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 .
Response to Amendments
The action is responsive to the Applicant’s Amendment filed on 6/26/2025. Claims 1-5, 7-18, and 20-24 are pending in the application. Claims 1, 9, and 15 are amended.
Response to Arguments
Applicant’s arguments with respect to the rejections previously made and the amended claims filed on 6/26/2025 have been fully considered but they are not persuasive. In view of the claim amendments, the rejections are being updated accordingly.
Furthermore, it is also submitted that all limitations in pending claims, including those not specifically argued, are properly addressed. The reason is set forth in the rejections. See claim analysis below for detail.
Claim Objections
Claim 24 is objected to because of the following informalities: Claim 24 appears on the “Applicant Arguments/Remarks” page. The claim or claims must commence on a separate physical sheet or electronic page. Any sheet including a claim or portion of a claim may not contain any other parts of the application or other material. See MPEP 608.01(m).
Appropriate correction is required.
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.
Claims 1, 9, and 15 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. Claims 1, 9, and 15 recite “serializing the structured data content into the formatted structured data content that is formatted in accordance with a second structure dialect different from the first structure dialect, the second structure dialect being in compliance with the known structure properties” that is directed to new matter and fails to comply with the written description requirement.” Applicant’s remarks states that “Support for the amendments to the claims can be found in the Applicant's specification at least at paragraphs [0051]-[0055].” However, the specification does not teach this limitation as recited in claims 1, 9, and 15.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-5, 7-18, and 20-24 are rejected under 35 U.S.C. 103 as being unpatentable over Saurav et al. (US 20180314883 A1, hereinafter Saurav) in view of Gerrit et al. (“Wrangling Messy CSV Files by Detecting Row and Type Patterns”, 27 November 2018).
Regarding Claim 1, Saurav discloses a method of transforming structured data content having unknown structure properties to formatted structured data content in compliance with known structure properties ([Abstract]: Methods, systems, and computer program products for automatic detection of string and column delimiters in tabular data files are provided herein; [0023]: a file for which the correct delimiters are unknown can be parsed with various candidate delimiters), identified by a trained machine learning model (Fig. 2; [0055]-[0064]: By applying machine learning to issues pertaining to automating the data wrangling process, the techniques detailed herein create a robust solution; [0062]: Subsequently, the alert and/or formatted data blocks are transmitted over a data channel to the user's wireless device), the method comprising:
inputting a textual sample of the structured data content including the unknown structure properties into the trained machine learning model (Fig. 1; [0023]: A training file can contain, for example, one row for each combination of (i) a sample data file… Once the model is trained, a file for which the correct delimiters are unknown can be parsed with various candidate delimiters; [0046]: score a candidate delimiter pair (when the system is deployed and applied to files with unknown delimiters)),
the textual sample being formatted in accordance with a first structure dialect ([0023]: A training file can contain… a candidate delimiter pair. Each such row contains one column for each feature used by the linear regression model (see, for example, Table 2 below)), the first structure dialect not being in compliance with the known structure properties ([0023]: Additionally, if the current candidate delimiters are the correct delimiters, such an embodiment can include writing “1” to the training file; otherwise, such an embodiment can include writing “0” to the training file… Once the model is trained, a file for which the correct delimiters are unknown can be parsed with various candidate delimiters),
wherein the trained machine learning model is trained by textual training samples of structured training data content with labeled structure properties corresponding to the unknown structure properties (Fig. 1; [0031]: the database 106, which can store tabular files with known delimiters, is only needed during the training of the model. Thereafter, only a determined set of feature weights needs to be retained for use when processing new files wherein the delimiters are unknown; Fig. 1; [0052]: To train the candidate evaluator 134 (also referred to herein as the classifier), at least one embodiment of the invention utilizes a collection of tabular data files; Fig. 2; [0058]: Step 204 includes… constructing one or more feature vectors corresponding to the detected candidate column delimiters and the detected candidate string delimiters) and
includes a loss function corresponding to each labeled structure property ([0053]: Tools (such as, for example, SciKit-Learn), can be used to train a logistic-regression classifier [Scikit-learn, a Python machine learning library, has several loss functions and metrics for measuring classification performance]; Fig. 5; [0105]: Examples of workloads and functions which may be provided from this layer include:… string and column delimiter detection 96);
predicting labels for the unknown structure properties of the structured data content using the trained machine learning model based on the textual sample (Figs. 1-2; [0054]-[0060]: Once the candidate evaluator 134 (classifier) has been trained, it can be used in one or more embodiments of the invention to analyze tabular data files for which the delimiters are unknown… [0060]: Step 206 includes outputting, to at least one user, the candidate column delimiter-candidate string delimiter pairing having the highest likelihood of validity),
wherein the labels identify known structure properties of the structured data content (Fig. 1; [0046]: train the linear regression model (during training on files with known delimiters); [0047]: Further, the candidate evaluator 134 uses a logistic-regression classifier… and the candidate pair generator 136 identifies the candidate pair with the highest score as the one most likely to be the correct choice; [0062]: an alert is generated containing one or more identified string delimiter and column delimiter pairs pertaining to the tabular data files);
parsing elements of the structured data content based on the known structure properties corresponding to the predicted labels (Fig.1; [0046]-[0049]: Additionally, as detailed herein, the file parser 132 parses (or partitions) the tabular data file in question using each selected candidate pair… As noted above, the candidate selector 120 uses one or more heuristics to select candidate string and column delimiter pairs);
However, Saurav does not explicitly teach “serializing the structured data content into formatted structured data content, wherein the formatted structured data content is formatted in compliance with the known structure properties; and outputting the formatted structured data content.”
