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 .
Claims 1-17 are presented for examination.
Priority
Receipt is acknowledged of certified copies of papers required by 37 CFR 1.55.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on October 20, 2023 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Specification
The disclosure is objected to because of the following informalities:
[0087]: "columns to be queries" should read "columns to be queried"
[0088]: "so the different between" should read "so the difference between"
Appropriate correction is required.
Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is invoked.
As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph:
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action.
This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because the claim limitation(s) uses a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitation(s) is/are: query-processing unit, machine-learning unit, and data storage unit in claims 1-7. The corresponding structure described in the specification as performing the claimed function for “data storage unit” is found in paragraph [0099]: “The memory 1030 and the storage 1060 may be storage media including at least one of a volatile medium, a nonvolatile medium, a detachable medium, a non-detachable medium, a communication medium, or an information delivery medium, or a combination thereof. For example, the memory 1030 may include ROM 1031 or RAM 1032.”
Because these claim limitation(s) is/are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, it/they is/are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof.
If applicant does not intend to have this/these limitation(s) interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, applicant may: (1) amend the claim limitation(s) to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitation(s) recite(s) sufficient structure to perform the claimed function so as to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph.
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 1-7 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as
failing to comply with the written description requirement. The claim(s) contains subject matter which
was not described in the specification in such a way as to reasonably convey to one skilled in the relevant
art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention.
Claim 1 recites the limitations “query-processing unit” and “machine-learning unit” which invoke 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. However, the written description fails to disclose the corresponding structure, material, or acts for performing the entire claimed functions and to clearly link the structure, material, or acts to the functions. Therefore, the written description is inadequate to show that the inventor had possession of the claimed invention at the time of filing. Claims 2-7 are rejected for being dependent on a rejected base claim. See rejections under 35 U.S.C. 112(b) below for further analysis.
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-7 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph,
as being indefinite for failing to particularly point out and distinctly claim the subject matter which the
inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards
as the invention.
Claim 1 recites limitations “query-processing unit” and “machine-learning unit” which invoke 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. However, the written description fails to disclose the corresponding structure, material, or acts for performing the entire claimed function and to clearly link the structure, material, or acts to the function. Therefore, the claim is indefinite and is rejected under 35 U.S.C. 112(b) or pre-AIA 35 U.S.C. 112, second paragraph. Claims 2-7 are rejected for being dependent on a rejected base claim. For examination purposes, Examiner is interpreting “query-processing unit” and “machine-learning unit” to be any combination of hardware/software capable of performing the function.
Applicant may:
(a) Amend the claim so that the claim limitation will no longer be interpreted as a limitation under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph;
(b) Amend the written description of the specification such that it expressly recites what structure, material, or acts perform the entire claimed function, without introducing any new matter (35 U.S.C. 132(a)); or
(c) Amend the written description of the specification such that it clearly links the structure, material, or acts disclosed therein to the function recited in the claim, without introducing any new matter (35 U.S.C. 132(a)).
If applicant is of the opinion that the written description of the specification already implicitly or inherently discloses the corresponding structure, material, or acts and clearly links them to the function so that one of ordinary skill in the art would recognize what structure, material, or acts perform the claimed function, applicant should clarify the record by either:
(a) Amending the written description of the specification such that it expressly recites the corresponding structure, material, or acts for performing the claimed function and clearly links or associates the structure, material, or acts to the claimed function, without introducing any new matter (35 U.S.C. 132(a)); or
(b) Stating on the record what the corresponding structure, material, or acts, which are implicitly or inherently set forth in the written description of the specification, perform the claimed function. For more information, see 37 CFR 1.75(d) and MPEP §§ 608.01(o) and 2181.
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.
Claims 1-17 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The analysis of the claims will follow the 2019 Revised Patent Subject Matter Eligibility Guidance (“2019 PEG”).
Claim 1
Step 1: The claim recites an apparatus, and therefore is directed to the statutory category of machines.
Step 2A Prong 1: The claim recites, inter alia:
“…analyzing a predictive spatiotemporal query of a user”; This limitation encompasses mentally analyzing a predictive spatiotemporal query of a user.
“…generating synthetic spatiotemporal data…”; This limitation encompasses mentally generating synthetic spatiotemporal data.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The claim further recites “returning a processing result” and “a data storage unit for storing raw spatiotemporal data and the generated synthetic spatiotemporal data, wherein the raw spatiotemporal data is stored in a form of a table including an identifier column and a position column,” however these limitations amount to the insignificant extra-solution activity of mere data gathering and outputting (MPEP 2106.05(g)). The claim further recites “a query-processing unit,” “a machine-learning unit,” and that the synthetic spatiotemporal data is generated “based on the machine-learning model,” however these limitations amount to mere instructions to apply a judicial exception using generic computer components programmed with a generic class of computer algorithms (MPEP 2106.05(f)). The claim further recites “training a machine-learning model in response to a request from the query-processing unit,” however this limitation amounts to merely generally linking the use of the judicial exception to the technological environment of model training (MPEP 2106.05(h)).
Step 2B: The claim does not contain significantly more than the judicial exception. The “returning a processing result” and “data storage unit for storing raw spatiotemporal data and the generated synthetic data” limitations, in addition to reciting insignificant extra solution activity, are also directed to the well-understood, routine, and conventional activity of storing and retrieving information in memory (MPEP 2106.05(d)(iv) Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015); OIP Techs., 788 F.3d at 1363, 115 USPQ2d at 1092-93). Otherwise, the analysis at this step mirrors that of Step 2A Prong 2 above. As an ordered whole, the claim is directed to a generic computer that performs a mentally performable process of analyzing a predictive spatiotemporal query and generating synthetic spatiotemporal data. Nothing in the claim provides significantly more than this. As such, the claim is not patent eligible.
Claim 2
Step 1: A machine, as above.
Step 2A Prong 1: The claim recites, inter alia:
“…selects a column of the raw spatiotemporal data to be learned”; This limitation encompasses mentally selecting a column of the raw spatiotemporal data to be learned.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The claim further recites that the selection is performed by “the machine-learning unit,” however this limitation amounts to mere instructions to apply a judicial exception using a generic computer programmed with a generic class of computer algorithms (MPEP 2106.05(f)).
Step 2B: The claim does not contain significantly more than the judicial exception. The analysis at this step mirrors that of Step 2A Prong 2 above.
Claim 3
Step 1: A machine, as above.
Step 2A Prong 1: The claim recites the same judicial exception as claim 2.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The claim further recites “the machine-learning unit trains the machine-learning model while changing a condition value for the column to be learned,” however this limitation amounts to merely generally linking the use of the judicial exception to the technological environment of model training (MPEP 2106.05(h)).
Step 2B: The claim does not contain significantly more than the judicial exception. The analysis at this step mirrors that of Step 2A Prong 2 above.
Claim 4
Step 1: A machine, as above.
Step 2A Prong 1: The claim recites the same judicial exception as claim 2.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The claim further recites “the machine-learning unit stores metadata corresponding to training of the machine-learning model, and the metadata includes information about the learned raw spatiotemporal data, information about a condition for the column, and information about a structure of the machine-learning model.” however this limitation amounts to the insignificant extra-solution activity of mere data gathering and outputting (MPEP 2106.05(g)).
Step 2B: The claim does not contain significantly more than the judicial exception. The “machine-learning unit stores metadata” limitation, in addition to reciting insignificant extra-solution activity, is also directed to the well-understood, routine, and conventional activity of storing and retrieving information in memory (MPEP 2106.05(d)(iv) Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015); OIP Techs., 788 F.3d at 1363, 115 USPQ2d at 1092-93).
