment, the processing unit (112) converts the selected structure to a structured object notation, e.g. javscript object notation (JSON) format, and the AI manager (158) employs the JSON format of the selected structure in the chatbot platform (132).” Paragraph [0053] of applications specification says the plan of action can be JSON and include directions and parameters to perform a task) sending a text-based prompt for mapping the plan of action to a version of a classification hierarchy for the organization to a natural language processor (NLP); ([0038] “The relationship manager (154) is shown herein operatively coupled to the NL manager (152). The relationship manager (154) leverages the intents and entities identified by the NL manager (152) to map a relationship between the generated intents and entities. “ and [0039] “The topology manager (156) is shown herein operatively coupled to the relationship manager (154). The topology manager (156) functions to construct a topology, also referred to herein as a topological structure, representing the mapped relationship.” Also see Fig 11.) receiving from the NLP a mapping of the plan of action to one or more nodes in the version of the classification hierarchy; and ([0041] “The topology, e.g. hierarchical structure shown and described in FIG. 11, is constructed to support a series of communications, also referred to herein as a dialog. Referring to the knowledge base (170), a constructed topology is shown operatively coupled to each document.”) storing the mapping of the plan of action to the one or more nodes in the version of the classification hierarchy in the knowledge base. ([0041] “Referring to the knowledge base (170), a constructed topology is shown operatively coupled to each document.”) Kumar does not teach: sending a text-based prompt for mapping the plan of action to a version of a classification hierarchy for the organization to a natural language processor (NLP); the classification hierarchy comprising a plurality of nodes that represent respective organizational units or roles of the organization, and the text-based prompt including at least one of one or more positive mapping examples or one or more negative mapping examples, wherein a positive mapping example includes one or more programmatic objects of a first example plan of action that are designated as being correctly mapped to a corresponding node in a first example classification hierarchy, and wherein a negative mapping example includes one or more programmatic objects of a second example plan of action that are designated as being incorrectly mapped to a corresponding node in a second example classification hierarchy. However, Qin does teach text-based prompts ([0019] “In on implementation, a query may be used to retrieve pieces of augmentation information that may be included in a prompt to the LLM.”). Kumar and Qin are considered analogous art to the claimed invention because they are in the same field of endeavor being natural language processing. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to substitute the natural language processing of Kumar with the large language model of Qin. One would want to use a large language model to improve accuracy (Qin [0028]). Homsi teaches: the classification hierarchy comprising a plurality of nodes that represent respective organizational units or roles of the organization, (Col 7 lines 14-24 “Organizational Unit (OU): Any logical grouping of resources. An OU represents the hierarchical organizational model for an enterprise. The OU usually reflects the current structure of cost centers/profit centers. Organizational Hierarchy: Given a finite set of OU's E, E is an organization hierarchy if the function hierarchically reports to is a tree dimension of E. This particular tree dimension is by convention the hierarchical tree dimension of E. The topmost OU is the root element of an organization hierarchy and the parent element of another OU (Parent OU).”) Kumar, Qin and Homsi are considered analogous art to the claimed invention because they are in the same field of endeavor being natural language processing/hierarchies. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to substitute the natural language processing of Kumar with the large language model of Qin with the workflow activities being organized into organizational hierarchy with organizational units of Homsi. One would want to do this to have a workflow model to process different tasks and who performs these tasks. (Homsi background). Pachet teaches: and the text-based prompt including at least one of one or more positive mapping examples or one or more negative mapping examples, wherein a positive mapping example includes one or more programmatic objects of a first example plan of action that are designated as being correctly mapped to a corresponding node in a first example classification hierarchy, and wherein a negative mapping example includes one or more programmatic objects of a second example plan of action that are designated as being incorrectly mapped to a corresponding node in a second example classification hierarchy. ([0040] “In particular, when developing a tag model for a tag T, the training set contains "positive" and "negative" examples, that is, prototypes which conform to tag T (notably, prototypes to which the user has assigned tag T) and prototypes which do not conform to tag T.”) Kumar, Qin, Homsi and Pachet are considered analogous art to the claimed invention because they are in the same field of endeavor being machine learning. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to substitute the natural language processing of Kumar with the large language model of Qin and use the positive and negative training examples of Pachet. A person of ordinary skill in the art would have wanted to do this to provide classification examples to the language model to help train the models and thus improve accuracy. (Pachet [0072]). Regarding claim 4, Kumar in view of Qin, Homsi and Pachet teaches claim 1 as outlined above. Kumar further teaches: the knowledge base stores organizational data of the organization that is used by an automated general knowledge worker to automatically provide the plan of action for performing the knowledge worker task for the organization. ([0048] “The AI platform (150) and the associated tools (152)-(158) leverage the knowledge base (170) and one or more libraries, such as library.sub.A (172.sub.A) and library.sub.B (172.sub.B), to support communication in the chatbot platform (132). The AI manager (158) leverages the structures representing the documents in the knowledge base (170) to facilitate and enable communications, including reply communications, in the chatbot platform (132).”) Regarding claim 5, Kumar in view of Qin, Homsi and Pachet teaches claim 1 as outlined above. Qin further teaches: the NLP includes a large language model (LLM). ([0019] “In on implementation, a query may be used to retrieve pieces of augmentation information that may be included in a prompt to the LLM.”). Regarding claim 6, Kumar in view of Qin, Homsi and Pachet teaches claim 5 as outlined above. Kumar further teaches: generating a new version of the classification hierarchy based on at least one of existing organizational data stored in the knowledge base or new organizational data provided to the knowledge base; ([0042] “The topology manager (156) re-generates the structure with the combined primary verbs. The re-generation of the structure includes the topology manager (156) to designate a primary noun of the second document, e.g. document.sub.A,1 (174.sub.A,1), as a second parent node in the combined topology, and to designates a second secondary nouns in the second document as a second child node in the combined topology.”) sending an additional text-based prompt for mapping the plan of action to the new version of the classification hierarchy to the NLP; ([0042] “The topology manager (156) re-generates the structure with the combined primary verbs. The re-generation of the structure includes the topology manager (156) to designate a primary noun of the second document, e.g. document.sub.A,1 (174.sub.A,1), as a second parent node in the combined topology, and to designates a second secondary nouns in the second document as a second child node in the combined topology.”) receiving from the NLP a new mapping of the plan of action to one or more additional nodes in the new version of the classification hierarchy; and ([0042] “Accordingly, two or more documents may be merged, with the merge reflected in a re-generated combined topological structure.”) storing the mapping of the plan of action to the one or more additional nodes in the new version of the classification hierarchy in the knowledge base. ([0041] “Referring to the knowledge base (170), a constructed topology is shown operatively coupled to each document.”) Regarding claim 7, Kumar in view of Qin, Homsi and Pachet teaches claim 1 as outlined above. Kumar further teaches: determining that a text-based request matches a node of the one or more nodes in the version of the classification hierarchy; and ([0055] “he variable N.sub.Total is assigned to represent a quantity of documents in the repository or documents subject to processing (304), and a corresponding document counting variable, N, is initialized (306). Document.sub.N is identified or retrieved from the repository and is subject to parsing (308), which includes analyzing a document table of contents (TOC) and document content.”) providing contextual data that are relevant to the text-based request to the NLP for generating a new plan of action, the contextual data including the plan of action that maps to the node in the version of the classification hierarchy or a subcomponent of the plan of action. ([0042] “The topology manager (156) re-generates the structure with the combined primary verbs. The re-generation of the structure includes the topology manager (156) to designate a primary noun of the second document, e.g. document.sub.A,1 (174.sub.A,1), as a second parent node in the combined topology, and to designates a second secondary nouns in the second document as a second child node in the combined topology. Accordingly, two or more documents may be merged, with the merge reflected in a re-generated combined topological structure.” ) Kumar does not teach the contextual data. However Qin does ([0039] “Pre-processor 202 may receive contextual information 215 and/or query 216. In embodiments, contextual information 215 may describe the context of the user (e.g., user identifier, user role, user profile, user location, browsing history, etc.) and/or the context of the query (e.g., the current webpage, the product or service associated with the current webpage, query timestamp information, etc.).”) Kumar and Qin are considered analogous art to the claimed invention because they are in the same field of endeavor being natural language processing. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the idea of using contextual data of Qin with Kumar system of generating new actions. One would want to do this to create better actions/responses (Qin [0017]). Regarding claim 8, Kumar in view of Qin, Homsi and Pachet teaches claim 1 as outlined above. Kumar further teaches: the version of the classification hierarchy is one of multiple versions of the classification hierarchy, ([0035] “The AI platform (150) may receive input across the network (105) and leverage a data source (170), also referred to herein as a corpus or knowledge base, to selectively access documents and corresponding topological structures.”) receiving a selection of a particular version of the multiple versions of the classification hierarchy for use during a generation of a new plan of action; and ([0035] “The AI platform (150) may receive input across the network (105) and leverage a data source (170), also referred to herein as a corpus or knowledge base, to selectively access documents and corresponding topological structures.”) Qin further teaches: providing contextual data that includes one or more plans of actions that maps to a node in the particular version of the classification hierarchy or one or more subcomponents of the one or more plans of action to the NLP for generating the new plan of action. ([0039] “Pre-processor 202 may receive contextual information 215 and/or query 216. In embodiments, contextual information 215 may describe the context of the user (e.g., user identifier, user role, user profile, user location, browsing history, etc.) and/or the context of the query (e.g., the current webpage, the product or service associated with the current webpage, query timestamp information, etc.).”) Regarding claim 9, Kumar in view of Qin, Homsi and Pachet teaches claim 8 as outlined above. Kumar further teaches: the node in the particular version of the classification hierarchy matches to a text-based request for generating the new plan of action. ([0040] “The topology manager (156) interfaces with the NL manager (152) and the relationship manager (154) to reflect and document the identified nouns and their relationship to a primary node set at the root node in the hierarchy. The topology manager (156) establishes the hierarchical arrangement of the identified nouns, and designates a primary noun as a parent node in the topology and further designates a secondary noun as a child node in the topology.”) Regarding claim 10, Kumar in view of Qin, Homsi and Pachet teaches claim 8 as outlined above. Kumar further teaches: comparing the new plan of action to an additional plan of action that is generated based on contextual data that includes one or more additional plans of action that map to an additional node in an additional version of the classification hierarchy or one or more additional subcomponents of the one or more additional plans of action; and ([0042] “The relationship manager (154) includes functionality to merge, or in one embodiment join, select documents. More specifically, the relationship manager (154) identifies shared content between at least two documents, for example document.sub.A,0 (174.sub.A,0) and document.sub.A,1 (174.sub.A,1), and merges the corresponding structures, e.g. structure.sub.A,0 (178.sub.A,0) and structure.sub.A,1 (178.sub.A,1).”) determining one or more differences between the new plan of action and the additional plan of action based on a comparison of the new plan of action to the additional plan of action. ([0042] “The merge of the structures entails the NL manager (152) to identify similarity of the primary verb of the first document, e.g. document.sub.A,0 (174.sub.A,0), with a primary verb of the second document, e.g. document.sub.A,1 (174.sub.A,1), and the relationship manager (154) combining the primary verbs of the first and second documents into the root node. The topology manager (156) re-generates the structure with the combined primary verbs. The re-generation of the structure includes the topology manager (156) to designate a primary noun of the second document, e.g. document.sub.A,1 (174.sub.A,1), as a second parent node in the combined topology, and to designates a second secondary nouns in the second document as a second child node in the combined topology. Accordingly, two or more documents may be merged, with the merge reflected in a re-generated combined topological structure.”) Regarding claim 11, Kumar in view of Qin, Homsi and Pachet teaches claim 1 as outlined above. Kumar further teaches: the plan of action is stored as programmatic objects in the knowledge base. ([0072] “Program modules (1342) generally carry out the functions and/or methodologies of embodiments to create and manage the topological structures representing corresponding documents, and leveraging the structure(s) in the chatbot platform to support and enable communications therein.”) Regarding claim 12, Kumar in view of Qin, Homsi and Pachet teaches claim 1 as outlined above. Qin further teaches: storing organizational data regarding the organization in