On the other hand, in the same field of endeavor, Gerrit teaches
serializing the structured data content into the formatted structured data content ([Page 3]: In this paper, we present a method for automatically determining the formatting parameters, which we call the dialect, of a CSV file; [Page 4]: we receive a text file from an unknown source, created with an unknown formatter using an unknown dialect, that contains unknown data, and are asked to choose the parameters that allow a faithful reconstruction of the original data; [Page 9]: Figure 2. Illustration of the data consistency measure for different dialect on a constructed ex- ample),
that is formatted in accordance with a second structure dialect different from the first structure dialect, the second structure dialect being in compliance with the known structure properties ([Abstract]: Comma-separated value (CSV) files are a popular format for tabular data due to their simplicity and ostensible ease of use; [Page 16]: CSV files are considered “standard” when they use the comma as the delimiter, use either no quotes or the " character as quote character, and do not use the escape character); and
outputting the formatted structured data content ([Abstract]: Our method achieves 97% overall accuracy on a large corpus of real- world CSV files; [Page 14]: Separating the files into those that are messy and those that follow the CSV standard further illustrates how our method improves over existing methods (see Table 2)).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Saurav to incorporate the teachings of Gerrit to include serializing the structured data content into formatted structured data content and outputting the formatted structured data content.
The motivation for doing so would be to improve the accuracy on messy CSV file, as recognized by Gerrit ([Abstract] of Gerrit: In this paper, we propose a dialect detection method based on a novel measure of data consistency of parsed data files. Our method achieves 97% overall accuracy on a large corpus of real- world CSV files and improves the accuracy on messy CSV files by almost 22% compared to existing approaches).
Regarding Claim 2, the combined teachings of Saurav and Gerrit disclose the method of claim 1.
Saurav further teaches wherein the data content includes a structured datastore having the unknown structure properties (Fig. 1; [0031]: the database 106, which can store tabular files with known delimiters, is only needed during the training of the model. Thereafter, only a determined set of feature weights needs to be retained for use when processing new files wherein the delimiters are unknown).
Regarding Claim 3 the combined teachings of Saurav and Gerrit disclose the method of claim 1.
Saurav further teaches wherein the unknown structure properties include a structure dialect of the structured data content ([0052]: Commonly, many tabular data files use double quotes and commas as delimiters. Therefore, the first group of training files includes files that employ these standard delimiters. The second group of training files includes files that have been modified to employ a delimiter pair randomly selected from all possible combinations of non-alphabetic, non-numeric characters found in the file).
Regarding Claim 4, the combined teachings of Saurav and Gerrit disclose the method of claim 1.
Saurav further teaches wherein the unknown structure properties include a low-level data type of the structured data content associated with a data parser of a programming language designated to parse the structured data content (Fig. 1; [0046]: the file parser 132 parses (or partitions) the tabular data file… when the system is deployed and applied to files with unknown delimiters; [0055]: By applying machine learning to issues pertaining to automating the data wrangling process, the techniques detailed herein create a robust solution that can work with arbitrary character sets and delimiter choices; See also para [0075]; [This is non-functional descriptive material describing the data, and is not functionally involved in the steps recited. None of the claimed steps are depending on any of the information being described. The inputting, predicting, and extracting steps would be performed the same to achieve a same outcome regardless of the data type]).