Claim 5
Step 1: A machine, as above.
Step 2A Prong 1: The claim recites, inter alia:
“…analyzes the predictive spatiotemporal query of the user, thereby extracting information about target data and columns to be queried”; This limitation encompasses mentally analyzing the predictive spatiotemporal query of the user by extracting information about target data and columns to be queried.
“…determines whether synthetic spatiotemporal data and a trained machine-learning model are present based on the information about the target data and columns to be queried”; This limitation encompasses mentally determining whether synthetic spatiotemporal data and a trained machine-learning model are present based on the information about the target data and columns to be queried.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The claim further recites that the “query-processing unit” performs the analyzing and the “machine-learning unit” performs the determining, however these limitations amount to mere instructions to apply a judicial exception using a generic computer programmed with a generic class of computer algorithms (MPEP 2106.05(f)). The claim further recites that the machine-learning unit “returns a result value for the predictive spatiotemporal query based on the synthetic spatiotemporal data,” however this limitation amounts to the insignificant extra-solution activity of mere data gathering and outputting (MPEP 2106.05(g)).
Step 2B: The claim does not contain significantly more than the judicial exception. The “returns a result value” limitation, in addition to reciting insignificant extra solution activity, is also directed to the well-understood, routine, and conventional activity of receiving or transmitting data over a network (MPEP 2106.05(d)(II)(i) OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015)). Otherwise, the analysis at this step mirrors that of Step 2A Prong 2.
Claim 6
Step 1: A machine, as above.
Step 2A Prong 1: The claim recites, inter alia:
“when synthetic data corresponding to the target data and columns to be queried is not present, the machine-learning unit determines whether a machine-learning model corresponding to the target data and columns to be queried is present”; This limitation encompasses, excepting the recitation of generic computer components (the machine-learning unit), mentally determining whether a machine-learning model corresponding to the target data and columns to be queried is present when synthetic data corresponding to the target data and columns to be queried is not present.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The claim further recites that the “machine-learning unit” performs the determining, however this limitation amounts to mere instructions to apply a judicial exception using a generic computer programmed with a generic class of computer algorithms (MPEP 2106.05(f)).
Step 2B: The claim does not contain significantly more than the judicial exception. The analysis at this step mirrors that of Step 2A Prong 2.
Claim 7
Step 1: A machine, as above.
Step 2A Prong 1: The claim recites, inter alia:
“when synthetic data corresponding to the target data and columns to be queried is not present but a machine-learning model corresponding thereto is present, the machine-learning unit generates synthetic data corresponding to the target data and columns…”; This limitation encompasses, excepting the recitation of generic computer components (the machine-learning unit), mentally generating synthetic data corresponding to the target data and columns when synthetic data corresponding to the target data and columns to be queried is not present but a machine-learning model corresponding thereto is present.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The claim further recites that the “machine-learning unit” performs the generating, and that the generating is “based on the machine-learning model,” however these limitations amount to mere instructions to apply a judicial exception using a generic computer programmed with a generic class of computer algorithms (MPEP 2106.05(f)).
Step 2B: The claim does not contain significantly more than the judicial exception. The analysis at this step mirrors that of Step 2A Prong 2.
Claim 8
Step 1: The claim recites a method, and therefore is directed to the statutory category of processes.
Step 2A Prong 1: The claim recites, inter alia:
“determining a structure of a machine-learning model for generating synthetic spatiotemporal data”; This limitation encompasses mentally determining a structure of a machine-learning model for generating synthetic spatiotemporal data.
“generating synthetic spatiotemporal data…”; This limitation encompasses mentally generating synthetic spatiotemporal data.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The claim further recites “wherein the raw spatiotemporal data is stored in a form of a table including an identifier column and a position column,” however this limitation amounts to the insignificant extra-solution activity of mere data gathering and outputting (MPEP 2106.05(g)). The claim further recites that the synthetic spatiotemporal data is generated “based on the machine-learning model,” however this limitation amounts to mere instructions to apply a judicial exception using generic computer components programmed with a generic class of computer algorithms (MPEP 2106.05(f)). The claim further recites “training the machine-learning model based on raw spatiotemporal data,” however this limitation amounts to merely generally linking the use of the judicial exception to the technological environment of model training (MPEP 2106.05(h)).
Step 2B: The claim does not contain significantly more than the judicial exception. The “raw spatiotemporal data is stored in the form of a table” limitation, in addition to reciting insignificant extra solution activity, are also directed to the well-understood, routine, and conventional activity of storing and retrieving information in memory (MPEP 2106.05(d)(iv) Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015); OIP Techs., 788 F.3d at 1363, 115 USPQ2d at 1092-93). Otherwise, analysis at this step mirrors that of Step 2A Prong 2 above. As an ordered whole, the claim is directed to a mentally performable process of determining a structure of a machine-learning model and generating synthetic spatiotemporal data. Nothing in the claim provides significantly more than this. As such, the claim is not patent eligible.
Claim 9
Step 1: A process, as above.
Step 2A Prong 1: The claim recites, inter alia:
“selecting a column of the raw spatiotemporal data to be learned”; This limitation encompasses mentally selecting a column of the raw spatiotemporal data to be learned.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. No further additional elements are recited, see analysis of claim 8.
Step 2B: The claim does not contain significantly more than the judicial exception. See analysis of claim 8.
Claim 10
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the same judicial exception as claim 9.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The claim further recites “training the machine-learning model while changing a condition value for the column to be learned,” however this limitation amounts to merely generally linking the use of the judicial exception to the technological environment of model training (MPEP 2106.05(h)).
Step 2B: The claim does not contain significantly more than the judicial exception. The analysis at this step mirrors that of Step 2A Prong 2 above.
Claim 11
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the same judicial exception as claim 9.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The claim further recites “storing metadata corresponding to training of the machine-learning model,” however this limitation amounts to the insignificant extra-solution activity of mere data gathering and outputting (MPEP 2106.05(g)).
Step 2B: The claim does not contain significantly more than the judicial exception. The storing metadata limitation, in addition to reciting insignificant extra-solution activity, is also directed to the well-understood, routine, and conventional activity of storing and retrieving information in memory (MPEP 2106.05(d)(iv) Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015); OIP Techs., 788 F.3d at 1363, 115 USPQ2d at 1092-93).
Claim 12
Step 1: A process, as above.
Step 2A Prong 1: The claim recites the same judicial exception as claim 9.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The claim further recites “wherein the metadata includes information about the learned raw spatiotemporal data, information about a condition for the column, and information about the structure of the machine-learning model,” however this limitation amounts to the insignificant extra-solution activity of mere data gathering and outputting (MPEP 2106.05(g)).
Step 2B: The claim does not contain significantly more than the judicial exception. The metadata limitation, in addition to reciting insignificant extra-solution activity, is also directed to the well-understood, routine, and conventional activity of storing and retrieving information in memory (MPEP 2106.05(d)(iv) Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015); OIP Techs., 788 F.3d at 1363, 115 USPQ2d at 1092-93).
Claim 13
Step 1: The claim recites a method, and therefore is directed to the statutory category of processes.
Step 2A Prong 1: The claim recites, inter alia:
“analyzing a predictive spatiotemporal query of a user, thereby extracting information about target data and columns to be queried”; This limitation encompasses mentally analyzing a predictive spatiotemporal query of a user by extracting information about target data and columns to be queries.
