Notice of Pre-AIA or AIA Status
1. The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Continued Examination
2. A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 05/26/2026 has been entered.
DETAILED ACTION
3. This Office Action is in response to the filing with the office dated 05/26/2026.
Claims 1, 16 and 20 have been amended. Claims 1, 16 and 20 are independent claims. Claims 1-20 are presented in this office action.
Response to amendment/arguments
4. Applicant’s arguments with respect to the rejection of claims under 35 U.S.C. § 101 as the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more, have been fully considered. However, Examiner respectfully disagrees with the applicant’s argument. See response to arguments section. The rejection has been maintained.
5. Applicant’s arguments with respect to the rejection of claims under 35 U.S.C. § 102 (a)(i) and 103(a) have been fully considered but are moot because the arguments are directed towards amended claims, thus necessitated the new ground of rejection as presented in this Office action. Please see the rejection below.
Response to 101 Rejection
6. Applicant’s arguments on pages 11, 13 regarding 101 rejection recites “the claims do not recite and are therefore also not "directed to" an alleged "mental process." In particular, the claimed embodiments (e.g., executing a query written in a programming language against a database management system) cannot practically be performed in a human mind, and thus do not recite a "mental process" and are not "directed to" an alleged "mental process." “the claims recite a practical application that improves the text-to-SQL conversion accuracy of a language model trained to retrieve data from a database management system, as well as the relevance of data retrieved by the language model” .
Examiner respectfully disagrees with the applicant because, executing a query written in a programming language against a database management system is recited at a high level of generality and do not place meaningful limits on the abstract idea. These limitations are essentially steps of generating and manipulating data at a high level of generality, which can be performed by a person using a computer as a tool using a trained language model to convert the text to SQL and execute the SQL query using the computer as a tool is an insignificant extra-solution activity of a data gathering process. These additional elements amount to nothing more than mere instructions to apply the recited abstract idea on a computer, under MPEP 2106.05(f). The additional elements of “executing the search query” amount to mere data outputting which are insignificant extra-solution activity. Combination of these additional elements is no more than mere instructions to apply the exception using series of steps and outputting the result of the mental process. Accordingly, even in combination, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea.
Additionally, the mere nominal recitation of a generic computer components, or a programmed computer does not take the claim limitation out of the mental processes grouping. These additional elements are no more than mere instructions to apply the exception using series of steps. Accordingly, even in combination, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. The limitations are steps involving processes that can be practically performed by a human with the aid of pen and paper, or as explained above, using a computer as a tool to perform the concept. For example, a person can obtain query logs, extracting plurality of features, associating the features with the structured query language queries, generating a summary, identify from the summary, identify subset of SQL queries that contain the features, wherein the subset of the plurality of features includes features of the plurality of features that are extracted from at least two different structured query language queries of the plurality of structured query language queries; converting each SQL query into a question phrased natural language and storing the question with the SQL as a prompt and executing the SQL query. These limitations do not improve the functioning of a computer, improve the technology, apply the abstract idea to a particular machine, effect a transformation, nor provide meaningful limitations beyond linking the abstract idea to computer technology. They do not recite specific details that amount to significantly more than the abstract idea or providing meaningful limits on the abstract idea. For at least these reasons, claims 1, 16 and 20 are nonstatutory because they are directed to a judicial exception without significantly more.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
7. Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more. Determining whether claims are statutory under 35 U.S.C. 101 involves a two-step analysis. Step 1 requires a determination of whether the claims are directed to the statutory categories of invention. Step 2 requires a determination of whether the claims are directed to a judicial exception without significantly more. Step 2 is divided into two prongs, with the first prong having a part 1 and part 2. See MPEP 2106; See 2019 Revised Patent Subject Matter Eligibility Guidance (2019 PEG).
Pursuant to Step 1, claims 16-19 recite a computer readable medium which are directed to the statutory category of a manufacture. Claim 20 recite a system, which are directed to a machine.
Pursuant to Step 2A, part 1, claims are analyzed to determine whether they are directed to an abstract idea. Under the 2019 PEG, claims are deemed to be directed to an abstract idea if they fall within one of the enumerated categories of (a) mathematical concepts, (b) certain methods of organizing human activity, and (c) mental processes. Here, claims 1, 16 and 20 are directed to an abstract idea categorized under mental processes. Courts consider a mental process if it “can be performed in the human mind, or by a human using a pen and paper.” MPEP 2016(a)(2)(III). Courts also consider a mental process as one that can be performed in the human mind and is merely using a computer as a tool to perform the concept. MPEP 2016(a)(2)(III)(C)(3). Claim 1 recites a mental process because the steps recite the actions of storing and manipulating data but is recited at a high level of generality that merely used computers as a tool to perform the processes. See MPEP 2106(a)(2)(III). For example, claims 1, 16 and 20 recites limitations of “obtaining….”, “extracting queries from the plurality of structured query language queries, wherein the subset of the plurality of features includes features of the plurality of features that are extracted from at least two different structured query language queries of the plurality of structured query language queries”, “associating…”, “generating…”, “identifying…”, converting…”. “executing, by the processing system, the query written in structured query language programming language against the database management system”. The “system”, “database management system”, “processor”, “computer readable medium”, are recited at a high level of generality and do not place meaningful limits on the abstract idea. These limitations are essentially steps of generating and manipulating data at a high level of generality, which can be performed by a person using a computer as a tool.