Regarding Claim 5, the combined teachings of Saurav and Gerrit disclose the method of claim 1.
Saurav further teaches wherein the unknown structure properties include a high-level data type of the structured data content associated with the machine learning framework (Fig. 1; [0046]: the file parser 132 parses (or partitions) the tabular data file… when the system is deployed and applied to files with unknown delimiters; [0055]: By applying machine learning to issues pertaining to automating the data wrangling process, the techniques detailed herein create a robust solution that can work with arbitrary character sets and delimiter choices; See also para [0075]; [This is non-functional descriptive material describing the data, and is not functionally involved in the steps recited. None of the claimed steps are depending on any of the information being described. The inputting, predicting, and extracting steps would be performed the same to achieve a same outcome regardless of the data type]).
Regarding Claim 7, the combined teachings of Saurav and Gerrit disclose the method of claim 1.
Saurav further teaches wherein the predicted labels include a high-level data type of the structured data content associated with a machine learning framework (Fig. 1; [0046]: the file parser 132 parses (or partitions) the tabular data file… when the system is deployed and applied to files with unknown delimiters; [0055]: By applying machine learning to issues pertaining to automating the data wrangling process, the techniques detailed herein create a robust solution that can work with arbitrary character sets and delimiter choices; See also para [0075]; [This is non-functional descriptive material describing the data, and is not functionally involved in the steps recited. None of the claimed steps are depending on any of the information being described. The inputting, predicting, and extracting steps would be performed the same to achieve a same outcome regardless of the data type]).
Regarding Claim 8, the combined teachings of Saurav and Gerrit disclose the method of claim 1.
Saurav further teaches wherein the inputting operation includes inputting a visual image sample of the structured data content, wherein the trained machine learning model is further trained by visual image training samples of the structured training data content with the labeled structure properties corresponding to the unknown structure properties ([0023]: Once the model is trained, a file for which the correct delimiters are unknown can be parsed with various candidate delimiters; [0072]: The present invention may be a system, a method, and/or a computer program product at any possible technical detail level of integration; [This is non-functional descriptive material describing the input data, and is not functionally involved in the steps recited. None of the claimed steps are depending on any of the information being described. The inputting, predicting, and extracting steps would be performed the same to achieve a same outcome regardless of the data.]), and
wherein the predicting operation further predicts the unknown structure properties of the structured data content using the trained machine learning model based on the visual image sample ([0046]: The purpose of parsing the file is… to train the linear regression model (during training on files with known delimiters), or to score a candidate delimiter pair (when the system is deployed and applied to files with unknown delimiters); Fig. 2; [0060]: Step 206 includes outputting, to at least one user, the candidate column delimiter-candidate string delimiter pairing having the highest likelihood of validity).
Regarding Claim 9, Saurav discloses a computing system for transforming structured data content having unknown structure properties to formatted structured data content in compliance with known structure properties ([Abstract]: Methods, systems, and computer program products for automatic detection of string and column delimiters in tabular data files are provided herein; [0023]: a file for which the correct delimiters are unknown can be parsed with various candidate delimiters) identified by a trained machine learning model (Fig. 2; [0055]-[0064]: By applying machine learning to issues pertaining to automating the data wrangling process, the techniques detailed herein create a robust solution; [0062]: Subsequently, the alert and/or formatted data blocks are transmitted over a data channel to the user's wireless device), the computing system comprising: one or more hardware processors ([0010]: FIG. 3 is a system diagram of an exemplary computer system on which at least one embodiment of the invention can be implemented);
a datastore sampler executable by the one or more hardware processors and configured to generate a textual sample of the structured data content including the unknown structure properties (Fig. 1; [0023]: A training file can contain, for example, one row for each combination of (i) a sample data file… Once the model is trained, a file for which the correct delimiters are unknown can be parsed with various candidate delimiters; [0046]: score a candidate delimiter pair (when the system is deployed and applied to files with unknown delimiters),
([0023]: A training file can contain… a candidate delimiter pair. Each such row contains one column for each feature used by the linear regression model (see, for example, Table 2 below)), the first structure dialect not being in compliance with the known structure properties ([0023]: Additionally, if the current candidate delimiters are the correct delimiters, such an embodiment can include writing “1” to the training file; otherwise, such an embodiment can include writing “0” to the training file… Once the model is trained, a file for which the correct delimiters are unknown can be parsed with various candidate delimiters);