“determining whether synthetic spatiotemporal data and a trained machine-learning model are present based on the information about the target data and columns to be queried”; This limitation encompasses mentally determining whether synthetic spatiotemporal data and a trained machine-learning model are present based on the information about the target data and columns to be queried.
“calculating a result value for the predictive spatiotemporal query based on the synthetic spatiotemporal data”; This limitation encompasses mentally calculating a result value for the predictive spatiotemporal query based on the synthetic spatiotemporal data.
“adjusting the result value”; This limitation encompasses mentally adjusting the result value.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The claim does not contain any additional elements beyond the judicial exception.
Step 2B: The claim does not contain significantly more than the judicial exception. The analysis at this step mirrors that of Step 2A Prong 2. As an ordered whole, the claim is directed to a mentally performable process of analyzing a predictive spatiotemporal query, determining if synthetic spatiotemporal data and a trained machine-learning model are present, calculating a result value for the query based on the synthetic data, and adjusting the result value. Nothing in the claim provides significantly more than this. As such, the claim is not patent eligible.
Claim 14
Step 1: A process, as above.
Step 2A Prong 1: The claim recites, inter alia:
“wherein the synthetic spatiotemporal data is generated based on raw spatiotemporal data…”; This limitation encompasses mentally generating synthetic spatiotemporal data based on raw spatiotemporal data.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The claim further recites that the raw spatiotemporal data is “stored in the form of a table including an identifier column and a position column,” however this limitation amounts to the insignificant extra-solution activity of mere data gathering and outputting (MPEP 2106.05(g)).
Step 2B: The claim does not contain significantly more than the judicial exception. The “raw spatiotemporal data stored in a form of a table” limitation, in addition to being insignificant extra solution activity, is also directed to the well-understood, routine, and conventional activity of storing and retrieving information in memory (MPEP 2106.05(d)(II)(iv) Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015); OIP Techs., 788 F.3d at 1363, 115 USPQ2d at 1092-93).
Claim 15
Step 1: A process, as above.
Step 2A Prong 1: The claim recites, inter alia:
“when synthetic data corresponding to the target data and columns to be queried is not present, determining whether a machine-learning model corresponding to the target data and columns to be queried is present”; This limitation encompasses mentally determining whether a machine-learning model corresponding to the target data and columns to be queried is present when synthetic data corresponding to the target data and columns to be queried is not present.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. See analysis of claim 14.
Step 2B: The claim does not contain significantly more than the judicial exception. See analysis of claim 14.
Claim 16
Step 1: A process, as above.
Step 2A Prong 1: The claim recites, inter alia:
“when the synthetic data corresponding to the target data and columns to be queried is not present but the machine-learning model corresponding thereto is present, generating synthetic data corresponding to the target data and columns…”; This limitation encompasses mentally generating synthetic data corresponding to the target data and columns when the synthetic data corresponding to the target data and columns to be queried is not present but the machine-learning model corresponding thereto is present.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. The claim further recites that the generating of synthetic data corresponding to the target data and columns is “based on the machine-learning model,” however this limitation amounts to mere instructions to apply a judicial exception using a generic computer programmed with a generic class of computer algorithms (MPEP 2106.05(f)).
Step 2B: The claim does not contain significantly more than the judicial exception. The analysis at this step mirrors that of Step 2A Prong 2 above.
Claim 17
Step 1: A process, as above.
Step 2A Prong 1: The claim recites, inter alia:
“adjusting the result value using a difference between the synthetic spatiotemporal data and the raw spatiotemporal data”; This limitation encompasses mentally adjusting the result value using a difference between the synthetic spatiotemporal data and the raw spatiotemporal data.
Step 2A Prong 2: This judicial exception is not integrated into a practical application. See analysis of claim 14.
Step 2B: The claim does not contain significantly more than the judicial exception. See analysis of claim 14.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1 and 2 are rejected under 35 U.S.C. 103 as being unpatentable over Segars et al. (US20240160897) (hereinafter “Segars”) in view of Hayashi et al. (US20130339371) (hereinafter “Hayashi”), further in view of Tong (US11429893).
Regarding claim 1, Segars discloses “An apparatus for processing a predictive spatiotemporal query based on synthetic data, comprising:
a query-processing unit for analyzing a… query of a user and returning a processing result (Segars, [0057]: “The query transformation module 120 may function to transform a received data synthetization query into a format and/or data structure that can be processed by one or more generative models (e.g., the generative models stored at the model training and storage module 140). In some examples, to transform a data synthetization query into a format and/or data structure that can be processed by a generative model, the query transformation module 120 may function to extract (or parse) attributes/properties of the received data synthetization query and, in turn, add/embed the extracted attributes/properties into a suitable model input data structure” and [0058]: “Additionally, or alternatively, in some embodiments, to construct the model input data structure, the query transformation module 120 may function to collect (or derive) data related to a user who submitted the data synthetization query (“user context data”) and, in turn, add/embed the collected user context data in the model input data structure”; Examiner notes that “query transformation module” corresponds to a “query processing unit,” “data synthetization query” corresponds to a “query of a user” and “model input data structure” corresponds to a “processing result”);
a machine-learning unit for training a machine-learning model… and generating synthetic… data based on the machine-learning model (Segars, [0065]: “The model training and storage module 140 may function to train one or more generative models based on one or more training datasets sourced by the data sourcing module 150 and/or may comprise one or more on-premise or cloud-based databases that store data related to the training of the one or more generative models (e.g., model weights, model parameters, model metrics, model architecture, and/or the like)” and [0091]: “S225, which includes generating synthetic data samples, may function to generate one or more distinct corpora of synthetic data samples using the one or more generative models trained in S220 (as generally shown in FIG. 5)”; Examiner notes that “model training and storage module” corresponds to a “machine-learning unit” and one of the “one or more generative models” corresponds to a “machine-learning model”); and
a data storage unit for storing raw… data and the generated synthetic… data” (Segars, [0093]: “Additionally, or alternatively, S225 may function to store, in memory, each distinct corpus of synthetic data samples generated by each distinct generative model in association with the corpus of (real) data samples used to train the distinct generative model”; Examiner notes that memory corresponds to a “data storage unit,” “real data samples” corresponds to “raw data”, and “synthetic data samples” corresponds to “the generated synthetic data”).
Segars does not appear to explicitly disclose the further limitations of the claim.
However, Hayashi discloses a “spatiotemporal query” and “spatiotemporal data” (Hayashi, [0118]: “First, when spatio-temporal data search conditions are input, the spatio-temporal data search module 124 creates a query sentence for the search (S201). In the processing of creating the query sentence, the spatio-temporal management table 111 is referred to and the spatio-temporal IDs of the spatio-temporal segments consistent with the search conditions are identified, to thereby create the query sentence including the identified spatio-temporal IDs. The processing of creating the query sentence is described in detail with reference to FIG. 13”) and “spatiotemporal data is stored in a form of a table including an identifier column and a position column” (Hayashi, [0147]: “As illustrated in FIG. 17A, the spatio-temporal data (point time series data) includes the fields of time, space (coordinate values), and object ID. In the "space" column in the table illustrated in FIG. 17A, coordinate values of a position at which an object has been located are written. The object ID is an identifier for uniquely identifying the object”).
Hayashi and the instant application both relate to spatiotemporal data and are analogous. It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the claimed invention, to have modified Segars with the teachings of Hayashi such that the query is a spatiotemporal query, the synthetic and raw data is spatiotemporal, and “wherein the raw spatiotemporal data is stored in a form of a table including an identifier column and a position column,” and one would have been motivated to do so for the purpose of speeding up search processing with conditions of time and space (see Hayashi, [0002]).
Neither Segars nor Hayashi appear to explicitly disclose the further limitations of the claim.
However, Tong discloses a “predictive… query” (Tong, Col 3, lines 1-9: “As indicated herein, embodiments improve on these approaches via a massively parallel real-time inference database engine that can integrated with cloud-based databases (or other large-scale computing environments) to allow users to use standard query interfaces to query ML models to answer predictive type questions. Embodiments thus ultimately enable the ability for users to query their own data warehouse or data lake through a database engine and directly obtain predictive query results”) and “training a machine-learning model in response to a request from… [a user device]” (Tong, Col 11, lines 64-65: “The user devices 602 can interact with the model training system 620 via frontend 629 of the model training system 620. For example, a user device 602 can provide a training request to the frontend 629…” and Col 12, lines 58-60: “The model training system 620 can use the information provided by the user device 602 to train a machine learning model…”).
Tong and the instant application both relate to machine learning and are analogous. It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the claimed invention, to have modified the combination of Segars/Hayashi such that the spatiotemporal query is a “predictive spatiotemporal query” and the training occurs “in response to a request from the query-processing unit,” and one would have been motivated to do so for the purpose of supporting real-time inference using data of a database (see Tong, Col 2, lines 9-27).
Regarding claim 2, the rejection of claim 1 is incorporated. Segars as modified by Hayashi and Tong further discloses “wherein the machine-learning unit selects a column of the raw spatiotemporal data to be learned” (Segars, [0087]: “Furthermore, it shall be noted that, when compared against one another, the generative models trained by S220 may differ in architecture, training parameters (“hyperparameters”), be trained on different columns of the same training data (“column subsetting”), be trained on different rows of the same training data, and/or differ by other training conditions/techniques”).
Claims 3 and 4 are rejected under 35 U.S.C. 103 as being unpatentable over Segars in view of Hayashi and Tong, and further in view of Xu et al. (“Modeling Tabular Data using Conditional GAN”) (hereinafter “Xu”).
Regarding claim 3, the rejection of claim 2 is incorporated. Neither Segars nor Hayashi nor Tong appear to explicitly disclose the further limitations of the claim.
However, Xu discloses “train…[ing a] machine learning model while changing a condition value for… [a] column…” (Xu, 4.3 Conditional Generator and Training-by-Sampling: “Specifically, the goal is to resample efficiently in a way that all the categories from discrete attributes are sampled evenly (but not necessary uniformly) during the training process, and to recover the (not-resampled) real data distribution during test. Let k∗ be the value from the i∗th discrete column Di∗ that has to be matched by the generated samples ˆr, then the generator can be interpreted as the conditional distribution of rows given that particular value at that particular column” and Condition vector: “We introduce the vector cond as the way for indicating the condition (Di∗ = k∗)… For instance, for two discrete columns, D1 = {1,2,3} and D2 = {1,2}, the condition (D2 = 1) is expressed by the mask vectors m1 = [0, 0, 0] and m2 = [1,0]; so cond = [0,0,0,1,0]” and Training-by-sampling: “2. Randomly select a discrete column Di out of all the Nd discrete columns… 4. Let k∗ be a randomly selected value according to the PMF above… 6. Calculate the vector cond”; Examiner notes that vector cond corresponds to a condition value for a column, and resampling during the training process corresponds to training a machine learning model while changing the condition value).
Xu and the instant application both relate to machine learning and are analogous. It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the claimed invention, to have modified the combination of Segars/Hayashi/Tong with the teachings of Xu such that the machine-learning unit trains the machine-learning model while changing a condition value for the column to be learned, and one would have been motivated to do so for the purpose of improving the performance of synthetic tabular data generation (see Xu, Abstract).
Regarding claim 4, the rejection of claim 2 is incorporated. Segars as modified by Hayashi and Tong further discloses “wherein the machine-learning unit stores metadata corresponding to training of the machine-learning model, and the metadata includes information about the learned raw spatiotemporal data, information about a… [data type] for the column, and information about a structure of the machine-learning model” (Segars, [0065]: “The model training and storage module 140 may function to train one or more generative models based on one or more training datasets sourced by the data sourcing module 150 and/or may comprise one or more on-premise or cloud-based databases that store data related to the training of the one or more generative models (e.g., model weights, model parameters, model metrics, model architecture, and/or the like)” and [0080]: “In one non-limiting example, pre-processing the corpus of training data may include collecting, from a subscriber, metadata information related to each data field (e.g., column) in the corpus of training data. For instance, in a non-limiting example, S210 may function to receive one or more user inputs and/or one or more data structures, such as a .csv file or the like, indicating the data type of each data field or column in the corpus of training data (e.g., whether a respective data field relates to datetime data, string data, unique identifier data, currency data, numerical data, location data, Boolean data, or the like)”; Examiner notes that “model parameters” correspond to information about the learned data, “model architecture” corresponds to information about the structure of the machine learning model, and the data type of the column to be learned corresponds to “information about a data type for the column”).
Neither Segars nor Hayashi nor Tong appear to explicitly disclose information about a “condition” for the column.
However, Xu discloses “a condition for… [a] column” (Xu, 4.3 Conditional Generator and Training-by-Sampling, Condition vector: “We introduce the vector cond as the way for indicating the condition (Di∗ = k∗)…For instance, for two discrete columns, D1 = {1,2,3} and D2 = {1,2}, the condition (D2 = 1) is expressed by the mask vectors m1 = [0, 0, 0] and m2 = [1,0]; so cond = [0,0,0,1,0]”).
Xu and the instant application both relate to machine learning and are analogous. It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the claimed invention, to have modified the combination of Segars/Hayashi/Tong with the teachings of Xu such that the metadata includes information about a condition for the column rather than a data type for the column, and one would have been motivated to do so for the purpose of improving the performance of synthetic tabular data generation (see Xu, Abstract).
Claims 5-7 are rejected under 35 U.S.C. 103 as being unpatentable over Segars in view of Hayashi and Tong, further in view of Hendawi et al. (“Panda*: A generic and scalable framework for predictive spatio-temporal queries”) (hereinafter “Hendawi”) and Pratik et al. (US11675817) (hereinafter “Pratik”).
Regarding claim 5, the rejection of claim 1 is incorporated. Segars as modified by Hayashi and Tong further discloses “wherein: the query-processing unit analyses the predictive spatiotemporal query of the user, thereby extracting information about target…columns to be queried (Segars, [0057]: “In some examples, to transform a data synthetization query into a format and/or data structure that can be processed by a generative model, the query transformation module 120 may function to extract (or parse) attributes/properties of the received data synthetization query and, in turn, add/embed the extracted attributes/properties into a suitable model input data structure. Example attributes/properties that may be extracted from a data synthetization query may include, but should not be limited to, the name of the generative database being queried (“target generative database”), the name of the generative database table being queried (“target generative database table”), the database fields (e.g., columns) to synthesize, the number of data samples (e.g., rows) to synthesize, the conditions of the data synthetization query (e.g., WHERE clause in an SQL query), and/or the like” and [0135]: “For instance, in a non-limiting example, the method 200 (e.g., via S250) may receive a generative database query such as ‘SELECT u.age, u.zip, u.income FROM users.’ In turn, based on receiving such generative database query, the method 200 may function to search the model election data matrix for generative models that were trained on data that enables the data fields “age,” “zip,” and “income” to be synthesized”; Examiner notes that “data synthetization query”/”generative database query” (note that these are the same thing, as the data synthetization query is a query of a generative database) corresponds to a query of a user, and “the database fields (e.g. columns) to synthetize” corresponds to information about target columns to be queried), and
the machine-learning unit determines whether… a trained machine-learning model… [is] present based on the information about the target…columns to be queried” (Segars, [0088]: “Accordingly, if a generative database query received by S250 requests any combination of the specific data fields used during a training of one or particular sets of generative models, the method 200 may determine that these one or more particular sets of generators are capable of fulfilling such generative database query. However, if the generative database query received by S240 extends beyond the specific fields used during a training of one or more particular sets of generative models, these one or more particular set of generators may not be applicable to a subject generative database query as such generative models are not capable of synthesizing such data”; Examiner notes that data fields correspond to columns, as shown above).
Neither Segars nor Hayashi nor Tong appear to explicitly disclose the further limitations of the claim.
However, Hendawi discloses “analyz…[ing a] predictive spatiotemporal query of… [a] user, thereby extracting information about target data… to be queried (Hendawi, Conclusion: “This paper introduces Panda∗; a system for evaluating predictive spatio-temporal queries. Panda∗ enables users to request a location-based service using the predicted locations of moving objects in a future time instance” and 4.2 Generic Query processing in Panda: “Upon the arrival of a new predictive spatio-temporal query Q, with an area of interest R, requesting a prediction about future time t, Panda∗ first divides Q into a sets of grid cells Cf that overlap with the query region of interest R”), and
…return…[ing] a result value for the predictive spatiotemporal query based on… synthetic spatiotemporal data (Hendawi, 4.2.1 Phase 1: result computation: “Phase I, result computation, receives a predictive query Q, either as range, aggregate, or knearest-neighbor, asking about future time t and a cell ci that overlaps with the query region of interest R. The output of this phase is the partial answer of Q computed from ci” and 6.1 Experiment setup: In our performance evaluation experiments, we use two data sets. Synthetic data We use the Network-based Generator of Moving Objects [5] to generate large sets of synthetic data of moving objects”).
Hendawi and the instant application both relate to predictive spatiotemporal queries and are analogous. It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the claimed invention, to have modified the combination of Segars/Hayashi/Tong with the teachings of Hendawi such that the query-processing unit analyzes the predictive spatiotemporal query of the user, thereby extracting information about target data and columns to be queried, and the machine-learning unit returns a result value for the predictive spatiotemporal query based on the synthetic spatiotemporal data, and one would have been motivated to do so for the purpose of implementing a scalable, efficient, and accurate predictive spatiotemporal query processor (see Hendawi, Introduction).
Neither Segars, Hayashi, Tong, nor Hendawi appear to explicitly disclose the further limitations of the claim.
However, Pratik discloses “determining whether synthetic…data… [is] present based on… information about… target data… to be… [synthesized]” (Pratik, Col 6, lines 1-6: “As used herein, the terms “configuration data parameter” may refer to a specification pertaining to the request for one or more synthetic datasets. In some embodiments, a configuration data parameter may, at least in part, set one or more conditions for the generation of one or more synthetic datasets” and Pratik, Col 19, lines 8-13: “Referring first to FIG. 5A, as shown in operation 501, the apparatus (e.g., synthetic data generator computing entity 106) includes means, such as processing element 205, volatile memory 215, non-volatile memory 210, or the like, for determining whether the one or more configuration data parameters are indicative of one or more base datasets” and Col 18, lines 51-63: “In some embodiments, the one or more generated synthetic datasets may be stored in an associated memory, such as storage subsystem 108… The one or more generated synthetic datasets may also be stored with the one or more corresponding configuration data parameters. In this way, synthetic data generator computing entity 106 may utilize the one or more corresponding configuration data parameters when determining whether one or more base datasets with similar configuration data parameters exist for future requests for synthetic data”; Examiner notes that “configuration data parameters” corresponds to “information about target data to be synthesized,” a base dataset comprising a previously-generated synthetic dataset corresponds to “synthetic data,” and determining whether the configuration data parameters are indicative of one or more base datasets corresponds to “determining whether synthetic data is present based on information about target data to be synthesized”).
Pratik and the instant application both relate to machine learning for synthetic data generation and are analogous. It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the claimed invention, to have modified the combination of Segars/Hayashi/Tong/Hendawi with the teachings of Pratik such that the machine-learning unit “determines whether synthetic spatiotemporal data and a trained machine-learning model are present based on the information about the target data and columns to be queried,” and one would have been motivated to do so for the purpose of utilizing previously generated synthetic data to fulfill future requests for synthetic data (see Pratik, Col 18, lines 51-63), thereby increasing the amount/variety of base data available.
Regarding claim 6, the rejection of claim 5 is incorporated. Segars as modified by Hayashi, Tong, Hendawi, and Pratik further discloses “wherein, when synthetic data corresponding to the target data and columns to be queried is not present, the machine-learning unit determines whether a machine-learning model corresponding to the target data and columns to be queried is present” (Segars, [0088]: “Accordingly, if a generative database query received by S250 requests any combination of the specific data fields used during a training of one or particular sets of generative models, the method 200 may determine that these one or more particular sets of generators are capable of fulfilling such generative database query. However, if the generative database query received by S240 extends beyond the specific fields used during a training of one or more particular sets of generative models, these one or more particular set of generators may not be applicable to a subject generative database query as such generative models are not capable of synthesizing such data”; Examiner notes that synthetic data corresponding to the target data and columns to be queried is not present as it has not yet been generated).
Regarding claim 7, the rejection of claim 5 is incorporated. Segars as modified by Hayashi, Tong, Hendawi, and Pratik further discloses “wherein, when synthetic data corresponding to the target data and columns to be queried is not present but a machine-learning model corresponding thereto is present, the machine-learning unit generates synthetic data corresponding to the target data and columns based on the machine-learning model” (Segars, [0091]: “S225, which includes generating synthetic data samples, may function to generate one or more distinct corpora of synthetic data samples using the one or more generative models trained in S220 (as generally shown in FIG. 5)” and [0135]: “For instance, in a non-limiting example, the method 200 (e.g., via S250) may receive a generative database query such as “SELECT u.age, u.zip, u.income FROM users.” In turn, based on receiving such generative database query, the method 200 may function to search the model election data matrix for generative models that were trained on data that enables the data fields “age,” “zip,” and “income” to be synthesized. The generative models that were not trained to generate these data fields may be deemed unsuitable and not considered for selection. Conversely, the suitable models trained on these data fields may then be ranked in one or more ways described herein (e.g., based on their performance on various synthetization tasks), and the model with optimal performance may be selected to fulfill the generative database query”; Examiner notes that synthetic data corresponding to the target data and columns to be queried is not present as it has not yet been generated).
Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Pratik et al. (US11675817) (hereinafter “Pratik”) in view of Hayashi et al. (US20130339371) (hereinafter “Hayashi”).