Pursuant to Step 2A, part 2, claims are analyzed to determine whether the recited abstract idea is integrated into a practical application. In this case, as explained above, claims 1, 16 and 20 merely recite a mental process. These limitations describe “associating…”, “generating…”, :identifying….”, “executing…” while claims 1, 16 and 20 recite additional components in the form of “system”, “database management system”, “processor”, “computer readable medium”, are recited at a high level of generality and do not place meaningful limits on the abstract idea to integrate it into a practical application by providing an improvement to the functioning of a computer or technology, implementing the abstract idea with a particular machine or manufacture that is integral to the claim, effecting a transformation or reduction of a particular article to a different state or thing, nor applying the abstract idea in some meaningful way beyond linking its use to computer technology. See 2019 PEG.
Since claims 1, 16 and 20 are directed to an abstract idea categorized as a mental process and does not integrate the judicial exception into a practical application.
Pursuant to Step 2B, claims are analyzed to determine whether they recite significantly more than the abstract idea. In other words, it is determined whether the claims provide an inventive concept. In this case, claims 1, 16 and 20 do not recite limitations that amount to significantly more than the abstract idea. The limitations are steps involving processes that can be practically performed by a human with the aid of pen and paper, or as explained above, using a computer as a tool to perform the concept. For example, a person can obtain query logs, extracting plurality of features, associating the features with the structured query language queries, generating a summary, identify from the summary, identify subset of SQL queries that contain the features, converting each SQL query into a question phrased natural language using a trained language model to generate a text to SQL query and execute the query. These limitations do not improve the functioning of a computer, improve the technology, apply the abstract idea to a particular machine, effect a transformation, nor provide meaningful limitations beyond linking the abstract idea to computer technology. They do not recite specific details that amount to significantly more than the abstract idea or providing meaningful limits on the abstract idea. For at least these reasons, claims 1, 16 and 20 are nonstatutory because they are directed to a judicial exception without significantly more.
Claim 2, 3, 4 recites “the plurality of features includes …” is insignificant extra-solution activity as mere data gathering and outputting when re-evaluated still does not provide significantly more. Considering the additional elements in combination and the claim as a whole does not change the analysis, and does not amount to significantly more. Thus the claims are abstract. Accordingly, the claim recites an abstract idea.
Claims 5, 6 recites, “at least some of the plurality of features are extracted … is a process, that under broadest reasonable interpretation, covers performance of the limitation in the mind. There is, nothing in the claim element precludes the steps from practically being performed by a human mentally or with pen and paper and likewise do not provide "significantly more" than the abstract idea for similar reasons as the independent claim.
Claims 7, 8, 9 recites, “…a count…” is a process, that under broadest reasonable interpretation, covers performance of the limitation in the mind. There is, nothing in the claim element precludes the steps from practically being performed by a human mentally or with pen and paper and likewise do not provide "significantly more" than the abstract idea for similar reasons as the independent claim.
Claim 10 recites “identifying…” is a process, that under broadest reasonable interpretation, covers performance of the limitation in the mind. There is, nothing in the claim element precludes the steps from practically being performed by a human mentally or with pen and paper and likewise do not provide "significantly more" than the abstract idea for similar reasons as the independent claim.
Claim 11 recites “converting comprises: identifying …,determining….providing….” is a process, that under broadest reasonable interpretation, covers performance of the limitation in the mind. There is, nothing in the claim element precludes the steps from practically being performed by a human mentally or with pen and paper and likewise do not provide "significantly more" than the abstract idea for similar reasons as the independent claim.
Claim 12, 13, 14, 15 recites “rephrasing…”, “question that is capable of being answered by an exact code…”, “determining…” s insignificant extra-solution activity as mere data gathering and outputting when re-evaluated still does not provide significantly more. Considering the additional elements in combination and the claim as a whole does not change the analysis, and does not amount to significantly more. Thus the claims are abstract. Accordingly, the claim recites an abstract idea.
Claim Rejections - 35 U.S.C. § 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, 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.
8. Claims 1-3, 5-8, 10-17 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over WAN; Bo (US 20210158176 A1) in view of Popescu; Octavian (US 20250077517 A1) and in further view of Johnson; Theodore (US 20260050591 A1).
Regarding independent claim 1, WAN; Bo (US 20210158176 A1) teaches, a method comprising: obtaining, by a processing system including at least one processor, a query log from a database management system, wherein the query log contains a plurality of structured query language queries previously executed against a target database (Paragraph [0037] he query logs 110 may include all historical queries, such as SQL (structured query language) queries for the one or more databases of the data center 106);
extracting, by the processing system from the plurality of structured query language queries, a plurality of features; associating, by the processing system, the plurality of features with the plurality of structured query language queries (Paragraphs [0038], [0039] discloses, extracting plurality of features from the query logs);
generating, by the processing system, a summary of the plurality of features (Paragraphs [0039], [0040] discloses, the analytic server 102 may implement a key phrase extraction model that utilizes natural language processing (NLP) to extract key phrases and features of each table's description (i.e., understanding each table's semantic, usage, and relationships to other tables within the data center);
identifying, by the processing system based on the summary, a subset of the plurality of features to use as a basis for extracting queries from the plurality of structured query language queries (Paragraph [0041] The analytic server 102 may also generate a graph database based on the output from the query parser. As discussed above, the query parser may extract the tables and the relationships among the tables based on the historical queries. Also see [0044]);
identifying, by the processing system, a subset of the plurality of structured query language queries that contains the subset of the plurality of features (Paragraphs [0040]- [0042] discloses, identifying the queries that are semantically similar to the user query. Also see [0051]);
WAN et al fails to explicitly teach, wherein the subset of the plurality of features includes features of the plurality of features that are extracted from at least two different structured query language queries of the plurality of structured query language queries; converting, by the processing system using a plurality of language models, each structured query language query of the subset of the plurality of structured query language queries into a question phrased in natural language; prompting, by the processing system, a language model that is trained to perform a text-to-structured query language conversion to generate query written in structured query language programming language, using the question and the each structured query language query together as part of a prompt that improves an accuracy of the text-to-structured query language conversion by the language model; and executing, by the processing system, the query written in structured query language programming language against the database management system.
Popescu; Octavian (US 20250077517 A1) teaches, (Figs. 7, 8, Paragraphs [0112], [0113] discloses, converting structured query language query into a question phrased in natural language);
Therefore it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention, to have modified the teachings of WAN et al by converting, by the processing system using a plurality of language models, each structured query language query of the subset of the plurality of structured query language queries into a question phrased in natural language;, as taught by Popescu et al (Paragraphs [0112], [0113]).
One of the ordinary skill in the art would have been motivated to make this modification, by doing so, the user can review the natural language text in query results 844 to confirm whether that natural language text corresponds to the user query as taught by Popescu et al (Paragraph [0112]).
WAN et al and Popescu et al fails to explicitly teach, wherein the subset of the plurality of features includes features of the plurality of features that are extracted from at least two different structured query language queries of the plurality of structured query language queries; prompting, by the processing system, a language model that is trained to perform a text-to-structured query language conversion to generate query written in structured query language programming language, using the question and the each structured query language query together as part of a prompt that improves an accuracy of the text-to-structured query language conversion by the language model; and executing, by the processing system, the query written in structured query language programming language against the database management system.
Johnson; Theodore (US 20260050591 A1) teaches, wherein the subset of the plurality of features includes features of the plurality of features that are extracted from at least two different structured query language queries of the plurality of structured query language queries (Paragraph [0026] discloses, plurality of features are extracted from plurality of SQL queries/ query logs that were previously executed against one or more tables contained in the database based on frequencies of occurrence of the features);
prompting, by the processing system, a language model that is trained to perform a text-to-structured query language conversion to generate query written in structured query language programming language, and executing, by the processing system, the query written in structured query language programming language against the database management system (Paragraphs [0026]-[0028] discloses, generating a prompt for a new SQL query 116 to be executed against the database using query logs);
using the question and the each structured query language query together as part of a prompt that improves an accuracy of the text-to-structured query language conversion by the language model (Paragraph [0018] both query log analysis and database profiling/ questions may be used together to improve the generation of prompts for text-to-SQL queries).
Note: the limitation “improves an accuracy of the text-to-structured query language conversion by the language model” is an intended use and it does not specify how the accuracy is performed by the model.
Therefore it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention, to have modified the teachings of WAN et al and Popescu et al by providing wherein the subset of the plurality of features includes features of the plurality of features that are extracted from at least two different structured query language queries of the plurality of structured query language queries; prompting, by the processing system, a language model that is trained to perform a text-to-structured query language conversion to generate query written in structured query language programming language, using the question and the each structured query language query together as part of a prompt that improves an accuracy of the text-to-structured query language conversion by the language model; and executing, by the processing system, the query written in structured query language programming language against the database management system, as taught by Johnson et al (Paragraphs [0026]-[0028]).
One of the ordinary skill in the art would have been motivated to make this modification, by doing so, would improve the generation of prompts for text-to-SQL queries. The disclosed approaches may assist an LLM in generating prompts for new SQL queries, even for SQL databases that the LLM has not previously encountered as taught by Johnson et al (Paragraph [0018]).
Regarding dependent claim 2, WAN et al, Popescu et al and Johnson et al teach, the method of claim 1.
WAN et al further teaches, wherein the plurality of features includes at least one of: a set of tables referenced in a query of the plurality of structured query language queries (Paragraph [0030] discloses, set of tables referenced in the historical query logs).
a collection of tables which are joined together in a query of the plurality of structured query language queries, a set of fields referenced by a query of the plurality of structured query language queries, a set of select elements in a query of the plurality of structured query language queries named using an as clause, a set of unnamed select elements in a query of the plurality of structured query language queries, a set of predicates which constrain only one range variable in a query of the plurality of structured query language queries, a set of predicates in an on clause in a query of the plurality of structured query language queries, a set of predicates which constrain multiple range variables in a query of the plurality of structured query language queries, a partitioning field of a window function in a query of the plurality of structured query language queries, an output field of a window function in a query of the plurality of structured query language queries, a set of named subqueries in a query of the plurality of structured query language queries, a set of named and unnamed fields in named subqueries in a query of the plurality of structured query language queries, a group-by field in a query of the plurality of structured query language queries, a normalized indication of a complexity of a query of the plurality of structured query language queries, or a normalized indicator of a shape of a query of the plurality of structured query language queries.