the trained machine learning model executable by the one or more hardware processors, trained by textual training samples of structured training data content with labeled structure properties corresponding to the unknown structure properties (Fig. 1; [0031]: the database 106, which can store tabular files with known delimiters, is only needed during the training of the model. Thereafter, only a determined set of feature weights needs to be retained for use when processing new files wherein the delimiters are unknown; Fig. 1; [0052]: To train the candidate evaluator 134 (also referred to herein as the classifier), at least one embodiment of the invention utilizes a collection of tabular data files; Fig. 2; [0058]: Step 204 includes… constructing one or more feature vectors corresponding to the detected candidate column delimiters and the detected candidate string delimiters),
including a loss function corresponding to each labeled structure property ([0053]: Tools (such as, for example, SciKit-Learn), can be used to train a logistic-regression classifier [Scikit-learn, a Python machine learning library, has several loss functions and metrics for measuring classification performance]; Fig. 5; [0105]: Examples of workloads and functions which may be provided from this layer include:… string and column delimiter detection 96), and
configured to predict labels for the unknown structure properties of the structured data content based on the textual sample (Figs. 1-2; [0054]-[0060]: Once the candidate evaluator 134 (classifier) has been trained, it can be used in one or more embodiments of the invention to analyze tabular data files for which the delimiters are unknown… [0060]: Step 206 includes outputting, to at least one user, the candidate column delimiter-candidate string delimiter pairing having the highest likelihood of validity),
wherein the labels identify the known structure properties of the structured data content (Fig. 1; [0046]: train the linear regression model (during training on files with known delimiters); [0047]: Further, the candidate evaluator 134 uses a logistic-regression classifier… and the candidate pair generator 136 identifies the candidate pair with the highest score as the one most likely to be the correct choice; [0062]: an alert is generated containing one or more identified string delimiter and column delimiter pairs pertaining to the tabular data files);
a data parser executable by the one or more hardware processors and configured to:
parse elements of the structured data content based on the known structure properties corresponding to the predicted labels (Fig.1; [0046]-[0049]: Additionally, as detailed herein, the file parser 132 parses (or partitions) the tabular data file in question using each selected candidate pair… As noted above, the candidate selector 120 uses one or more heuristics to select candidate string and column delimiter pairs);
However, Saurav does not explicitly teach “serialize the structured data content into formatted structured data content, wherein the formatted structured data content is formatted in compliance with the known structure properties; and outputting the formatted structured data content.”
On the other hand, in the same field of endeavor, Gerrit teaches
serialize the structured data content into formatted structured data content ([Page 3]: In this paper, we present a method for automatically determining the formatting parameters, which we call the dialect, of a CSV file; [Page 4]: we receive a text file from an unknown source, created with an unknown formatter using an unknown dialect, that contains unknown data, and are asked to choose the parameters that allow a faithful reconstruction of the original data; [Page 9]: Figure 2. Illustration of the data consistency measure for different dialect on a constructed ex- ample),
that is formatted in accordance with a second format different from the first format, the second format being in compliance with the known structure properties ([Abstract]: Comma-separated value (CSV) files are a popular format for tabular data due to their simplicity and ostensible ease of use; [Page 16]: CSV files are considered “standard” when they use the comma as the delimiter, use either no quotes or the " character as quote character, and do not use the escape character); and
output the formatted structured data content ([Abstract]: Our method achieves 97% overall accuracy on a large corpus of real- world CSV files; [Page 14]: Separating the files into those that are messy and those that follow the CSV standard further illustrates how our method improves over existing methods (see Table 2)).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Saurav to incorporate the teachings of Gerrit to include serializing the structured data content into formatted structured data content and outputting the formatted structured data content.
The motivation for doing so would be to improve the accuracy on messy CSV file, as recognized by Gerrit ([Abstract] of Gerrit: In this paper, we propose a dialect detection method based on a novel measure of data consistency of parsed data files. Our method achieves 97% overall accuracy on a large corpus of real- world CSV files and improves the accuracy on messy CSV files by almost 22% compared to existing approaches).
Regarding Claim 10, the combined teachings of Saurav and Gerrit disclose the computing system of claim 9.
Saurav further teaches wherein the unknown structure properties include a structure dialect of the structured data content ([0052]: Commonly, many tabular data files use double quotes and commas as delimiters. Therefore, the first group of training files includes files that employ these standard delimiters. The second group of training files includes files that have been modified to employ a delimiter pair randomly selected from all possible combinations of non-alphabetic, non-numeric characters found in the file).
