DETAILED ACTION
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 .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.
Applicant's submission filed on 01/02/2026 has been entered.
Status of the Claims
Claims 1, 4-5, 9, 11, 15 and 18 have been amended. Claims 10 and 17 have been canceled. Claims 1-9, 11-16, 18-20 are pending.
Claim Rejections - 35 USC § 103
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 1-9, 11-16, 18-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Vignesh THIRUKAZHUKUNDRAM SUBRAHMANIAM et al. (US 20250272505), hereafter Vignesh in view of Jones et al. (US 12,321,791) and in further view of Hoang et al. (US 20250068627).
Regarding claim 1, Vignesh teaches a system comprising: a memory storing program code: and one or more processing units to execute the program code to cause the system to:
receive a natural language query ([0023], [0037]);
generate a Structured Query Language (SQL) query based on the received natural language query ([0026], [0039]) and using a large language model (LLM) trained with an domain-specific data dictionary ([0031] “LLM uses a metadata dictionary to generate the sequence of operations”, [0038], [0042] “database may be domain-specific”),
the generated SQL query including an endpoint and one or more fields based on the application-specific data dictionary trained LLM ([0040]);
determine, via a
invoke the
determine the natural language query is answerable with the received response ([0042] “determine one or more available data for one or more parameters”, [0045] “operations component may be constrained by the available data”; “operation related to invoice amounts may not be used when the dataset does not include invoice data”, [0046]);
generate a natural language response from the response to the SQL query in a case the natural language query is determined answerable ([0048] “generate a natural language response, which is provided to user as part of answers”, [0077], [0086]); and transmit the natural language response to an entity ([0068]).
Vignesh does not explicitly teach no contract-based Application Programming Interface (API). Instead, Vignesh teaches - a REST API which is a flexible, loosely coupled contract based on HTTP standards, documentation, which allows for greater adaptability. However, it is only obvious variation to use a no contract-based API instead of flexible, loosely coupled contract API per intended design choice and a desired adaptation.
Still, Vignesh does not explicitly teach, however Jones discloses determine, via a no contract-based Application Programming Interface (API) (C6L44 -50 “API is not specified using predefined or preprogramed software”; “the API may be a codeless API that is flexibly adapted”), the endpoint and an API call from data included in the generated SQL query (C8L48-53, C28L65-66, C30L44-54); invoke the no contract-based API (C36L32-37); receive a response to the SQL query from a data source via the no contract-based API (C28L65-67 – C29L1-5, C30L44-54, C35L1-22).
It would have been obvious to one of ordinary skill in the art at the time of invention to modify the teachings of Vignesh to include no contract-based Application Programming Interface as disclosed by Jones. Doing so would allow data in database to be efficiently, flexibly and robustly accessed in real time in a wide variety of applications (Jones C42L43-45).
Vignesh does not explicitly teach application-specific data dictionary. Instead, Vignesh teaches LLM is based on a domain-specific metadata dictionary, which is which is and answering questions for real-world domains with large datasets. However, applicant’s specification discloses – “data dictionary may be application-specific and generated for each application with data from an application-specific database (e.g., application database)” [0046]. Thus, it is reasonable to conclude that application-specific data corresponds to application-specific database and is a an obvious variation for the domain-specific tabular dataset disclosed by Vignesh. The Vignesh domain-specific metadata dictionary com[rises - information associated with columns, such as a column name, data type, and description, column aliases (i.e., possible alternative names for the column, minimum and maximum values for date and numerical columns, distinct M values for categorical columns [0042]. The applicant’s specification similarly discloses – “The data dictionary 212 may include a data structure including, but not limited to, tables, entities, fields, etc. for a given application. Pursuant to embodiments, the data dictionary 212 may be application-specific and generated for each application with data from an application-specific database” [0046]. Thus, the domain-specific metadata dictionary, disclosed by Vignesh, is an obvious variation of the limitation application-specific data dictionary.
However, to merely obviate such reasoning, Hoang discloses application-specific data [0188]. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the teachings of Vignesh to include application-specific data as disclosed by Hoang. Doing so would improve conversational experience (Hoang [0003]).
Claims 9 and 15 recite substantially the same limitations as claim 1, and is rejected for substantially the same reasons.