Regarding dependent claim 3, WAN et al, Popescu et al and Johnson et al teach, the method of claim 1.
Wan et al further teaches, wherein the plurality of features includes at least one feature that is not extracted directly from text of a query of the plurality of structured query language queries [0093] the analytic server may then execute the query command received at step 210 based on the subset of tables identified (e.g., tables that satisfy a similarity threshold or top number or portion of tables after the tables have been ranked). As a result, the analytic server may expand the end user's query command to other tables not identified by the end user. For instance, the analytic server may receive a query command and a corresponding table to search and implement the query command. However, using the methods and systems described herein, the analytic server may further identify additional tables not identified (or not known) to the end user. The analytic server may display a prompt and notify the end user regarding the additional tables. The analytic server may also automatically execute the query command (on the additional tables) and provide additional results to the user that were not specifically requested by the end user.
Regarding dependent claim 5, WAN et al, Popescu et al and Johnson et al teach, the method of claim 1.
Johnson et al further teaches, wherein at least some of the plurality of features are extracted from abstract syntax trees corresponding to the plurality of structured query language queries (Paragraph [0004] discloses, plurality of features are extracted from abstract syntax trees corresponding to plurality of structured language queries that were previously executed).
Regarding dependent claim 6, WAN et al, Popescu et al and Johnson et al teach, the method of claim 5.
Johnson et al further teaches, wherein the extracting comprises replacing a literal in a query of the plurality of structured query language queries with a token (Paragraph [0096] Normalization may include, for instance, conversion of names and keywords to lowercase, sorting lists, and/or converting literals to placeholders (e.g., st_cd=‘UU’).
Regarding dependent claim 7, WAN et al, Popescu et al and Johnson et al teach, the method of claim 1.
WAN et al further teaches, wherein the generating comprises determining, for each feature of the plurality of features, a count for the each feature among the plurality of structured query language queries (Fig. 5, Paragraph [0069] discloses frequency/count of each feature among the plurality of structured query language queries. Examiner interprets feature as table. Also see [0030]).
Johnson et al also further teaches, wherein the generating comprises determining, for each feature of the plurality of features, a count for the each feature among the plurality of structured query language queries (Paragraph [0096] Normalization may help to compute, for each feature, a count of the feature's occurrence across the plurality of queries. Thus, the statistical report may include a list of features extracted from the plurality of queries and frequencies of occurrence for each of the extracted features).
Regarding dependent claim 8, WAN et al, Popescu et al and Johnson et al teach, the method of claim 7.
WAN et al further teaches, wherein the count comprises a count of how many times the each feature occurs in the plurality of structured query language queries (Paragraph [0069] For the purpose of clarification, the following example will contain only three tables (table A, B, and C). However, a person skilled in the art will appreciate that the analytic server may implement the methods described herein using numerous tables. From the historical log data, the analytic server may find the relations of the three tables, such as the frequency of the tables being used together. FIG. 5A illustrates a matrix showing the frequency of tables being used together, according to one embodiment. The number may represent how often the tables are used together, such as the frequency of Tables A, B, C being used with each other. For example, the number “5” 502 may indicate that the frequency of Table A and Table B being used together is 5. The number “2” 504 may indicate that the frequency of Table B and Table C being used together is 2. Also see [0030]).
Johnson et al also further teaches, wherein the count comprises a count of how many times the each feature occurs in the plurality of structured query language queries (Paragraph [0096] Normalization may help to compute, for each feature, a count of the feature's occurrence across the plurality of queries. Thus, the statistical report may include a list of features extracted from the plurality of queries and frequencies of occurrence for each of the extracted features).
Regarding dependent claim 10, WAN et al, Popescu et al and Johnson et al teach, the method of claim 1.
WAN et al further teaches, wherein the identifying the subset of the plurality of features is performed with an input from a subject matter expert who has familiarity with the target database (Paragraph [0095] the analytic server may retrain the models based on the users' feedback. For example, when the analytic server recommends a list of tables, a user may select which tables are relevant, which tables are not related to the query. The analytic server may leverage such information from the user's feedback and retrain the models to dynamically fit the user's needs (Examiner interprets user selecting the relevant tables as input from a subject matter expert who has familiarity with the target database)).
Johnson et al also further teaches, wherein the identifying the subset of the plurality of features is performed with an input from a subject matter expert who has familiarity with the target database (Paragraph [0018] queries that were previously written by subject matter experts might be mined for useful information that may not be well-documented).
Regarding dependent claim 11, WAN et al, Popescu et al and Johnson et al teach, the method of claim 1.
Popescu et al further teaches, wherein the converting comprises: identifying, by the processing system, a field name referenced in a selected structured query language query from the subset of the plurality of structured query language queries (Paragraph [0040] discloses, identifying the field names from SQL queries);
Paragraph [0074] SQL generator 216 can identify in the selected SQL query values assigned to fields,
determining, by the processing system, a meaning of the field name (Paragraph [0042] discloses determining the meaning of the field names. Also see [0044]);
and providing, by the processing system to a first language model of the plurality of language models, the selected structured query language query and metadata describing the field name and the meaning, wherein the first language model uses the selected structured query language query and the metadata to convert the selected structured query language query into an initial version of the question phrased in natural language. (Paragraphs [0111]-[0113] discloses, the SQL query generated from user query uses the SQL to convert the selected SQL into natural language using the machine learning trained SQL-to-text model).