the knowledge base as programmatic objects; and ([0036] “Dataset(s) 112 may include one or more databases storing augmentation information that is used to respond to the query from GUI 114. In embodiments, augmentation information stored in dataset(s) 112 may include, but are not limited to, domain-specific information (e.g., information related to specific topics or fields), entity-specific information (e.g., internal or proprietary corporate information), product-specific information (e.g., information related to products of an entity), recent information unavailable at generation of the large language model (e.g., new facts that postdate the generation of the LLM), and/or information changed after generation of the large language model (e.g., facts that have changed since the generation of the LLM)”) Kumar further teaches: sending at least the programmatic objects of the organizational data to an NLP to prompt the NLP to generate the version of the classification hierarchy. ([0042] “The topology manager (156) re-generates the structure with the combined primary verbs. The re-generation of the structure includes the topology manager (156) to designate a primary noun of the second document, e.g. document.sub.A,1 (174.sub.A,1), as a second parent node in the combined topology, and to designates a second secondary nouns in the second document as a second child node in the combined topology.” Where Qin teaches the organizational data as outlined above) Regarding claim 13, Kumar in view of Qin, Homsi and Pachet teaches claim 12 as outlined above. Kumar further teaches: the sending at least the programmatic objects of the organizational data includes sending the programmatic objects and an existing classification hierarchy to the NLP. ([0072] “Program modules (1342) generally carry out the functions and/or methodologies of embodiments to create and manage the topological structures representing corresponding documents, and leveraging the structure(s) in the chatbot platform to support and enable communications therein.”) Regarding claim 14, Kumar teaches: A computer-implemented method, comprising: ([0005] “The embodiments include a system, computer program product, and method to support a chatbot platform.”) storing a plan of action for performing a knowledge worker task for an organization in a knowledge base, the plan of action comprising a plurality of programmatic objects; ([0043] “The processing unit (112) consumes the selected structure to an AI schema for use and implementation in the chatbot platform to support real-time communication flow. In one embodiment, the processing unit (112) converts the selected structure to a structured object notation, e.g. javscript object notation (JSON) format, and the AI manager (158) employs the JSON format of the selected structure in the chatbot platform (132).” Paragraph [0053] of applications specification says the plan of action can be JSON and include directions and parameters to perform a task) sending a text-based prompt for mapping the plan of action to a version of a classification hierarchy for the organization to a natural language processor (NLP); ([0038] “The relationship manager (154) is shown herein operatively coupled to the NL manager (152). The relationship manager (154) leverages the intents and entities identified by the NL manager (152) to map a relationship between the generated intents and entities. “ and [0039] “The topology manager (156) is shown herein operatively coupled to the relationship manager (154). The topology manager (156) functions to construct a topology, also referred to herein as a topological structure, representing the mapped relationship.” Also see Fig 11.) receiving from the NLP a mapping of the plan of action to one or more nodes in the version of the classification hierarchy; and ([0041] “The topology, e.g. hierarchical structure shown and described in FIG. 11, is constructed to support a series of communications, also referred to herein as a dialog. Referring to the knowledge base (170), a constructed topology is shown operatively coupled to each document.”) storing the mapping of the plan of action to the one or more nodes in the version of the classification hierarchy in the knowledge base. ([0041] “Referring to the knowledge base (170), a constructed topology is shown operatively coupled to each document.”) Kumar does not teach: sending a text-based prompt for mapping the plan of action to a version of a classification hierarchy for the organization to a natural language processor (NLP); the classification hierarchy comprising a plurality of nodes that represent respective organizational units or roles of the organization, and the text-based prompt including at least one of one or more positive mapping examples or one or more negative mapping examples, wherein a positive mapping example includes one or more programmatic objects of a first example plan of action that are designated as being correctly mapped to a corresponding node in a first example classification hierarchy, and wherein a negative mapping example includes one or more programmatic objects of a second example plan of action that are designated as being incorrectly mapped to a corresponding node in a second example classification hierarchy. However, Qin does teach text-based prompts ([0019] “In on implementation, a query may be used to retrieve pieces of augmentation information that may be included in a prompt to the LLM.”). Kumar and Qin are considered analogous art to the claimed invention because they are in the same field of endeavor being natural language processing. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to substitute the natural language processing of Kumar with the large language model of Qin. One would want to use a large language model to improve accuracy (Qin [0028]). Homsi teaches: the classification hierarchy comprising a plurality of nodes that represent respective organizational units or roles of the organization, (Col 7 lines 14-24 “Organizational Unit (OU): Any logical grouping of resources. An OU represents the hierarchical organizational model for an enterprise. The OU usually reflects the current structure of cost centers/profit centers. Organizational Hierarchy: Given a finite set of OU's E, E is an organization hierarchy if the function hierarchically reports to is a tree dimension of E. This particular tree dimension is by convention the hierarchical tree dimension of E. The topmost OU is the root element of an organization hierarchy and the parent element of another OU (Parent OU).”) Kumar, Qin and Homsi are considered analogous art to the claimed invention because they are in the same field of endeavor being natural language processing/hierarchies. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to substitute the natural language processing of Kumar with the large language model of Qin with the workflow activities being organized into organizational hierarchy with organizational units of Homsi. One would want to do this to have a workflow model to process different tasks and who performs these tasks. (Homsi background). Pachet teaches: and the text-based prompt including at least one of one or more positive mapping examples or one or more negative mapping examples, wherein a positive mapping example includes one or more programmatic objects of a first example plan of action that are designated as being correctly mapped to a corresponding node in a first example classification hierarchy, and wherein a negative mapping example includes one or more programmatic objects of a second example plan of action that are designated as being incorrectly mapped to a corresponding node in a second example classification hierarchy. ([0040] “In particular, when developing a tag model for a tag T, the training set contains "positive" and "negative" examples, that is, prototypes which conform to tag T (notably, prototypes to which the user has assigned tag T) and prototypes which do not conform to tag T.”) Kumar, Qin, Homsi and Pachet are considered analogous art to the claimed invention because they are in the same field of endeavor being machine learning. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to substitute the natural language processing of Kumar with the large language model of Qin and use the positive and negative training examples of Pachet. A person of ordinary skill in the art would have wanted to do this to provide classification examples to the language model to help train the models and thus improve accuracy. (Pachet [0072]). Regarding claim 16, Kumar in view of Qin, Homsi and Pachet teaches claim 14 as outlined above. Qin further teaches: the NLP includes a large language model (LLM). ([0019] “In on implementation, a query may be used to retrieve pieces of augmentation information that may be included in a prompt to the LLM.”). Regarding claim 17, Kumar in view of Qin, Homsi and Pachet teaches claim 14 as outlined above. Kumar further teaches: generating a new version of the classification hierarchy based on at least one of existing organizational data stored in the knowledge base or new organizational data provided to the knowledge base; ([0042] “The topology manager (156) re-generates the structure with the combined primary verbs. The re-generation of the structure includes the topology manager (156) to designate a primary noun of the second document, e.g. document.sub.A,1 (174.sub.A,1), as a second parent node in the combined topology, and to designates a second secondary nouns in the second document as a second child node in the combined topology.”) sending an additional text-based prompt for mapping the plan of action to the new version of the classification hierarchy to the NLP; ([0042] “The topology manager (156) re-generates the structure with the combined primary verbs. The re-generation of the structure includes the topology manager (156) to designate a primary noun of the second document, e.g. document.sub.A,1 (174.sub.A,1), as a second parent node in the combined topology, and to designates a second secondary nouns in the second document as a second child node in the combined topology.”) receiving from the NLP a new mapping of the plan of action to one or more additional nodes in the new version of the classification hierarchy; and ([0042] “Accordingly, two or more documents may be merged, with the merge reflected in a re-generated combined topological structure.”) storing the mapping of the plan of action to the one or more additional nodes in the new version of the classification hierarchy in the knowledge base. ([0041] “Referring to the knowledge base (170), a constructed topology is shown operatively coupled to each document.”) Regarding claim 18, Kumar in view of Qin, Homsi and Pachet teaches claim 14 as outlined above. Kumar further teaches: determining that a text-based request matches a node of the one or more nodes in the version of the classification hierarchy; and ([0055] “he variable N.sub.Total is assigned to represent a quantity of documents in the repository or documents subject to processing (304), and a corresponding document counting variable, N, is initialized (306). Document.sub.N is identified or retrieved from the repository and is subject to parsing (308), which includes analyzing a document table of contents (TOC) and document content.”) providing contextual data that are relevant to the text-based request to the NLP for generating a new plan of action, the contextual data including the plan of action that maps to the node in the version of the classification hierarchy or a subcomponent of the plan of action. ([0042] “The topology manager (156) re-generates the structure with the combined primary verbs. The re-generation of the structure includes the topology manager (156) to designate a primary noun of the second document, e.g. document.sub.A,1 (174.sub.A,1), as a second parent node in the combined topology, and to designates a second secondary nouns in the second document as a second child node in the combined topology. Accordingly, two or more documents may be merged, with the merge reflected in a re-generated combined topological structure.” ) Kumar does not teach the contextual data. However Qin does ([0039] “Pre-processor 202 may receive contextual information 215 and/or query 216. In embodiments, contextual information 215 may describe the context of the user (e.g., user identifier, user role, user profile, user location, browsing history, etc.) and/or the context of the query (e.g., the current webpage, the product or service associated with the current webpage, query timestamp information, etc.).”) Kumar and Qin are considered analogous art to the claimed invention because they are in the same field of endeavor being natural language processing. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the idea of using contextual data of Qin with Kumar system of generating new actions. One would want to do this to create better actions/responses (Qin [0017]). Regarding claim 19, Kumar in view of Qin, Homsi and Pachet teaches claim 14 as outlined above. Kumar further teaches: the version of the classification hierarchy is one of multiple versions of the classification hierarchy, ([0035] “The AI platform (150) may receive input across the network (105) and leverage a data source (170), also referred to herein as a corpus or knowledge base, to selectively access documents and corresponding topological structures.”) receiving a selection of a particular version of the multiple versions of the classification hierarchy for use during a generation of a new plan of action; and ([0035] “The AI platform (150) may receive input across the network (105) and leverage a data source (170), also referred to herein as a corpus or knowledge base, to selectively access documents and corresponding topological structures.”) Qin further teaches: providing contextual data that includes one or more plans of actions that maps to a node in the particular version of the classification hierarchy or one or more subcomponents of the one or more plans of action to the NLP for generating the new plan of action. ([0039] “Pre-processor 202 may receive contextual information 215 and/or query 216. In embodiments, contextual information 215 may describe the context of the user (e.g., user identifier, user role, user profile, user location, browsing history, etc.) and/or the context of the query (e.g., the current webpage, the product or service associated with the current webpage, query timestamp information, etc.).”) Regarding claim 20, A system, comprising: one or more processors; and ([0005] “[0005] The embodiments include a system, computer program product, and method to support a chatbot platform.”) memory including a plurality of computer-executable components that are executable by the one or more processors to perform a plurality of actions, the plurality of actions comprising: ([0070] “The components of host (1302) may include, but are not limited to, one or more processors or processing units (1304), e.g. hardware processors, a system memory (1306), and a bus (1308) that couples various system components including system memory (1306) to processor (1304).”) storing a plan of action for performing a knowledge worker task for an organization in a knowledge base, the plan of action comprising a plurality of programmatic objects; ([0043] “The processing unit (112) consumes the selected structure to an AI schema for use and implementation in the chatbot platform to support real-time communication flow. In one embodiment, the processing unit (112) converts the selected structure to a structured object notation, e.g. javscript object notation (JSON) format, and the AI manager (158) employs the JSON format of the selected structure in the chatbot platform (132).” Paragraph [0053] of applications specification says the plan of action can be JSON and include directions and parameters to perform a task) sending a text-based prompt for mapping the plan of action to a version of a classification hierarchy for the organization