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
This is a first final Office Action for application 18/966,940, in response to arguments and amendments filed on 04/30/2026. Claims 1, 5-7, 9-10, and 15 are currently amended. Claims 1-20 are pending and examined below.
Specification
The amendments to specification are sufficient to overcome previous objections, therefore previous objections to the specification are withdrawn.
Claim Objections
The amendment to claim 9 is sufficient to overcome previous objection. The addition of a conjunction clarified the scope of the SQL statements. Objection “an SQL SELECT” as opposed to “a SQL SELECT” is withdrawn because both forms are acceptable. Therefore previous objections are withdrawn.
Response to Arguments
Applicant's arguments filed 04/30/2026 have been fully considered but they are not persuasive.
Applicant argues that reference arts Lin and Modi do not teach the new claim 1 language
A computer-implemented method, comprising: generating, by computer hardware, a tokenized schema from a Structured Query Language (SQL) schema, a tokenized natural language question, and a plurality of tokenized candidates, wherein the plurality of tokenized candidates are generated by tokenizing a plurality of candidates from the SQL schema, wherein each of the plurality of tokenized candidates is formed by a string of tokens having a first token representing an initial delimiter and a last token representing an end delimiter, and wherein the tokenized natural language question is generated by tokenizing a natural language question;
generating a plurality of concatenated vectors from each of the transformed vectorial representations of the plurality of tokenized candidates, wherein each of the concatenated vectors is generated by concatenating a first transformed vectorial representation corresponding to the first token with a last transformed vectorial representation corresponding to the last token of each tokenized candidate of the plurality of tokenized candidates;
and generating quantitative relevancies for the plurality of candidates from the SQL schema based on the concatenated vectors, wherein the quantitative relevancies are represented as predicted probabilities corresponding to a likelihood that a column of the SQL schema is relevant to the natural language question.
However, as shown below, Modi teaches
a plurality of tokenized candidates, wherein the plurality of tokenized candidates are generated by tokenizing a plurality of candidates from the SQL schema
because it shows that a plurality of potential views (i.e. candidates) of database views.
Further, Lin teaches and generating quantitative relevancies for the plurality of candidates from the SQL schema based on the concatenated vectors, wherein the quantitative relevancies are represented as predicted probabilities corresponding to a likelihood that a column of the SQL schema is relevant to the natural language question
because fuzzy matching itself involves similarity scores between two data elements, and similarity scores are “predicted probabilities corresponding to a likelihood”, particularly when read under the broadest reasonable interpretation standard for claim interpretation. Therefore arguments are unpersuasive.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1, 5-10, 14-15 and 19-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Lin et al. (US Pub. 2021/0374133) in view of Modi et al. (US Pub. 2026/0072901).
Regarding claim(s) 1, Lin teaches
A computer-implemented method, comprising: generating, by computer hardware, a tokenized schema from a Structured Query Language (SQL) schema, a tokenized natural language question, and a plurality of tokenized candidates, wherein the plurality of tokenized candidates are generated by tokenizing a plurality of candidates from the SQL schema, wherein each of the plurality of tokenized candidates is formed by a string of tokens having a first token representing an initial delimiter and a last token representing an end delimiter, and wherein the tokenized natural language question is generated by tokenizing a natural language question; (Fig. 2; Par. [0024] in the process of tokenizing table values for a value-aware schema encoder (#170) and user question (i.e. tokenized natural language question; #140), each table and field name have special tokens and can be concatenated by surrounding them (i.e. initial and end) with two special separator (delimiter) tokens (examiner notes that the plurality of questions is taught by Modi as shown below, but the language is left here for referential clarity))
generating, by the computer hardware, vectorial representations of the tokenized schema, tokenized natural language question, and each of the plurality of tokenized candidates; (Fig. 1; Par. [0019, 24, 26-7] a natural language query (#140) and table schema are tokenized and then vectorized in the process of encoding in a serialized question-schema representation)
generating, by the computer hardware, transformed vectorial representations by processing the vectorial representations through a decoder model; (Fig. 1; Par. [0019, 24, 26-7] a natural language query (#140) and table schema are tokenized and then vectorized in the process of encoding in a serialized question-schema representation)
generating a plurality of concatenated vectors from each of the transformed vectorial representations of the plurality of tokenized candidates, wherein each of the concatenated vectors is generated by concatenating a first transformed vectorial representation corresponding to the first token with a last transformed vectorial representation corresponding to the last token of each tokenized candidate of the plurality of tokenized candidates; and (Fig. 1; Par. [0019, 24, 26-7] a natural language query (#140) and table schema are tokenized and then vectorized in the process of encoding in a serialized (i.e. concatenated) question-schema representation)
and generating quantitative relevancies for the plurality of candidates from the SQL schema based on the concatenated vectors, wherein the quantitative relevancies are represented as predicted probabilities corresponding to a likelihood that a column of the SQL schema is relevant to the natural language question. (Par. [0028] fuzzy matching (i.e. based on quantitative relevancies represented as predicted probabilities) is used for field matching (i.e. likelihood that a column of the SQL schema is relevant))
Lin does not explicitly teach
a plurality of tokenized candidates, wherein the plurality of tokenized candidates are generated by tokenizing a plurality of candidates from the SQL schema
through a decoder-only model
However, from the same field Modi teaches
a plurality of tokenized candidates, wherein the plurality of tokenized candidates are generated by tokenizing a plurality of candidates from the SQL schema (Fig. 4; Par. [0019, 106] the schema definition (#412) can include a set (i.e. a plurality of candidates) of database views)
through a decoder-only model (Par. [0087] decoder-only language models can be used in a natural language to database query generation system)
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to combine the SQL schema candidates and the decoder-only model in the query construction system of Modi into the natural language to SQL query system of Lin. The motivation for this combination would have been to improve system discovery as explained in Modi (Par. [0019]).