Regarding dependent claim 12, WAN et al, Popescu et al and Johnson et al teach, the method of claim 11.
Popescu et al further teaches, further comprising: rephrasing, by the processing system using a second language model of the plurality of language models, the initial version of the question (Paragraph [0113] discloses, rephrasing the initial version of the query using second language model such as SQL-to-text model).
Regarding dependent claim 13, WAN et al, Popescu et al and Johnson et al teach, the method of claim 11.
Popescu et al further teaches, wherein the question is a question that is capable of being answered by an exact code of the selected structured query language query (Paragraph [0112], [0113] discloses, Once the user have reviewed the rephrased query and confirms the SQL query is capable of executing the SQL query).
Regarding dependent claim 14, WAN et al, Popescu et al and Johnson et al teach, the method of claim 11.
Popescu et al further teaches, wherein the determining utilizes an analysis of the selected structured query language query to determine the meaning (Paragraph [0040] discloses, analysis of the SQL to determine the meaning such as a description of a database to which a question and/or SQL query pertain, actual strings recorded in a database to which a question and/or SQL query pertain, etc).
Regarding dependent claim 15, WAN et al, Popescu et al and Johnson et al teach, the method of claim 11.
Popescu et al further teaches, wherein the determining pulls a meaning of the field name from a schema of a table referenced in the selected structured query language query (Paragraph [0042] discloses determining the meaning of the field names from a schema of a table. Also see [0044]).
Regarding independent claim 16, WAN; Bo (US 20210158176 A1) teaches, a non-transitory computer readable medium storing instructions which, when executed by a processing system including at least one processor, cause the processing system to perform operations (Paragraph [0012]), the operations comprising: obtaining a query log from a database management system, wherein the query log contains a plurality of structured query language queries previously executed against a target database (Paragraph [0037] he query logs 110 may include all historical queries, such as SQL (structured query language) queries for the one or more databases of the data center 106);
extracting, from the plurality of structured query language queries, a plurality of features; associating the plurality of features with the plurality of structured query language queries (Paragraphs [0038], [0039] discloses, extracting plurality of features from the query logs);
generating a summary of the plurality of features (Paragraphs [0039], [0040] discloses, the analytic server 102 may implement a key phrase extraction model that utilizes natural language processing (NLP) to extract key phrases and features of each table's description (i.e., understanding each table's semantic, usage, and relationships to other tables within the data center);
identifying, based on the summary, a subset of the plurality of features to use as a basis for extracting queries from the plurality of structured query language queries (Paragraph [0041] The analytic server 102 may also generate a graph database based on the output from the query parser. As discussed above, the query parser may extract the tables and the relationships among the tables based on the historical queries. Also see [0044]);
identifying a subset of the plurality of structured query language queries that contains the subset of the plurality of features (Paragraph [0042] discloses, identifying the queries that are semantically similar to the user query. Also see [0051]);
WAN et al fails to explicitly teach, wherein the subset of the plurality of features includes features of the plurality of features that are extracted from at least two different structured query language queries of the plurality of structured query language queries; converting, using a plurality of language models, each structured query language query of the subset of the plurality of structured query language queries into a question phrased in natural language; prompting a language model that is trained to perform a text-to-structured query language conversion to generate query written in structured query language programming language, using the question and the each structured query language query together as part of a prompt that improves an accuracy of the text- to-structured query language conversion by the language model; and executing the query written in structured query language programming language against the database management system.
Popescu; Octavian (US 20250077517 A1) teaches, converting, using a plurality of language models, each structured query language query of the subset of the plurality of structured query language queries into a question phrased in natural language (Figs. 7, 8, Paragraphs [0112], [0113] discloses, converting structured query language query into a question phrased in natural language);
Therefore it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention, to have modified the teachings of WAN et al by converting, by the processing system using a plurality of language models, each structured query language query of the subset of the plurality of structured query language queries into a question phrased in natural language;, as taught by Popescu et al (Paragraphs [0112], [0113]).
One of the ordinary skill in the art would have been motivated to make this modification, by doing so, the user can review the natural language text in query results 844 to confirm whether that natural language text corresponds to the user query as taught by Popescu et al (Paragraph [0112]).
WAN et al and Popescu et al fails to explicitly teach, wherein the subset of the plurality of features includes features of the plurality of features that are extracted from at least two different structured query language queries of the plurality of structured query language queries; prompting a language model that is trained to perform a text-to-structured query language conversion to generate
Johnson; Theodore (US 20260050591 A1) teaches, wherein the subset of the plurality of features includes features of the plurality of features that are extracted from at least two different structured query language queries of the plurality of structured query language queries (Paragraph [0026] discloses, plurality of features are extracted from plurality of SQL queries/ query logs that were previously executed against one or more tables contained in the database based on frequencies of occurrence of the features);
prompting a language model that is trained to perform a text-to-structured query language conversion to generate (Paragraphs [0026]-[0028] discloses, generating a prompt for a new SQL query 116 to be executed against the database using query logs);
using the question and the each structured query language query together as part of a prompt that improves an accuracy of the text-to-structured query language conversion by the language model (Paragraph [0018] both query log analysis and database profiling/ questions may be used together to improve the generation of prompts for text-to-SQL queries).