to a natural language processor (NLP); ([0038] “The relationship manager (154) is shown herein operatively coupled to the NL manager (152). The relationship manager (154) leverages the intents and entities identified by the NL manager (152) to map a relationship between the generated intents and entities. “ and [0039] “The topology manager (156) is shown herein operatively coupled to the relationship manager (154). The topology manager (156) functions to construct a topology, also referred to herein as a topological structure, representing the mapped relationship.” Also see Fig 11.) receiving from the NLP a mapping of the plan of action to one or more nodes in the version of the classification hierarchy; and ([0041] “The topology, e.g. hierarchical structure shown and described in FIG. 11, is constructed to support a series of communications, also referred to herein as a dialog. Referring to the knowledge base (170), a constructed topology is shown operatively coupled to each document.”) storing the mapping of the plan of action to the one or more nodes in the version of the classification hierarchy in the knowledge base. ([0041] “Referring to the knowledge base (170), a constructed topology is shown operatively coupled to each document.”) Kumar does not teach: sending a text-based prompt for mapping the plan of action to a version of a classification hierarchy for the organization to a natural language processor (NLP); the classification hierarchy comprising a plurality of nodes that represent respective organizational units or roles of the organization, and the text-based prompt including at least one of one or more positive mapping examples or one or more negative mapping examples, wherein a positive mapping example includes one or more programmatic objects of a first example plan of action that are designated as being correctly mapped to a corresponding node in a first example classification hierarchy, and wherein a negative mapping example includes one or more programmatic objects of a second example plan of action that are designated as being incorrectly mapped to a corresponding node in a second example classification hierarchy. However, Qin does teach text-based prompts ([0019] “In on implementation, a query may be used to retrieve pieces of augmentation information that may be included in a prompt to the LLM.”). Kumar and Qin are considered analogous art to the claimed invention because they are in the same field of endeavor being natural language processing. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to substitute the natural language processing of Kumar with the large language model of Qin. One would want to use a large language model to improve accuracy (Qin [0028]) Homsi teaches: the classification hierarchy comprising a plurality of nodes that represent respective organizational units or roles of the organization, (Col 7 lines 14-24 “Organizational Unit (OU): Any logical grouping of resources. An OU represents the hierarchical organizational model for an enterprise. The OU usually reflects the current structure of cost centers/profit centers. Organizational Hierarchy: Given a finite set of OU's E, E is an organization hierarchy if the function hierarchically reports to is a tree dimension of E. This particular tree dimension is by convention the hierarchical tree dimension of E. The topmost OU is the root element of an organization hierarchy and the parent element of another OU (Parent OU).”) Kumar, Qin and Homsi are considered analogous art to the claimed invention because they are in the same field of endeavor being natural language processing/hierarchies. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to substitute the natural language processing of Kumar with the large language model of Qin with the workflow activities being organized into organizational hierarchy with organizational units of Homsi. One would want to do this to have a workflow model to process different tasks and who performs these tasks. (Homsi background). Pachet teaches: and the text-based prompt including at least one of one or more positive mapping examples or one or more negative mapping examples, wherein a positive mapping example includes one or more programmatic objects of a first example plan of action that are designated as being correctly mapped to a corresponding node in a first example classification hierarchy, and wherein a negative mapping example includes one or more programmatic objects of a second example plan of action that are designated as being incorrectly mapped to a corresponding node in a second example classification hierarchy. ([0040] “In particular, when developing a tag model for a tag T, the training set contains "positive" and "negative" examples, that is, prototypes which conform to tag T (notably, prototypes to which the user has assigned tag T) and prototypes which do not conform to tag T.”) Kumar, Qin, Homsi and Pachet are considered analogous art to the claimed invention because they are in the same field of endeavor being machine learning. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to substitute the natural language processing of Kumar with the large language model of Qin and use the positive and negative training examples of Pachet. A person of ordinary skill in the art would have wanted to do this to provide classification examples to the language model to help train the models and thus improve accuracy. (Pachet [0072]). Claims 2-3 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Kumar in view of Qin and Pachet (US 2010/0299328 A1). Regarding claim 2, Kumar in view of Qin teaches claim 1 as outlined above. Neither of them teach: the text- based prompt includes at least one of one or more first positive mapping examples and one or more negative mapping examples. However, Pachet does: the text- based prompt includes at least one of one or more first positive mapping examples and one or more negative mapping examples. ([0040] “In particular, when developing a tag model for a tag T, the training set contains "positive" and "negative" examples, that is, prototypes which conform to tag T (notably, prototypes to which the user has assigned tag T) and prototypes which do not conform to tag T.”) Kumar, Qin, and Pachet are considered analogous art to the claimed invention because they are in the same field of endeavor being machine learning. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to substitute the natural language processing of Kumar with the large language model of Qin and use training practices of Pachet. One would want to do this for better accuracy (Pachet [0072]). Regarding claim 3, Kumar in view of Qin teaches claim 1 as outlined above. Neither of them teach: a positive mapping example includes one or more programmatic objects of a first example plan of action that are designated as being correctly mapped to a corresponding node in a first example classification hierarchy, and wherein a negative mapping example includes one or more programmatic objects of a second example plan of action that are designated as being incorrectly mapped to a corresponding node in a second example classification hierarchy. However, Pachet does: a positive mapping example includes one or more programmatic objects of a first example plan of action that are designated as being correctly mapped to a corresponding node in a first example classification hierarchy, and wherein a negative mapping example includes one or more programmatic objects of a second example plan of action that are designated as being incorrectly mapped to a corresponding node in a second example classification hierarchy. ([0040] “In particular, when developing a tag model for a tag T, the training set contains "positive" and "negative" examples, that is, prototypes which conform to tag T (notably, prototypes to which the user has assigned tag T) and prototypes which do not conform to tag T.”) Kumar, Qin, and Pachet are considered analogous art to the claimed invention because they are in the same field of endeavor being machine learning. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to substitute the natural language processing of Kumar with the large language model of Qin and use training practices of Pachet. One would want to do this for better accuracy (Pachet [0072]). Regarding claim 15, Kumar in view of Qin teaches claim 14 as outlined above. Neither of them teach: the text-based prompt includes at least one of one or more first positive mapping examples and one or more negative mapping examples. However, Pachet does: the text-based prompt includes at least one of one or more first positive mapping examples and one or more negative mapping examples. ([0040] “In particular, when developing a tag model for a tag T, the training set contains "positive" and "negative" examples, that is, prototypes which conform to tag T (notably, prototypes to which the user has assigned tag T) and prototypes which do not conform to tag T.”) Kumar, Qin, and Pachet are considered analogous art to the claimed invention because they are in the same field of endeavor being machine learning. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to substitute the natural language processing of Kumar with the large language model of Qin and use training practices of Pachet. One would want to do this for better accuracy (Pachet [0072]). Conclusion 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 DANIEL P GRUSZKA whose telephone number is (571)272-5259. The examiner can normally be reached M-F 9:00 AM - 6:00 PM ET. 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, Li Zhen can be reached at (571) 272-3768. 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. /DANIEL GRUSZKA/Examiner, Art Unit 2121 /Li B. Zhen/Supervisory Patent Examiner, Art Unit 2121
Read full office action

Prosecution Timeline

Sep 01, 2023
Application Filed
Apr 21, 2026
Non-Final Rejection mailed — §103
Aug 04, 2026
Response Filed
Oct 21, 2026
Final Rejection mailed — §103 (current)

Strategy Recommendation AI-generated — please review before filing

Get a prosecution strategy drawn from examiner precedents, rejection analysis, and claim mapping.
Typically takes 5-10 seconds — AI-generated, attorney review required before filing

Prosecution Projections

3-4
Expected OA Rounds
33%
Grant Probability
99%
With Interview (+66.7%)
4y 2m (~1y 0m remaining)
Median Time to Grant
Moderate
PTA Risk
Based on 6 resolved cases by this examiner. Grant probability derived from career allowance rate.

Sign in with your work email

Enter your email to receive a magic link. No password needed.

Personal email addresses (Gmail, Yahoo, etc.) are not accepted.

Free tier: 3 strategy analyses per month