Regarding claim 8, Pratik discloses “A method for generating synthetic spatiotemporal data, comprising:
determining a structure of a machine-learning model for generating synthetic…data (Pratik, Col 6 line 60- Col 7 line 24: “The terms “machine learning generation model” may refer to an electronically-stored data construct that is configured to describe parameters, hyper-parameters, and/or stored operations of a machine learning generation model that is configured to process one or more base datasets and/or one or more configuration data parameters in order to generate one or more synthetic datasets…By way of a nonlimiting example, a number of nodes in a neural network, a number of generator training runs for each associated training run of a discriminator, a loss and optimizer function, one or more early stopping criteria, quality metrics, and/or the like may be configured based upon iterative operation of the one or more machine learning generation models. In some embodiments, the parameters and/or hyper-parameters of a machine learning generation model may be represented as values in an n by n dimensional array, such as a matrix. In some embodiments, the number of parameters and/or hyperparameters may be determined based at least in part on the size of the one or more base datasets and/or one or more configuration data parameters”);
training the machine-learning model based on raw… data (Pratik, Col 10, lines 29-45: “In some embodiments, the synthetic data generator computing entity 106 may be configured to train one or more processing models based at least in part on the training data store 122 stored in the storage subsystem 108 and store the one or more trained processing models as part of the model definition data store 121 stored in the storage subsystem 108…The external computing entity 102 (e.g., a management computing entity or the like) may periodically update/provide raw input data (e.g., real and/or synthetic generated datasets) to the synthetic data generator system 101 to serve as training data, which may be stored in the training data store 122 and/or storage subsystem 108” and Col 17, lines 44-46: “In some embodiments, the one or more processing models may include…one or more machine learning generation models”); and
generating synthetic…data based on the machine-learning model” (Pratik, Col 18, lines 1-8: “Thereafter, as shown in operation 403, the apparatus (e.g., synthetic data generator computing entity 106, includes means, such as processing element 205, rules-based generation engine 110, machine learning generation engine 112, noise generation engine 114, and/or obfuscation generation engine 116, and/or the like, for generating one or more synthetic datasets comprising one or more synthetic data values via the selected at least one processing model”).
Pratik does not appear to explicitly disclose the further limitations of the claim.
However, Hayashi discloses “spatiotemporal data” and “spatiotemporal data is stored in a form of a table including an identifier column and a position column” (Hayashi, [0147]: “As illustrated in FIG. 17A, the spatio-temporal data (point time series data) includes the fields of time, space (coordinate values), and object ID. In the "space" column in the table illustrated in FIG. 17A, coordinate values of a position at which an object has been located are written. The object ID is an identifier for uniquely identifying the object”).
Hayashi and the instant application both relate to spatiotemporal data and are analogous. It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the claimed invention, to have modified Pratik with the teachings of Hayashi such that the synthetic and raw data is spatiotemporal data, and “wherein the raw spatiotemporal data is stored in a form of a table including an identifier column and a position column,” and one would have been motivated to do so for the purpose of speeding up search processing with conditions of time and space (see Hayashi, [0002]).
Claims 9 and 11 are rejected under 35 U.S.C. 103 as being unpatentable over Pratik in view of Hayashi, and further in view of Segars et al. (US20240160897) (hereinafter “Segars”).
Regarding claim 9, the rejection of claim 8 is incorporated. Neither Pratik nor Hayashi appear to explicitly disclose the further limitations of the claim.
However, Segars discloses “wherein training… [a] machine-learning model comprises selecting a column of the raw spatiotemporal data to be learned” (Segars, [0087]: “Furthermore, it shall be noted that, when compared against one another, the generative models trained by S220 may differ in architecture, training parameters (“hyperparameters”), be trained on different columns of the same training data (“column subsetting”), be trained on different rows of the same training data, and/or differ by other training conditions/techniques”).
Segars and the instant application both relate to machine learning for generating synthetic data and are analogous. It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the claimed invention, to have modified the combination of Pratik/Hayashi with the teachings of Segars such that the training comprises selecting a column of the raw spatiotemporal data to be learned, and one would have been motivated to do so for the purpose of efficiently handling/processing queries relating to a plurality of different data synthetization intents and a plurality of different synthetization queries (see Segars, [0086]).
Regarding claim 11, the rejection of claim 9 is incorporated. Pratik as modified by Hayashi and Segars further discloses “storing metadata corresponding to training of the machine-learning model” (Segars, [0065]: “The model training and storage module 140 may function to train one or more generative models based on one or more training datasets sourced by the data sourcing module 150 and/or may comprise one or more on-premise or cloud-based databases that store data related to the training of the one or more generative models (e.g., model weights, model parameters, model metrics, model architecture, and/or the like)”).
Segars and the instant application both relate to machine learning for generating synthetic data and are analogous. It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the claimed invention, to have modified the combination of Pratik/Hayashi with the teachings of Segars to include “storing metadata corresponding to training of the machine-learning model,” and one would have been motivated to do so for the purpose of efficiently handling/processing queries relating to a plurality of different data synthetization intents and a plurality of different synthetization queries (see Segars, [0086]).
Claims 10 and 12 are rejected under 35 U.S.C. 103 as being unpatentable over Pratik in view of Hayashi and Segars, and further in view of Xu et al. (“Modeling Tabular Data using Conditional GAN”) (hereinafter “Xu”).
Regarding claim 10, the rejection of claim 9 is incorporated. Neither Pratik nor Hayashi nor Segars appear to explicitly disclose the further limitation of the claim.
However, Xu discloses “training… [a] machine learning model while changing a condition value for… [a] column…” (Xu, 4.3 Conditional Generator and Training-by-Sampling: “Specifically, the goal is to resample efficiently in a way that all the categories from discrete attributes are sampled evenly (but not necessary uniformly) during the training process, and to recover the (not-resampled) real data distribution during test. Let k∗ be the value from the i∗th discrete column Di∗ that has to be matched by the generated samples ˆr, then the generator can be interpreted as the conditional distribution of rows given that particular value at that particular column” and Condition vector: “We introduce the vector cond as the way for indicating the condition (Di∗ = k∗)… For instance, for two discrete columns, D1 = {1,2,3} and D2 = {1,2}, the condition (D2 = 1) is expressed by the mask vectors m1 = [0, 0, 0] and m2 = [1,0]; so cond = [0,0,0,1,0]” and Training-by-sampling: “2. Randomly select a discrete column Di out of all the Nd discrete columns… 4. Let k∗ be a randomly selected value according to the PMF above… 6. Calculate the vector cond”; Examiner notes that vector cond corresponds to a condition value for a column, and resampling during the training process corresponds to training a machine learning model while changing the condition value).
Xu and the instant application both relate to machine learning and are analogous. It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the claimed invention, to have modified the combination of Pratik/Hayashi/Segars with the teachings of Xu such that the machine-learning unit trains the machine-learning model while changing a condition value for the column to be learned, and one would have been motivated to do so for the purpose of improving the performance of synthetic tabular data generation (see Xu, Abstract).
Regarding claim 12, the rejection of claim 11 is incorporated. Pratik as modified by Hayashi and Segars further discloses “wherein the metadata includes information about the learned raw spatiotemporal data, information about a… [data type] for the column, and information about the structure of the machine-learning model” (Segars, [0065]: “The model training and storage module 140 may function to train one or more generative models based on one or more training datasets sourced by the data sourcing module 150 and/or may comprise one or more on-premise or cloud-based databases that store data related to the training of the one or more generative models (e.g., model weights, model parameters, model metrics, model architecture, and/or the like)” and [0080]: “In one non-limiting example, pre-processing the corpus of training data may include collecting, from a subscriber, metadata information related to each data field (e.g., column) in the corpus of training data. For instance, in a non-limiting example, S210 may function to receive one or more user inputs and/or one or more data structures, such as a .csv file or the like, indicating the data type of each data field or column in the corpus of training data (e.g., whether a respective data field relates to datetime data, string data, unique identifier data, currency data, numerical data, location data, Boolean data, or the like)”; Examiner notes that “model parameters” correspond to information about the learned data, “model architecture” corresponds to information about the structure of the machine learning model, and the data type of the column used for training the model corresponds to “information about a data type for the column”).
Segars and the instant application both relate to machine learning for generating synthetic data and are analogous. It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the claimed invention, to have modified the combination of Pratik/Hayashi with the teachings of Segars such that the metadata includes information about the learned raw spatiotemporal data, information about a data type for the column, and information about the structure of the machine-learning model, and one would have been motivated to do so for the purpose of efficiently handling/processing queries relating to a plurality of different data synthetization intents and a plurality of different synthetization queries (see Segars, [0086]).
Neither Pratik nor Hayashi nor Segars appear to explicitly disclose information about a “condition” for the column.
However, Xu discloses “a condition for… [a] column” (Xu, 4.3 Conditional Generator and Training-by-Sampling, Condition vector: “We introduce the vector cond as the way for indicating the condition (Di∗ = k∗)…For instance, for two discrete columns, D1 = {1,2,3} and D2 = {1,2}, the condition (D2 = 1) is expressed by the mask vectors m1 = [0, 0, 0] and m2 = [1,0]; so cond = [0,0,0,1,0]”).
Xu and the instant application both relate to machine learning and are analogous. It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the claimed invention, to have modified the combination of Pratik/Hayashi/Segars with the teachings of Xu such that the metadata includes information about a condition for the column instead of a data type for the column, and one would have been motivated to do so for the purpose of improving the performance of synthetic tabular data generation (see Xu, Abstract).
Claim 13 is rejected under 35 U.S.C. 103 as being unpatentable over Hendawi et al. (“Panda*: A generic and scalable framework for predictive spatio-temporal queries”) (hereinafter “Hendawi”) in view of Segars et al. (US20240160897) (hereinafter “Segars”), and further in view of Pratik et al. (US11675817) (hereinafter “Pratik”).
Regarding claim 13, Hendawi discloses “A method for processing a predictive spatiotemporal query based on synthetic data, comprising:
analyzing a predictive spatiotemporal query of a user, thereby extracting information about target data… to be queried… (Hendawi, Conclusion: “This paper introduces Panda∗; a system for evaluating predictive spatio-temporal queries. Panda∗ enables users to request a location-based service using the predicted locations of moving objects in a future time instance” and 4.2 Generic Query processing in Panda: “Upon the arrival of a new predictive spatio-temporal query Q, with an area of interest R, requesting a prediction about future time t, Panda∗ first divides Q into a sets of grid cells Cf that overlap with the query region of interest R”)
calculating a result value for the predictive spatiotemporal query based on… synthetic spatiotemporal data (Hendawi, 4.2.1 Phase 1: result computation: “Phase I, result computation, receives a predictive query Q, either as range, aggregate, or knearest-neighbor, asking about future time t and a cell ci that overlaps with the query region of interest R. The output of this phase is the partial answer of Q computed from ci” and 6.1 Experiment setup: In our performance evaluation experiments, we use two data sets. Synthetic data We use the Network-based Generator of Moving Objects [5] to generate large sets of synthetic data of moving objects”); and
adjusting the result value (Hendawi, 4.2.1 Phase 1: result computation, Algorithm 1, lines 3 and 16; Examiner notes QueryResult is updated by CellResult for each cell ci in Cf, thus the partial answer of Q computed from ci (result value) is adjusted when CellResult for cell ci+1 is computed).
Hendawi does not appear to explicitly disclose the further limitations of the claim.
However, Segars discloses
“analyzing a… query of a user, thereby extracting information about target… columns to be queried (Segars, [0057]: “In some examples, to transform a data synthetization query into a format and/or data structure that can be processed by a generative model, the query transformation module 120 may function to extract (or parse) attributes/properties of the received data synthetization query and, in turn, add/embed the extracted attributes/properties into a suitable model input data structure. Example attributes/properties that may be extracted from a data synthetization query may include, but should not be limited to, the name of the generative database being queried (“target generative database”), the name of the generative database table being queried (“target generative database table”), the database fields (e.g., columns) to synthesize, the number of data samples (e.g., rows) to synthesize, the conditions of the data synthetization query (e.g., WHERE clause in an SQL query), and/or the like” and [0135]: “For instance, in a non-limiting example, the method 200 (e.g., via S250) may receive a generative database query such as ‘SELECT u.age, u.zip, u.income FROM users.’ In turn, based on receiving such generative database query, the method 200 may function to search the model election data matrix for generative models that were trained on data that enables the data fields “age,” “zip,” and “income” to be synthesized”; Examiner notes that “data synthetization query”/”generative database query” (note that these are the same thing, as the data synthetization query is a query of a generative database) corresponds to a query of a user and “the database fields (e.g. columns) to synthetize” corresponds to information about target columns to be queried);
determining whether… a trained machine-learning model… [is] present based on the information about the target…columns to be queried… (Segars, [0088]: “Accordingly, if a generative database query received by S250 requests any combination of the specific data fields used during a training of one or particular sets of generative models, the method 200 may determine that these one or more particular sets of generators are capable of fulfilling such generative database query. However, if the generative database query received by S240 extends beyond the specific fields used during a training of one or more particular sets of generative models, these one or more particular set of generators may not be applicable to a subject generative database query as such generative models are not capable of synthesizing such data”; Examiner notes that data fields correspond to columns, as shown above).
Segars and the instant application both relate to machine learning for synthetic data generation and are analogous. It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the claimed invention, to have modified Hendawi with the teachings of Segars such that the extracted information is “about target data and columns to be queried” and to include “determining whether a trained machine-learning model is present based on the information about the columns to be queried,” and one would have been motivated to do so for the purpose of efficiently handling/processing queries relating to a plurality of different data synthetization intents and a plurality of different synthetization queries (see Segars, [0086]).
Neither Hendawi nor Segars appear to explicitly disclose the further limitations of the claim.
However, Pratik discloses “determining whether synthetic…data… [is] present based on… information about… target data… to be… [synthesized]” (Pratik, Col 6, lines 1-6: “As used herein, the terms “configuration data parameter” may refer to a specification pertaining to the request for one or more synthetic datasets. In some embodiments, a configuration data parameter may, at least in part, set one or more conditions for the generation of one or more synthetic datasets” and Pratik, Col 19, lines 8-13: “Referring first to FIG. 5A, as shown in operation 501, the apparatus (e.g., synthetic data generator computing entity 106) includes means, such as processing element 205, volatile memory 215, non-volatile memory 210, or the like, for determining whether the one or more configuration data parameters are indicative of one or more base datasets” and Col 18, lines 51-63: “In some embodiments, the one or more generated synthetic datasets may be stored in an associated memory, such as storage subsystem 108… The one or more generated synthetic datasets may also be stored with the one or more corresponding configuration data parameters. In this way, synthetic data generator computing entity 106 may utilize the one or more corresponding configuration data parameters when determining whether one or more base datasets with similar configuration data parameters exist for future requests for synthetic data”; Examiner notes that “configuration data parameters” corresponds to “information about target data to be synthesized,” a base dataset comprising a previously-generated synthetic dataset corresponds to “synthetic data,” and determining whether the configuration data parameters are indicative of one or more base datasets corresponds to “determining whether synthetic data is present based on information about target data to be synthesized”).
Pratik and the instant application both relate to machine learning for synthetic data generation and are analogous. It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the claimed invention, to have modified the combination of Hendawi/Segars with the teachings of Pratik such that the determining step consists of “determining whether synthetic spatiotemporal data and a trained machine-learning model are present based on the information about the target data and columns to be queried,” and one would have been motivated to do so for the purpose of utilizing previously generated synthetic data to fulfill future requests for synthetic data (see Pratik, Col 18, lines 51-63), thereby increasing the amount/variety of base data available.
Claims 14-17 are rejected under 35 U.S.C. 103 as being unpatentable over Hendawi in view of Segars and Pratik, and further in view of Hayashi et al. (US20130339371) (hereinafter “Hayashi”).
Regarding claim 14, the rejection of claim 13 is incorporated. Hendawi as modified by Segars and Pratik further discloses “wherein the synthetic spatiotemporal data is generated based on raw spatiotemporal data…” (Segars, [0091]: “S225, which includes generating synthetic data samples, may function to generate one or more distinct corpora of synthetic data samples using the one or more generative models trained in S220 (as generally shown in FIG. 5). That is, once the generative models are trained using the sourced corpus of training data sourced in S210, S225 may instantiate or invoke each trained generative model of the target ensemble and, subsequently, cause each model to generate a respective corpus of synthetic data samples (e.g., a synthetic dataset)”; Examiner notes that the sourced corpus of training data corresponds to raw data).
Segars and the instant application both relate to machine learning for synthetic data generation and are analogous. It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the claimed invention, to have modified the combination of Hendawi/Pratik with the teachings of Segars such that the synthetic spatiotemporal data is generated based on raw spatiotemporal data, and one would have been motivated to do so for the purpose of efficiently handling/processing queries relating to a plurality of different data synthetization intents and a plurality of different synthetization queries (see Segars, [0086]).
Neither Hendawi nor Segars nor Pratik appear to explicitly disclose the further limitations of the claim.
However, Hayashi discloses “spatiotemporal data stored in a form of a table including an identifier column and a position column” (Hayashi, [0147]: “As illustrated in FIG. 17A, the spatio-temporal data (point time series data) includes the fields of time, space (coordinate values), and object ID. In the "space" column in the table illustrated in FIG. 17A, coordinate values of a position at which an object has been located are written. The object ID is an identifier for uniquely identifying the object”).
Hayashi and the instant application both relate to spatiotemporal data and are analogous. It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the claimed invention, to have modified the combination of Hendawi/Segars/Pratik with the teachings of Hayashi such that the raw spatiotemporal data is “stored in a form of a table including an identifier column and a position column” and one would have been motivated to do so for the purpose of speeding up search processing with conditions of time and space (see Hayashi, [0002]).
Regarding claim 15, the rejection of claim 14 is incorporated. Hendawi as modified by Pratik, Segars, and Hayashi further discloses “wherein determining whether the synthetic spatiotemporal data and the trained machine-learning model are present comprises, when synthetic data corresponding to the target data and columns to be queried is not present, determining whether a machine-learning model corresponding to the target data and columns to be queried is present” (Segars, [0088]: “Accordingly, if a generative database query received by S250 requests any combination of the specific data fields used during a training of one or particular sets of generative models, the method 200 may determine that these one or more particular sets of generators are capable of fulfilling such generative database query. However, if the generative database query received by S240 extends beyond the specific fields used during a training of one or more particular sets of generative models, these one or more particular set of generators may not be applicable to a subject generative database query as such generative models are not capable of synthesizing such data”; Examiner notes that synthetic data corresponding to the target data and columns to be queried is not present as it has not yet been generated).
Segars and the instant application both relate to machine learning for synthetic data generation and are analogous. It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the claimed invention, to have modified the combination of Hendawi/Pratik/Hayashi with the teachings of Segars such that adjusting the result value comprises adjusting the result value using a difference between the synthetic spatiotemporal data and the raw spatiotemporal data, and one would have been motivated to do so for the purpose of efficiently handling/processing queries relating to a plurality of different data synthetization intents and a plurality of different synthetization queries (see Segars, [0086]).
Regarding claim 16, the rejection of claim 15 is incorporated. Hendawi as modified by Pratik, Segars, and Hayashi further discloses “wherein determining whether the synthetic spatiotemporal data and the trained machine-learning model are present comprises, when the synthetic data corresponding to the target data and columns to be queried is not present but the machine-learning model corresponding thereto is present, generating synthetic data corresponding to the target data and columns based on the machine-learning model” (Segars, [0091]: “S225, which includes generating synthetic data samples, may function to generate one or more distinct corpora of synthetic data samples using the one or more generative models trained in S220 (as generally shown in FIG. 5)” and [0135]: “For instance, in a non-limiting example, the method 200 (e.g., via S250) may receive a generative database query such as “SELECT u.age, u.zip, u.income FROM users.” In turn, based on receiving such generative database query, the method 200 may function to search the model election data matrix for generative models that were trained on data that enables the data fields “age,” “zip,” and “income” to be synthesized. The generative models that were not trained to generate these data fields may be deemed unsuitable and not considered for selection. Conversely, the suitable models trained on these data fields may then be ranked in one or more ways described herein (e.g., based on their performance on various synthetization tasks), and the model with optimal performance may be selected to fulfill the generative database query”; Examiner notes that synthetic data corresponding to the target data and columns to be queried is not present as it has not yet been generated).
Segars and the instant application both relate to machine learning for synthetic data generation and are analogous. It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the claimed invention, to have modified the combination of Hendawi/Pratik/Hayashi with the teachings of Segars such that adjusting the result value comprises adjusting the result value using a difference between the synthetic spatiotemporal data and the raw spatiotemporal data, and one would have been motivated to do so for the purpose of efficiently handling/processing queries relating to a plurality of different data synthetization intents and a plurality of different synthetization queries (see Segars, [0086]).
Regarding claim 17, the rejection of claim 14 is incorporated. Neither Hendawi, Pratik, nor Hayashi appear to explicitly disclose the further limitations of the claim.
However, Segars further discloses “adjusting… [a] result value using a difference between… synthetic…data and…raw…data” (Segars, [0096]: “In a first implementation, S225 may function to perform granular sensitivity testing or evaluation of a given synthetic dataset that evaluates each distinct synthetic data sample for similarity or closeness to real data samples within a corpus of source training data samples. That is, in this first implementation, S225 may perform a similarity analysis that includes scanning or comparing a target synthetic data sample of a synthetic dataset against various real data samples of a corpus of source training data samples. In one embodiment, S225 may produce a resemblance or closeness metric or score based on each pairwise evaluation of the target synthetic data sample and each real data sample of a corpus of source training data samples. Accordingly, if the resemblance or closeness metric satisfies or exceeds a resemblance or similarity threshold (e.g., a maximum similarity value), S225 may jettison the target synthetic data sample from the synthetic dataset”; Examiner notes that the generated synthetic dataset corresponds to a result, and jettisoning a target synthetic data sample from the synthetic dataset based on resemblance to its corresponding real data sample corresponds to adjusting a result value using a difference between synthetic data and raw data).
Segars and the instant application both relate to machine learning for synthetic data generation and are analogous. It would have been obvious to one of ordinary skill in the art, prior to the effective filing date of the claimed invention, to have modified the combination of Hendawi/Pratik/Hayashi with the teachings of Segars such that adjusting the result value comprises adjusting the result value using a difference between the synthetic spatiotemporal data and the raw spatiotemporal data, and one would have been motivated to do so for the purpose of efficiently handling/processing queries relating to a plurality of different data synthetization intents and a plurality of different synthetization queries (see Segars, [0086]).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to GWYNEVERE A DETERDING whose telephone number is (571)272-7657. The examiner can normally be reached Mon-Fri. 9am-5pm.
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, Kamran Afshar can be reached at (571) 272-7796. 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.
/G.A.D./Examiner, Art Unit 2125
/KAMRAN AFSHAR/Supervisory Patent Examiner, Art Unit 2125
Prosecution Insights