Note: the limitation “improves an accuracy of the text-to-structured query language conversion by the language model” is an intended use and it does not specify how the accuracy is performed by the model.
Therefore it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention, to have modified the teachings of WAN et al and Popescu et al by providing wherein the subset of the plurality of features includes features of the plurality of features that are extracted from at least two different structured query language queries of the plurality of structured query language queries; prompting, by the processing system, a language model that is trained to perform a text-to-structured query language conversion to generate query written in structured query language programming language, using the question and the each structured query language query together as part of a prompt that improves an accuracy of the text-to-structured query language conversion by the language model; and executing, by the processing system, the query written in structured query language programming language against the database management system, as taught by Johnson et al (Paragraphs [0026]-[0028]).
One of the ordinary skill in the art would have been motivated to make this modification, by doing so, would improve the generation of prompts for text-to-SQL queries. The disclosed approaches may assist an LLM in generating prompts for new SQL queries, even for SQL databases that the LLM has not previously encountered as taught by Johnson et al (Paragraph [0018]).
Regarding dependent claim 17, WAN et al, Popescu et al and Johnson et al teach, the non-transitory computer readable medium of claim 16.
WAN et al further teaches, wherein the plurality of features includes at least one of: a set of tables referenced in a query of the plurality of structured query language queries (Paragraph [0030] discloses, set of tables referenced in the historical query logs).
a collection of tables which are joined together in a query of the plurality of structured query language queries, a set of fields referenced by a query of the plurality of structured query language queries, a set of select elements in a query of the plurality of structured query language queries named using an as clause, a set of unnamed select elements in a query of the plurality of structured query language queries, a set of predicates which constrain only one range variable in a query of the plurality of structured query language queries, a set of predicates in an on clause in a query of the plurality of structured query language queries, a set of predicates which constrain multiple range variables in a query of the plurality of structured query language queries, a partitioning field of a window function in a query of the plurality of structured query language queries, an output field of a window function in a query of the plurality of structured query language queries, a set of named subqueries in a query of the plurality of structured query language queries, a set of named and unnamed fields in named subqueries in a query of the plurality of structured query language queries, a group-by field in a query of the plurality of structured query language queries, a normalized indication of a complexity of a query of the plurality of structured query language queries, or a normalized indicator of a shape of a query of the plurality of structured query language queries.
Regarding independent claim 20, WAN; Bo (US 20210158176 A1) teaches, a system comprising: a processing system comprising at least one processor; and a non-transitory computer readable medium storing instructions which, when executed by the processing system, cause the processing system to perform operations (Paragraph [0012]), the operations comprising: obtaining a query log from a database management system, wherein the query log contains a plurality of structured query language queries previously executed against a target database (Paragraph [0037] he query logs 110 may include all historical queries, such as SQL (structured query language) queries for the one or more databases of the data center 106);
extracting, from the plurality of structured query language queries, a plurality of features; associating the plurality of features with the plurality of structured query language queries (Paragraphs [0038], [0039] discloses, extracting plurality of features from the query logs);
generating a summary of the plurality of features (Paragraphs [0039], [0040] discloses, the analytic server 102 may implement a key phrase extraction model that utilizes natural language processing (NLP) to extract key phrases and features of each table's description (i.e., understanding each table's semantic, usage, and relationships to other tables within the data center);
identifying, based on the summary, a subset of the plurality of features to use as a basis for extracting queries from the plurality of structured query language queries (Paragraph [0041] The analytic server 102 may also generate a graph database based on the output from the query parser. As discussed above, the query parser may extract the tables and the relationships among the tables based on the historical queries. Also see [0044]);
identifying a subset of the plurality of structured query language queries that contains the subset of the plurality of features (Paragraph [0042] discloses, identifying the queries that are semantically similar to the user query. Also see [0051]);
WAN et al fails to explicitly teach, wherein the subset of the plurality of features includes features of the plurality of features that are extracted from at least two different structured query language queries of the plurality of structured query language queries; converting, using a plurality of language models, each structured query language query of the subset of the plurality of structured query language queries into a question phrased in natural language; prompting a language model that is trained to perform a text-to- structured query language conversion to generate query written in structured query language programming language, using the question and the each structured query language query together as part of a prompt that improves an accuracy of the text-to-structured query language conversion by the language model; and executing the query written in structured query language programming language against the database management system.
Popescu; Octavian (US 20250077517 A1) teaches, converting, using a plurality of language models, each structured query language query of the subset of the plurality of structured query language queries into a question phrased in natural language (Figs. 7, 8, Paragraphs [0112], [0113] discloses, converting structured query language query into a question phrased in natural language);
Therefore it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention, to have modified the teachings of WAN et al by converting, by the processing system using a plurality of language models, each structured query language query of the subset of the plurality of structured query language queries into a question phrased in natural language;, as taught by Popescu et al (Paragraphs [0112], [0113]).
One of the ordinary skill in the art would have been motivated to make this modification, by doing so, the user can review the natural language text in query results 844 to confirm whether that natural language text corresponds to the user query as taught by Popescu et al (Paragraph [0112]).
WAN et al and Popescu et al fails to explicitly teach, herein the subset of the plurality of features includes features of the plurality of features that are extracted from at least two different structured query language queries of the plurality of structured query language queries; prompting a language model that is trained to perform a text-to- structured query language conversion to generate query written in structured query language programming language, using the question and the each structured query language query together as part of a prompt that improves an accuracy of the text-to-structured query language conversion by the language model; and executing the query written in structured query language programming language against the database management system.
Johnson; Theodore (US 20260050591 A1) teaches, wherein the subset of the plurality of features includes features of the plurality of features that are extracted from at least two different structured query language queries of the plurality of structured query language queries (Paragraph [0026] discloses, plurality of features are extracted from plurality of SQL queries/ query logs that were previously executed against one or more tables contained in the database based on frequencies of occurrence of the features);
prompting, by the processing system, a language model that is trained to perform a text-to-structured query language conversion to generate query written in structured query language programming language, and executing, by the processing system, the query written in structured query language programming language against the database management system (Paragraphs [0026]-[0028] discloses, generating a prompt for a new SQL query 116 to be executed against the database using query logs);
using the question and the each structured query language query together as part of a prompt that improves an accuracy of the text-to-structured query language conversion by the language model (Paragraph [0018] both query log analysis and database profiling/ questions may be used together to improve the generation of prompts for text-to-SQL queries).
Note: the limitation “improves an accuracy of the text-to-structured query language conversion by the language model” is an intended use and it does not specify how the accuracy is performed by the model.
Therefore it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention, to have modified the teachings of WAN et al and Popescu et al by providing wherein the subset of the plurality of features includes features of the plurality of features that are extracted from at least two different structured query language queries of the plurality of structured query language queries; prompting, by the processing system, a language model that is trained to perform a text-to-structured query language conversion to generate query written in structured query language programming language, using the question and the each structured query language query together as part of a prompt that improves an accuracy of the text-to-structured query language conversion by the language model; and executing, by the processing system, the query written in structured query language programming language against the database management system, as taught by Johnson et al (Paragraphs [0026]-[0028]).
One of the ordinary skill in the art would have been motivated to make this modification, by doing so, would improve the generation of prompts for text-to-SQL queries. The disclosed approaches may assist an LLM in generating prompts for new SQL queries, even for SQL databases that the LLM has not previously encountered as taught by Johnson et al (Paragraph [0018]).
9. Claims 4, 9 is rejected under 35 U.S.C. 103 as being unpatentable over WAN; Bo (US 20210158176 A1) in view of Popescu; Octavian (US 20250077517 A1), Johnson; Theodore (US 20260050591 A1) and in further view of Yuyama; Norihisa (US 20170075955 A1).
Regarding dependent claim 4, WAN et al, Popescu et al and Johnson et al teach, the method of claim 3.
WAN et al, Popescu et al and Johnson et al fails to explicitly teach, wherein the at least one feature is at least one of: a start time for a query of the plurality of structured query language queries, a finish time for a query of the plurality of structured query language queries, a result indication for a query of the plurality of structured query language queries, or an ordinal position of a query of the plurality of structured query language queries in a list of the plurality of structured query language queries.
Yuyama; Norihisa (US 20170075955 A1) teaches, a result indication for a query of the plurality of structured query language queries (Fig. 4 Paragraph [0071] discloses, there is a SQL command sq which is executed a plurality of number of (n) times and the SQL command sq has not been executed n times).
Therefore it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention, to have modified the teachings of WAN et al, Popescu et al and Johnson et al by wherein the at least one feature is at least one of: a start time for a query of the plurality of structured query language queries, a finish time for a query of the plurality of structured query language queries, a result indication for a query of the plurality of structured query language queries, or an ordinal position of a query of the plurality of structured query language queries in a list of the plurality of structured query language queries as taught by Yuyama et al (Paragraphs [0071]).
One of the ordinary skill in the art would have been motivated to make this modification, by doing so, the processing time of the SQL command sq is reduced and the performance of the access to the database 205 is improved as taught by Yuyama et al (Paragraphs [0072]).
Regarding dependent claim 9, WAN et al, Popescu et al and Johnson et al teach, the method of claim 7.
Yuyama et al further teaches, wherein the count comprises a count of how many queries of the plurality of structured query language queries include the each feature (Fig. 4 Paragraph [0071] discloses, there is a SQL command sq which is executed a plurality of number of (n) times and the SQL command sq has not been executed n times. Also see Fig. 21, Paragraph [0216]).
10. Claims 18 is rejected under 35 U.S.C. 103 as being unpatentable over WAN; Bo (US 20210158176 A1) in view of Popescu; Octavian (US 20250077517 A1), Johnson; Theodore (US 20260050591 A1) and in further view of Panda; Srikant (US 20250238614 A1).
Regarding dependent claim 18, WAN et al, Popescu et al and Johnson et al teach, the non-transitory computer readable medium of claim 16.