Regarding Claim 11, the combined teachings of Saurav and Gerrit disclose the computing system of claim 9.
Saurav further teaches wherein the unknown structure properties include a low-level data type of the structured data content associated with the data parser (Fig. 1; [0046]: the file parser 132 parses (or partitions) the tabular data file… when the system is deployed and applied to files with unknown delimiters; [0055]: By applying machine learning to issues pertaining to automating the data wrangling process, the techniques detailed herein create a robust solution that can work with arbitrary character sets and delimiter choices; See also para [0075]; [This is non-functional descriptive material describing the data, and is not functionally involved in the steps recited. None of the claimed steps are depending on any of the information being described. The inputting, predicting, and extracting steps would be performed the same to achieve a same outcome regardless of the data type]).
Regarding Claim 12, the combined teachings of Saurav and Gerrit disclose the computing system of claim 9.
Saurav further teaches wherein the unknown structure properties include a high-level data type of the structured data content associated with the machine learning framework (Fig. 1; [0046]: the file parser 132 parses (or partitions) the tabular data file… when the system is deployed and applied to files with unknown delimiters; [0055]: By applying machine learning to issues pertaining to automating the data wrangling process, the techniques detailed herein create a robust solution that can work with arbitrary character sets and delimiter choices; See also para [0075]; [This is non-functional descriptive material describing the data, and is not functionally involved in the steps recited. None of the claimed steps are depending on any of the information being described. The inputting, predicting, and extracting steps would be performed the same to achieve a same outcome regardless of the data type]).
Regarding Claim 13, the combined teachings of Saurav and Gerrit disclose the computing system of claim 9.
Saurav further teaches wherein the loss functions of all labeled structure properties are trained concurrently ([0053]: Tools (such as, for example, SciKit-Learn), can be used to train a logistic-regression classifier [Scikit-learn, a Python machine learning library, has several loss functions and metrics for measuring classification performance]; Fig. 5; [0105]: Examples of workloads and functions which may be provided from this layer include:… string and column delimiter detection 96).
Regarding Claim 14, the combined teachings of Saurav and Gerrit disclose the computing system of claim 9.
Saurav further teaches wherein the datastore sampler further configured to input a visual image sample of the structured data content ([0023]: Once the model is trained, a file for which the correct delimiters are unknown can be parsed with various candidate delimiters; [0072]: The present invention may be a system, a method, and/or a computer program product at any possible technical detail level of integration; [This is non-functional descriptive material describing the input data, and is not functionally involved in the steps recited. None of the claimed steps are depending on any of the information being described. The inputting, predicting, and extracting steps would be performed the same to achieve a same outcome regardless of the data.]),
wherein the trained machine learning model is further trained by visual image training samples of the structured training data content with the labeled structure properties corresponding to the unknown structure properties ([0046]: The purpose of parsing the file is… to train the linear regression model (during training on files with known delimiters), or to score a candidate delimiter pair (when the system is deployed and applied to files with unknown delimiters), and
wherein the trained machine learning model further predicts the unknown structure properties of the structured data content using the trained machine learning model based on the visual image sample (Fig. 2; [0060]: Step 206 includes outputting, to at least one user, the candidate column delimiter-candidate string delimiter pairing having the highest likelihood of validity).