Regarding claim 2, Vignesh as modified teaches the system of claim 1, further comprising program code to cause the system to:
extract one or more intents from the natural language query (Hoang [0052], [0080], [0104]); and
transmit the extracted one or more intents to a text generation tool for generation of the SQL query (Hoang [0122], [0126], [0165]).
Regarding claims 3 and 16, Vignesh as modified teaches the system and the media wherein the one or more intents are extracted based on action verbs in the natural language query (Hoang [0103]-[0104], [0106]).
Regarding claim 4, Vignesh as modified teaches the system of claim 2, wherein the text generation tool is the large language model (LLM) (Vignesh [0030], Hoang [0131]).
Regarding claims 5, 11 and 18, Vignesh as modified teaches the system, the method and the media, wherein the LLM is trained with two or more data dictionaries (Vignesh [0031], [0038], [0042], Hoang [0085], [0103] [0113], [0130], [0133]).
Regarding claims 6, 12 and 19, Vignesh as modified teaches the system, the method and the media, wherein each data dictionary is generated for a respective application-specific database (Vignesh [0031], [0038], [0042], Hoang [0187] “data repositories 1414, 1416 may be used to store information such as information related to chatbot”, [0225] “communication subsystem may be used to communicate with a chatbot system selected for an application”, Jones C31L4-6, C20L15-31).
Regarding claims 7 and 13, Vignesh as modified teaches the system and the method, wherein the SQL query includes a data source identifier (Hoang [0158], [0167] Table 5, [0187], [0208], Jones C27L56-67, C30L1-8, C44L44-50, C46L10-13).
Regarding claim 8, Vignesh as modified teaches the system of claim 1, wherein the API call provides security to the data source (Hoang [0203], [0210], Jones C9L13-14, C16L22-23, C17L66-67, C19L1-5).
Regarding claims 14 and 20, Vignesh as modified teaches the method and the media, wherein the natural language response is transmitted as a text response or a voice response (Vignesh [0035]- [0036], Hoang [0056], [0064], Jones C31L7-8)).
Claims 5-6, 11-12, 18-19 is/are additionally rejected under 35 U.S.C. 103 as being unpatentable over Vignesh as modified and in further view of Guan et al. (US 20240427742).
Regarding claims 5, 11 and 18, if Vignesh as modified does not explicitly teach, however Guan discloses the system, the method and the media, wherein the LLM is trained with two or more data dictionaries (F1:1010, [0028], [0046] [0113], [0130], [0133]).
It would have been obvious to one of ordinary skill in the art at the time of invention to modify the teachings of Vignesh as modified to include LLM is trained with one or more data dictionaries as disclosed by Guan. Doing so provides more prevalent, efficient, and accurate in building conversational outputs (Guan [0002]).
Regarding claims 6, 12 and 19, Vignesh as modified teaches the system, the method and the media, wherein each data dictionary is generated for a respective application-specific database (Guan [0189], [0196]).
Claims 7-8, 13 is/are additionally rejected under 35 U.S.C. 103 as being unpatentable over Vignesh as modified and in further view of Zhao et al. (US 20250077511).
Regarding claims 7 and 13, if Vignesh as modified does not explicitly teach, however Zhao discloses the system and the method, wherein the SQL query includes a data source identifier ([0022], [0167] Table 5, [0187], [0208]).
It would have been obvious to one of ordinary skill in the art at the time of invention to modify the teachings of Vignesh as modified to include data source identifier as disclosed by Zhao. Doing so providing tenants with a robust system for retrieving information (Zhao [0012]).
Regarding claim 8, if Vignesh as modified does not explicitly teach, however Zhao discloses the system of claim 1, wherein the API call provides security to the data source ([0015], [0029], [0036], [0038]).
It would have been obvious to one of ordinary skill in the art at the time of invention to modify the teachings of Vignesh as modified to include API call that provides security as disclosed by Zhao. Doing so providing tenants with information security across a variety of knowledge sources from a single interface via natural language interactions (Zhao [0012]).
◊ Claims 1, 9 and 15 is/are alternatively rejected under 35 U.S.C. 103 as being unpatentable over Vignesh THIRUKAZHUKUNDRAM SUBRAHMANIAM et al. (US 20250272505), hereafter Vignesh in view of Jones et al. (US 12,321,791) and in further view of Guan et al. (US 20240427742).
Regarding claim 1, Vignesh teaches a system comprising: a memory storing program code: and one or more processing units to execute the program code to cause the system to:
receive a natural language query ([0023], [0037]);
generate a Structured Query Language (SQL) query based on the received natural language query ([0026], [0039]) and using a large language model (LLM) trained with an domain-specific data dictionary ([0031] “LLM uses a metadata dictionary to generate the sequence of operations”, [0038], [0042] “database may be domain-specific”),
the generated SQL query including an endpoint and one or more fields based on the application-specific data dictionary trained LLM ([0040]);
determine, via a
invoke the
determine the natural language query is answerable with the received response ([0042] “determine one or more available data for one or more parameters”, [0045] “operations component may be constrained by the available data”; “operation related to invoice amounts may not be used when the dataset does not include invoice data”, [0046]);
generate a natural language response from the response to the SQL query in a case the natural language query is determined answerable ([0048] “generate a natural language response, which is provided to user as part of answers”, [0077], [0086]); and transmit the natural language response to an entity ([0068]).
Vignesh does not explicitly teach no contract-based Application Programming Interface (API). Instead, Vignesh teaches - a REST API which is a flexible, loosely coupled contract based on HTTP standards, documentation, which allows for greater adaptability. However, it is only obvious variation to use a no contract-based API instead of flexible, loosely coupled contract API per intended design choice and a desired adaptation.
Still, Vignesh does not explicitly teach, however Jones discloses determine, via a no contract-based Application Programming Interface (API) (C6L44 -50 “API is not specified using predefined or preprogramed software”; “the API may be a codeless API that is flexibly adapted”), the endpoint and an API call from data included in the generated SQL query (C8L48-53, C28L65-66, C30L44-54); invoke the no contract-based API (C36L32-37); receive a response to the SQL query from a data source via the no contract-based API (C28L65-67 – C29L1-5, C30L44-54, C35L1-22).
It would have been obvious to one of ordinary skill in the art at the time of invention to modify the teachings of Vignesh to include no contract-based Application Programming Interface as disclosed by Jones. Doing so would allow data in database to be efficiently, flexibly and robustly accessed in real time in a wide variety of applications (Jones C42L43-45).
Vignesh does not explicitly teach application-specific data dictionary. Instead, Vignesh teaches LLM is based on a domain-specific metadata dictionary, which is which is and answering questions for real-world domains with large datasets. However, applicant’s specification discloses – “data dictionary may be application-specific and generated for each application with data from an application-specific database (e.g., application database)” [0046]. Thus, it is reasonable to conclude that application-specific data corresponds to application-specific database and is a an obvious variation for the domain-specific tabular dataset disclosed by Vignesh. The Vignesh domain-specific metadata dictionary com[rises - information associated with columns, such as a column name, data type, and description, column aliases (i.e., possible alternative names for the column, minimum and maximum values for date and numerical columns, distinct M values for categorical columns [0042]. The applicant’s specification similarly discloses – “The data dictionary 212 may include a data structure including, but not limited to, tables, entities, fields, etc. for a given application. Pursuant to embodiments, the data dictionary 212 may be application-specific and generated for each application with data from an application-specific database” [0046]. Thus, the domain-specific metadata dictionary, disclosed by Vignesh, is an obvious variation of the limitation application-specific data dictionary.
However, to merely obviate such reasoning, Guan discloses using a large language model (LLM) trained with an application-specific data dictionary ([0027]-[0028], [0031]-[0032]), the generated SQL query including an endpoint and one or more fields ([0042], [0044], [0069], [0179]) based on the application-specific data dictionary trained LLM ([0027], [0029], [0046]).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the teachings of Vignesh to include application-specific data as disclosed by Guan. Doing so provides more prevalent, efficient, and accurate in building conversational outputs (Guan [0002]).
Claims 9 and 15 recite substantially the same limitations as claim 1, and is rejected for substantially the same reasons.
Response to Arguments
Applicant’s arguments, filed 01/02/2026, in regard to the presently amended claims, are addressed in the updated rejections to the claims above.
Please note an alternative rejection to the claims 1, 9 and 15 immediately above.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure is indicated on PTO-892.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to POLINA G PEACH whose telephone number is (571)270-7646. The examiner can normally be reached Monday-Friday, 9:30 - 5:30.
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, Aleksandr Kerzhner can be reached at 571-270-1760. 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.
/POLINA G PEACH/Primary Examiner, Art Unit 2165 April 5, 2026
Prosecution Insights