Popescu et al further teaches, wherein the converting comprises: converting, using a first language model of the plurality of language models, a selected structured query language query from the subset of the plurality of structured query language queries into an initial version of the question phrased in natural language (Paragraphs [0111]-[0113] discloses, the SQL query generated from user query uses the SQL to convert the selected SQL into natural language using the machine learning trained SQL-to-text model).
and rephrasing, using a third language model, the initial version of the question associated with the target database (Paragraph [0113] discloses, rephrasing the initial version of the query using second language model such as SQL-to-text model).
WAN et al and Popescu et al fails to explicitly teach, domain-specific language
Panda; Srikant (US 20250238614 A1) teaches, and rephrasing, using a third language model, the initial version of the question using a domain-specific language associated with the target database (Paragraph [0128] Prompt information that includes a prompt along with a dynamic vocabulary for the prompt can be provided to the language model, which in turn can predict tokens for an output statement that are contextually consistent with the tokens included the dynamic vocabulary. In this way, by taking into account the contents of the input prompt and domain-specific information. (i.e., rephrasing initial version of the query using a domain-specific language)).
Therefore it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention, to have modified the teachings of WAN et al, Popescu et al and Johnson et al by rephrasing, using a third language model, the initial version of the question using a domain-specific language associated with the target database as taught by Panda et al (Paragraph [0128]).
One of the ordinary skill in the art would have been motivated to make this modification, by doing so, accuracy of results from queries executed using these output statements can be improved as taught by Panda et al (Paragraph [0128]).
11. Claims 19 is rejected under 35 U.S.C. 103 as being unpatentable over WAN; Bo (US 20210158176 A1) in view of Popescu; Octavian (US 20250077517 A1), Johnson; Theodore (US 20260050591 A1), Panda; Srikant (US 20250238614 A1), and in further view of SHIMIZU; Toshihiro (US 20120310648 A1).
Regarding dependent claim 19, WAN et al, Popescu et al, Johnson et al and Panda et al teach, the non-transitory computer readable medium of claim 18.
WAN et al, Popescu et al, Panda et al fails to explicitly teach, further comprising, prior to the converting the selected structured query language query but prior to the rephrasing: replacing, using a second language model, a field name in the initial version of the question with a natural language phrase.
SHIMIZU; Toshihiro (US 20120310648 A1) teaches, further comprising, prior to the converting the selected structured query language query but prior to the rephrasing: replacing, using a second language model, a field name in the initial version of the question with a natural language phrase (Fig. 9 Paragraphs [0086]-[0088] discloses, plurality of delete statements comprises replacing a literal in a query of the plurality of structured query language queries with a token. Examiner interprets a literal as character string "LNAME").
Therefore it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention, to have modified the teachings of WAN et al, Popescu et al, Johnson et al and Panda et al by prior to the converting the selected structured query language query but prior to the rephrasing: replacing, using a second language model, a field name in the initial version of the question with a natural language phrase as taught by SHIMIZU et al (Paragraphs [0085]-[0088]).
One of the ordinary skill in the art would have been motivated to make this modification, by doing so, mapping between an event character string of an input sentence and an abstract syntax tree may be preserved as taught by SHIMIZU et al (Paragraph [0084]).
Closest Prior Art
12. The prior art made of record and not relied upon is considered pertinent to the applicant’s disclosure.
HAN; Wook Shin (US 20250061111 A1) teaches, [0004] In methods employing an SQL-to-text model, a model for translating SQL into a natural language is used. First, a database corresponding to a large number of SQL queries is crawled from a web or SQL queries are automatically synthesized with predefined grammar and a given database, and then a natural language query sentence corresponding to each SQL query is generated using a model. Compared to the template-based methods, these methods have an advantage in that it is possible to generate various natural language query expressions. However, it is still difficult to collect or generate various data for training an SQL-to-text model.
Agarwal, Manoj K ( US 20050172306 A1) teaches, (Paragraph [0085] An SQL event is sent to the management server 80 for each SQL request execution captured from a monitoring interface of the database server 110. Each event comprises an SQL query string, a time stamp that is defined to be the end time of the SQL request, and the start time of the query).
13. Examiner has pointed out particular references contained in the prior arts of record in the body of this action for the convenience of the applicant. Although the specified citations are representative of the teachings in the art and are applied to the specific limitations within the individual claim, other passages and Figures may apply as well. It is respectfully requested from the applicant, in preparing the response, to consider fully the entire references as potentially teaching all or part of the claimed invention, as well as the context of the passage as taught by the prior arts or disclosed by the examiner. It is noted that any citation to specific pages, columns, figures, or lines in the prior art references any interpretation of the references should not be considered to be limiting in any way. A reference is relevant for all it contains and may be relied upon for all that it would have reasonably suggested to one having ordinary skill in the art. In re Heck, 699 F.2d 1331-33, 216 USPQ 1038-39 (Fed. Cir. 1983) (quoting In re Lemelson, 397 F.2d 1006, 1009, 158 USPQ 275, 277 (CCPA 1968))).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SUMAN RAJAPUTRA whose telephone number is (571) 272-4669. The examiner can normally be reached between 8:00 AM - 5:00 PM.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Tony Mahmoudi (571) 272-4078 can be reached. 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.
/S. R./
Examiner, Art Unit 2163
/TONY MAHMOUDI/Supervisory Patent Examiner, Art Unit 2163
Prosecution Insights