Regarding Claim 15, Saurav discloses one or more tangible processor-readable storage media embodied with instructions for executing on one or more processors and circuits of a computing device a process of transforming structured data content having unknown structure properties to formatted structured data content in compliance with known structure properties identified by a trained machine learning model ([Abstract]: Methods, systems, and computer program products for automatic detection of string and column delimiters in tabular data files are provided herein; [0023]: a file for which the correct delimiters are unknown can be parsed with various candidate delimiters), identified by a trained machine learning model (Fig. 2; [0055]-[0064]: By applying machine learning to issues pertaining to automating the data wrangling process, the techniques detailed herein create a robust solution; [0062]: Subsequently, the alert and/or formatted data blocks are transmitted over a data channel to the user's wireless device), the process comprising:
inputting a randomly-selected textual sample of the structured data content including the unknown structure properties into the trained machine learning model (Fig. 1; [0023]: A training file can contain, for example, one row for each combination of (i) a sample data file… Once the model is trained, a file for which the correct delimiters are unknown can be parsed with various candidate delimiters; [0046]: score a candidate delimiter pair (when the system is deployed and applied to files with unknown delimiters)),
wherein the textual sample is formatted in accordance with a first structure dialect ([0023]: A training file can contain… a candidate delimiter pair. Each such row contains one column for each feature used by the linear regression model (see, for example, Table 2 below)), the first structure dialect not being in compliance with the known structure properties ([0023]: Additionally, if the current candidate delimiters are the correct delimiters, such an embodiment can include writing “1” to the training file; otherwise, such an embodiment can include writing “0” to the training file… Once the model is trained, a file for which the correct delimiters are unknown can be parsed with various candidate delimiters),
wherein the trained machine learning model is trained by textual training samples of structured training data contents with labeled structure properties corresponding to the unknown structure properties (Fig. 1; [0031]: the database 106, which can store tabular files with known delimiters, is only needed during the training of the model. Thereafter, only a determined set of feature weights needs to be retained for use when processing new files wherein the delimiters are unknown; Fig. 1; [0052]: To train the candidate evaluator 134 (also referred to herein as the classifier), at least one embodiment of the invention utilizes a collection of tabular data files; Fig. 2; [0058]: Step 204 includes… constructing one or more feature vectors corresponding to the detected candidate column delimiters and the detected candidate string delimiters) and
includes a loss function corresponding to each labeled structure property ([0053]: Tools (such as, for example, SciKit-Learn), can be used to train a logistic-regression classifier [Scikit-learn, a Python machine learning library, has several loss functions and metrics for measuring classification performance]; Fig. 5; [0105]: Examples of workloads and functions which may be provided from this layer include:… string and column delimiter detection 96),
wherein the labels identify the known structure properties of the structured data content Fig. 1; [0046]: train the linear regression model (during training on files with known delimiters); [0047]: Further, the candidate evaluator 134 uses a logistic-regression classifier… and the candidate pair generator 136 identifies the candidate pair with the highest score as the one most likely to be the correct choice; [0062]: an alert is generated containing one or more identified string delimiter and column delimiter pairs pertaining to the tabular data files);
predicting labels for the unknown structure properties of the structured data content using the trained machine learning model based on the randomly-selected textual sample Figs. 1-2; [0054]-[0060]: Once the candidate evaluator 134 (classifier) has been trained, it can be used in one or more embodiments of the invention to analyze tabular data files for which the delimiters are unknown… [0060]: Step 206 includes outputting, to at least one user, the candidate column delimiter-candidate string delimiter pairing having the highest likelihood of validity),
parsing elements of the structured data content based on the known structure properties corresponding to the predicted labels (Fig.1; [0046]-[0049]: Additionally, as detailed herein, the file parser 132 parses (or partitions) the tabular data file in question using each selected candidate pair… As noted above, the candidate selector 120 uses one or more heuristics to select candidate string and column delimiter pairs),
However, Saurav does not explicitly teach “serializing the structured data content into formatted structured data content, wherein the formatted structured data content is formatted in compliance with the known structure properties; and outputting the formatted structured data content.”
On the other hand, in the same field of endeavor, Gerrit teaches
serializing the structured data content into the formatted structured data content ([Page 3]: In this paper, we present a method for automatically determining the formatting parameters, which we call the dialect, of a CSV file; [Page 4]: we receive a text file from an unknown source, created with an unknown formatter using an unknown dialect, that contains unknown data, and are asked to choose the parameters that allow a faithful reconstruction of the original data; [Page 9]: Figure 2. Illustration of the data consistency measure for different dialect on a constructed ex- ample),
formatted in accordance with a second format different from the first format, the first format not being in compliance with the known structure properties, the second format being in compliance with the known structure properties ([Abstract]: Comma-separated value (CSV) files are a popular format for tabular data due to their simplicity and ostensible ease of use; [Page 16]: CSV files are considered “standard” when they use the comma as the delimiter, use either no quotes or the " character as quote character, and do not use the escape character); and
outputting the formatted structured data content ([Abstract]: Our method achieves 97% overall accuracy on a large corpus of real- world CSV files; [Page 14]: Separating the files into those that are messy and those that follow the CSV standard further illustrates how our method improves over existing methods (see Table 2)).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Saurav to incorporate the teachings of Gerrit to include serializing the structured data content into formatted structured data content and outputting the formatted structured data content.
The motivation for doing so would be to improve the accuracy on messy CSV file, as recognized by Gerrit ([Abstract] of Gerrit: In this paper, we propose a dialect detection method based on a novel measure of data consistency of parsed data files. