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 .
DETAILED ACTION
Claim Rejections - 35 USC § 101
Applicants arguments in concert with amendments to claims 1-20 as filed 3/9/26 suffice to obviate the rejection of the claims under 35 U.S.C. 101 as the claims are considered to resolve a practical application resulting in the improvement of visualizations.
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 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 of this title, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1-20 rejected under 35 U.S.C. 103 as being unpatentable over Zhao: 20250077511 hereinafter Zha further in view of Dibia: “LIDA: A Tool for Automatic Generation of Grammar-Agnostic Visualizations and Infographics using Large Language Models,” (copy provided by Examiner; copyright 6/23; and hereinafter Dib) and further in view of Sidhu: 20250335799 hereinafter Sid.
Regarding claim 1
Zha teaches:
A system for causing a visualization of data responsive to a one or more data requests to be provided for display based on large language model (LLM) generated code (Zha: Abstract; ¶ 12-16, 29, 33; Fig 1, 2: chatbot operative of an LLM to generate SQL queries based on user input of a natural language query), the system comprising: one or more memories; and one or more processors, communicatively coupled to the one or more memories (Zha: ¶ 58; Fig 4: such as operative upon, in concert with, etc. the system of figure 4), configured to:
generate, in response to a data request (Zha: ¶ 35; Fig 3A: such as in response to obtention of a user natural language input), an LLM-generated query associated with retrieving data wherein the LLM-generated query is generated by an LLM that is configured based on metadata associated with a plurality of datasets, and wherein the LLM- generated query comprises executable code (Zha: ¶ 12-16, 20, 29, 33-48; Fig 3A, 3B: a query converter for deriving SQL queries from the natural language input said query operative to query one more databases for results and return a summary thereof to a user such as in a conversation window; said query operative in concert with database metadata, schema, etc.; said query comprising SQL for searching the database);
execute, in response to the data request, the LLM-generated query to retrieve the data (Zha: ¶ 12-16, 34-38; Fig 3A; system determines parameters for a query and executes query upon the one or more datasets);
generate, in response to the data request (Zha: ¶ 35; Fig 3A: such as in response to obtention of a user natural language input), LLM-generated code associated with the data for display (Zha: ¶ 12-16, 20, 36-48; Fig 3A, 3B: a query converter for deriving SQL queries from the natural language input; said query operative to query a database for results and return a summary thereof to a user such as in a conversation window); and
provide, in response to the data request, visualization of the data based on the LLM- generated code (Zha: ¶ 14-16, 34-45; Fig 3A, 3B: system provides query results to the user such as in the form of a formatted chart, table, etc.).
Zha is not explicit that the query is configured based on metadata nor configuring processing by an LLM based on metadata associated with a plurality of datasets without accessing the plurality of datasets; nor the generation of a response to the data request based on additionally receipt of input identifying a type of visualization thereby providing the returned data to the user in concert with the visualization type.
In a related field of endeavor Dib teaches a system and method for causing a visualization of data responsive to a one or more data requests to be provided for display based on large language model (LLM) generated code (Dib: Abstract; § 1, 3.2, 3.2.1, 3.2.2, 3.3; Fig 1: a multi stage LLM pipeline comprising a visualization stage wherein the visualization stage receives input enumerating visualization goals for determined data and specifies an output based thereon such as in executable code; and a grapher stage which generates a visualization based thereon), configured to:
generate, in response to a data request, an LLM-generated query associated with retrieving data wherein the LLM-generated query is generated by an LLM that is configured based on metadata associated with a plurality of datasets without accessing the plurality of datasets, and wherein the LLM- generated query comprises executable code (Dib: § 3.1-3.4; Fig 1, 3-5: summary stage receives an input dataset, such as part of a user request for a visualization and determines and outputs metadata and metadata summaries based thereon which is passed to the explorer stage which displays a summary of the dataset and goals related thereto in a manner sufficient to assist a user in generating a prompt in the form of a determined visualization goal to provide to the visualization stage which generates executable code for the grapher stage; the summary stage does not access the data within the dataset, merely the metadata thereof);
generate, in response to the data request and in response to receiving input identifying a type of visualization, LLM-generated code associated with providing the visualization of the data for display (Dib: § 3.1-3.4; Fig 1, 3-5: visualization stage outputs code based on the input dataset and the input user provided visualization goal); and
provide, in response to the data request, the visualization of the data based on the LLM- generated code (Dib: § 3.1-3.4; Fig 1, 3-5: grapher stage in receipt of visualization code generates and provides a visualization based thereon).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the instant application to utilize the Dib taught LLM driven visualization pipeline to enhance the visualization capabilities as taught or suggested by Zha for at least the purpose of allow user selection of visualizations based on natural language requests therefor; one of ordinary skill in the art would have expected only predictable results therefrom.
Zha in view of Sai does not explicitly discuss the recited the LLM explicitly operating without accessing the plurality of datasets; however Examiner considers such implementation of learning operations to be well known.
As evidence consider Sid which teaches techniques for operating and improving artificial intelligence models (Sid: Abstract); said models including large language models ((LLM); Sid: ¶ 3, 18: artificial intelligence models comprise an LLM); wherein the model operates without receiving explicit data and receives only metadata descriptors of data (Sid: ¶ 44, 72: model receives only metadata and operates without receipt of underlying data; see also provided portions of provisional application). It would have been obvious to one of ordinary skill in the art before the effective filing date of the instant application to provide visualizations of the Zha in view of Sai data, requests for data, query of data, etc. by provision of only metadata, or data values without receiving the underlying data, such as from a user, client, or remote dataset, such as in the manner discussed by Sid and for at least the purpose of enhancing privacy of data, a user, a client or edge device, etc. while still providing analysis and insight gleaned from or based on the data and said data analysis, insight, etc. relevant to a query or request upon the underlying data; one of ordinary skill in the art would have expected only predictable results therefrom. Further Sid is considered exemplary of well-known operations of LLM’s, learning models, outputs thereof, etc. based on only metadata, indeed this is the basis of a federated learning domain (please see additionally: Brende: 20250293998—LLM system operative to only transfer and operate upon metadata, models thereof without actual data for privacy reasons; Sommers: 20240378125—system provides learning based on only parameters and metadata associated with a model; Arasachetty: 20190220550—a federated learning system where a centralized learning model receives only metadata, portions thereof from a plurality of clients; Bhowmick: 11989634—adapting a remote model based only upon metadata received from a plurality of clients).
Regarding claim 2
Zha in view of Dib in view of Sid teaches or suggests:
The system of claim 1, wherein the one or more processors are further configured to:
obtain information that identifies one or more datasets of the plurality of datasets, the one or more datasets being identified by the LLM and based on the data request (Zha: ¶ 34-36; Fig 3A, 3B: chatbot in concert with LLM determines a knowledge source by reifying query terms relevant to the user query); (Dib: § 3.1-3.4; Fig 1, 3-5: explorer stage enumerates dataset metadata based goals, displays same); (Sid: ¶ 61, 62: chatbot identifies and displays relevant data). The claim is considered obvious over Zha as modified by Dib and Sid as addressed in the base claim as it would have been obvious to apply the further teaching of Zha, Dib, and/or Sid to the modified device of Zha, Dib, and Sid; one of ordinary skill in the art would have expected only predictable results therefrom.
Regarding claim 3
Zha in view of Dib in view of Sid teaches or suggests:
The system of claim 2, wherein the one or more processors are further configured to receive user input indicating at least one selected dataset of the one or more datasets, wherein the LLM-generated query is a query associated with the at least one selected dataset (Zha: ¶ 12-16, 34-45: query directed to particular datasets); (Dib: § 3.1-3.4; Fig 1, 3-5: summarizer stage receives an input dataset; determines and outputs metadata and metadata summaries based thereon to the explorer stage); (Sid: ¶ 61-63: chatbot identifies and displays relevant data; such as indicated in a conversation, conversation history, etc.). The claim is considered obvious over Zha as modified by Dib and Sid as addressed in the base claim as it would have been obvious to apply the further teaching of Zha, Dib, and/or Sid to the modified device of Zha, Dib, and Sid; one of ordinary skill in the art would have expected only predictable results therefrom.
Regarding claim 4
Zha in view of Dib in view of Sid teaches or suggests:
The system of claim 1, wherein the one or more processors are further configured to determine that the user is authorized to access a dataset of plurality of datasets that includes the data responsive to the data request (Zha: ¶ 23: system ensures information security, does not share protected information); (Sid: ¶ 44: system applies security policies). Examiner has taken official notice which Applicant has failed to timely and explicitly traverse and it is thus accepted as Admitted Prior Art (APA: please see MPEP 2144.03) that requiring authorization to access particular databases would have comprised an obvious inclusion for at least the purpose of maintaining the integrity of a database; one of ordinary skill in the art would have expected only predictable results therefrom. The claim is thus considered obvious over Zha as modified by Dib and Sid as addressed in the base claim as it would have been obvious to apply the further teaching of Zha, Dib, and/or Sid to the modified device of Zha, Dib, and Sid; one of ordinary skill in the art would have expected only predictable results therefrom.
Regarding claim 5
Zha in view of Dib in view of Sid teaches or suggests:
The system of claim 1, wherein the one or more processors, to cause the visualization to be provided for display to the user, are configured to provide the LLM-generated code for execution by a user device associated with the user (Zha: ¶ 14-18; Figs 103: such as by iterating over tabular data mapped to particular terms); (Dib: § 1, 3.1-3.4; Fig 1, 3-5:system treats visualizations as code which is exported by the visualization stage and executed by the grapher stage). The claim is thus considered obvious over Zha as modified by Dib and Sid as addressed in the base claim as it would have been obvious to apply the further teaching of Zha, Dib, and/or Sid to the modified device of Zha, Dib, and Sid; one of ordinary skill in the art would have expected only predictable results therefrom.
Regarding claim 6
Zha in view of Dib in view of Sid teaches or suggests:
The system of claim 1, wherein the one or more processors, to cause the visualization to be provided for display to the user, are configured to: generate a static visualization based on the LLM-generated code, and provide information associated with the static visualization to a user device associated with the user (Zha: ¶ 14-18, 47; Figs 1-3: static results such as those depicted at 108 and/or charts, tables etc. provided, displayed, etc. to a user which may be in the form of a formatted table or downloadable file); (Dib: § 3.1-3.4; Fig 2: system outputs stylized visualizations). The claim is thus considered obvious over Zha as modified by Dib and Sid as addressed in the base claim as it would have been obvious to apply the further teaching of Zha, Dib, and/or Sid to the modified device of Zha, Dib, and Sid; one of ordinary skill in the art would have expected only predictable results therefrom.
Regarding claim 7
Zha in view of Dib in view of Sid teaches or suggests:
The system of claim 1, wherein the one or more processors, to obtain the LLM-generated query, are configured to:
provide the data request as an input to the LLM that is configured based on the metadata associated with the plurality of datasets (Zha: Abstract: system queries datasets based on metadata therein, retrieves results thereof based thereon); and receive the LLM-generated query as an output of the LLM (Zha: ¶ 12-15, 34-36; Fig 3A: system determines a particular knowledge source based on parameters of the language query; such as by determination of terms therein relevant to particular knowledge sources, databases, etc. and by parsing the query into metadata expected by the knowledge sources, such as sql which is received by processors of the system operable to provide the query to relevant data sources; said data sources receiving the query and returning a response thereto). The claim is considered obvious over Zha as modified by Dib and Sid as addressed in the base claim as it would have been obvious to apply the further teaching of Zha, Dib, and/or Sid to the modified device of Zha, Dib, and Sid; one of ordinary skill in the art would have expected only predictable results therefrom.
Regarding claim 8
Zha teaches:
A method for causing a visualization of data responsive to a data request to be provided for display, comprising:
generating, by a system, a query associated with retrieving data responsive to a data request, wherein the query is a first output of a large language model (LLM) that is trained based on metadata associated with a plurality of datasets and wherein the query comprises executable code (Zha: ¶ 4, 12-16, 20, 29, 33-48; Fig 3A, 3B: a query converter for deriving SQL queries from the natural language input said query operative to query one or more databases for results and return a summary thereof to a user such as in a conversation window; said query operative in concert with database metadata, schema, etc.; said query comprising SQL for searching the database; said model trained and refined with respect to generating SQL queries of a database);
executing, by the system the query to retrieve the data responsive to the data request, the data being retrieved from at least one dataset of the one or more of datasets; (Zha: ¶ 14-16, 34-38; Fig 3A; system determines parameters for a query and executes query upon one or more datasets);
generating, by the system (Zha: ¶ 35; Fig 3A: such as in response to obtention of a user natural language input), code associated with providing the visualization of the data responsive to the data request for display, wherein the code is a second output of the LLM (Zha: ¶ 12-16, 20, 36-48; Fig 3A, 3B: a query converter for deriving SQL queries from the natural language input; said query operative to query a database for results and return a summary thereof to a user such as in a conversation window); and
providing , by the system and based on the code, the visualization of the data responsive to the data request to be provided for display (Zha: ¶ 14-16, 34-45; Fig 3A, 3B: system provides query results to the user such as in the form of a formatted chart, table, etc.).
Zha is not explicit that the query is configured based on metadata nor configuring processing by an LLM based on metadata associated with a plurality of datasets without accessing the plurality of datasets; nor the generation of a response to the data request based on additionally receipt of input identifying a type of visualization thereby providing the returned data to the user in concert with the visualization type.
In a related field of endeavor Dib teaches a system and method for causing a visualization of data responsive to a one or more data requests to be provided for display based on large language model (LLM) generated code (Dib: Abstract; § 1, 3.2, 3.2.1, 3.2.2, 3.3; Fig 1: a multi stage LLM pipeline comprising a visualization stage wherein the visualization stage receives input enumerating visualization goals for determined data and specifies an output based thereon such as in executable code; and a grapher stage which generates a visualization based thereon), configured to:
generate, in response to a data request, an LLM-generated query associated with retrieving data wherein the LLM-generated query is generated by an LLM that is configured based on metadata associated with a plurality of datasets without accessing the plurality of datasets, and wherein the LLM- generated query comprises executable code(Dib: § 3.1-3.4; Fig 1, 3-5: summary stage receives an input dataset, such as part of a user request for a visualization and determines and outputs metadata and metadata summaries based thereon which is passed to the explorer stage which displays a summary of the dataset and goals related thereto in a manner sufficient to assist a user in generating a prompt in the form of a determined visualization goal to provide to the visualization stage which generates executable code for the grapher stage; the summary stage does not access the data within the dataset, merely the metadata thereof);
generate, in response to the data request and in response to receiving input identifying a type of visualization, LLM-generated code associated with providing the visualization of the data for display (Dib: § 3.1-3.4; Fig 1, 3-5: visualization stage outputs code based on the input dataset and the input user provided visualization goal); and
provide, in response to the data request, the visualization of the data based on the LLM- generated code (Dib: § 3.1-3.4; Fig 1, 3-5: grapher stage in receipt of visualization code generates and provides a visualization based thereon).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the instant application to utilize the Dib taught LLM driven visualization pipeline to enhance the visualization capabilities as taught or suggested by Zha for at least the purpose of allow user selection of visualizations based on natural language requests therefor; one of ordinary skill in the art would have expected only predictable results therefrom.
Zha in view of Sai does not explicitly discuss the recited the LLM explicitly operating without accessing the plurality of datasets; however Examiner considers such implementation of learning operations to be well known.
As evidence consider Sid which teaches techniques for operating and improving artificial intelligence models (Sid: Abstract); said models including large language models ((LLM); Sid: ¶ 3, 18: artificial intelligence models comprise an LLM); wherein the model operates without receiving explicit data and receives only metadata descriptors of data (Sid: ¶ 44, 72: model receives only metadata and operates without receipt of underlying data; see also provided portions of provisional application). It would have been obvious to one of ordinary skill in the art before the effective filing date of the instant application to provide visualizations of the Zha in view of Sai data, requests for data, query of data, etc. by provision of only metadata, or data values without receiving the underlying data, such as from a user, client, or remote dataset, such as in the manner discussed by Sid and for at least the purpose of enhancing privacy of data, a user, a client or edge device, etc. while still providing analysis and insight gleaned from or based on the data and said data analysis, insight, etc. relevant to a query or request upon the underlying data; one of ordinary skill in the art would have expected only predictable results therefrom. Further Sid is considered exemplary of well-known operations of LLM’s, learning models, outputs thereof, etc. based on only metadata, indeed this is the basis of a federated learning domain (please see additionally: Brende: 20250293998—LLM system operative to only transfer and operate upon metadata, models thereof without actual data for privacy reasons; Sommers: 20240378125—system provides learning based on only parameters and metadata associated with a model; Arasachetty: 20190220550—a federated learning system where a centralized learning model receives only metadata, portions thereof from a plurality of clients; Bhowmick: 11989634—adapting a remote model based only upon metadata received from a plurality of clients).
Regarding independent claim 15—the claim recites substantially similar subject matter to that of claims 1, 8 and is similarly rejected.
Regarding claim 9, 16—the claims recite substantially similar subject matter to that of claim 2 and are similarly rejected.
Regarding claim 10, 17—the claims recite substantially similar subject matter to that of claim 3 and are similarly rejected.
Regarding claim 11, 18—the claims recite substantially similar subject matter to that of claim 4 and are similarly rejected.
Regarding claim 12, 19—the claims recite substantially similar subject matter to that of claim 5 and are similarly rejected.
Regarding claim 13, 20—the claims recite substantially similar subject matter to that of claim 6 and are similarly rejected.
Regarding claim 14—the claim recites substantially similar subject matter to that of claim 7 and is similarly rejected.
Response to Arguments
Applicant’s arguments in concert with amendments to the independent claims, see Remarks and Claims, filed 3/9/26, with respect to the rejection(s) of claim(s) 1-20 under 35 USC 103 over Zhao, Sainani, and Sidhu have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made in view of Zhao, Dibia, and Sidhu.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
20240362405 – prompt includes a first parameter specifying visualization type; second parameter data specifying a data request.
20250094703 – LLM trained on visualization types.
20240338232 – prompt to an LLM specifies visualization type and data request.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to PAUL C MCCORD whose telephone number is (571)270-3701. The examiner can normally be reached 730-630 M-F.
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, CAROLYN EDWARDS can be reached at (571) 270-7136. 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.
/PAUL C MCCORD/ Primary Examiner, Art Unit 2692
Prosecution Insights