Regarding claim(s) 5, Lin and Modi teach claim 1 as shown above, and Modi further teaches
The computer implemented method of claim 1, further comprising: for each tokenized candidate, selecting the first transformed vectorial representation of the candidate and the last transformed vectorial representation of the candidate and excluding each intervening transformed vectorial representation of the candidate from the concatenating. (Fig. 4; Par. [0106] the schema definition (#412) can include partial concatenations of blocks via properties)
Regarding claim(s) 6, Lin and Modi teach claim 1 as shown above, and Lin further teaches
The computer-implemented method of claim 1, wherein each of the plurality of candidates from the SQL schema uniquely corresponds to the column of the SQL schema. (Par. [0021, 24] encoding of the table includes table and field (i.e. column) name, including tokens specific to each)
Regarding claim(s) 7, Lin and Modi teach claim 1 as shown above, and Lin further teaches
The computer-implemented method of claim 1, wherein each of the plurality of candidates from the SQL schema is formed by pairing a name of an SQL table and a column of the SQL table, wherein the column of the SQL table comprises the column of the SQL schema. (Par. [0021, 24] encoding of the table includes table and field (i.e. column) name, including tokens specific to each)
Regarding claim(s) 8, Lin and Modi teach claim 1 as shown above, and Lin further teaches
The computer-implemented method of claim 1, wherein the decoder-only model is trained using a ground truth schema linking produced by a static analysis of a ground truth SQL statement. (Par. [0021, 8] pick lists are used as anchor values (i.e. ground truth) for analyzing field values associated with incoming queries )
Regarding claim(s) 9, Lin and Modi teach claim 8 as shown above, and Lin further teaches
The computer-implemented method of claim 8, wherein the static analysis identifies all SQL select columns of the SQL schema as determined by an SQL SELECT and determines all columns relevant to the ground truth SQL statement using JOIN, WHERE, HAVING, GROUP BY, and ORDER statements. (Par. [0021, 8, 36] pick lists are used as anchor values for analyzing field values (i.e. columns) associated with incoming queries and include using joins and other relevant filtering criteria in the training set)
Regarding claim(s) 10, while worded slightly different, Is rejected under the same rationale as claim 1. Lin further teaches
one or more processors capable of initiating operations (Fig. 1 #110, #120)
Regarding claim(s) 14, while worded slightly different, Is rejected under the same rationale as claim 5.
Regarding claim(s) 15, while worded slightly different, Is rejected under the same rationale as claim 10.
Regarding claim(s) 19, while worded slightly different, Is rejected under the same rationale as claim 5.
Regarding claim(s) 20, while worded slightly different, Is rejected under the same rationale as claim 8.
Claim(s) 2-4, 11-13 and 16-18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Lin et al. (US Pub. 2021/0374133) in view of Modi et al. (US Pub. 2026/0072901), and further in view of Lazarev et al. (US Pub. 2025/0225128).
Regarding claim(s) 2, Lin and Modi teach claim 1 as show above but does not explicitly teach
The computer-implemented method of claim 1, further comprising: selecting one or more candidates from the plurality of candidates from the SQL schema based on the quantitative relevancies for inclusion in an SQL query for the natural language question.
However, from the same field Lazarev teaches
The computer-implemented method of claim 1, further comprising: selecting one or more candidates from the plurality of candidates from the SQL schema based on the quantitative relevancies for inclusion in an SQL query for the natural language question. (Par. [0028] a fuzzy matching (i.e. based on quantitative relevancies) is used for field matching in the process of turning a natural language query into a SQL query)
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to combine the fuzzy matching of Lazarev into the natural language to SQL query system of Lin. The motivation for this combination would have been to improve a natural language query to database query conversion as explained in Lazarev (Par. [0027]).
Regarding claim(s) 3, Lin, Modi and Lazarev teach claim 1 as shown above, and Lazarev further teaches
The computer-implemented method of claim 2, further comprising: generating the SQL query using the one or more candidates selected. (Par. [0028] a fuzzy matching (i.e. based on quantitative relevancies) is used for field matching in the process of turning a natural language query into a SQL query)
Regarding claim(s) 4, Lin, Modi and Lazarev teach claim 1 as shown above, and Lin further teaches
The computer-implemented method of claim 1, wherein each of the plurality of candidates from the SQL schema uniquely corresponds to a column of the SQL schema. (Par. [0019] the SQL query is executable)
Regarding claim(s) 11, while worded slightly different, Is rejected under the same rationale as claim 2.
Regarding claim(s) 12, while worded slightly different, Is rejected under the same rationale as claim 3.
Regarding claim(s) 13, while worded slightly different, Is rejected under the same rationale as claim 4.
Regarding claim(s) 16, while worded slightly different, Is rejected under the same rationale as claim 2.
Regarding claim(s) 17, while worded slightly different, Is rejected under the same rationale as claim 3.
Regarding claim(s) 18, while worded slightly different, Is rejected under the same rationale as claim 4.
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to J MITCHELL CURRAN whose telephone number is (469)295-9081. The examiner can normally be reached M-F 8:00am - 5:00pm.
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, Sherief Badawi can be reached at (571) 272-9782. 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.
/J MITCHELL CURRAN/Examiner, Art Unit 2169
/BRITTANY N ALLEN/Primary Examiner, Art Unit 2169
Prosecution Insights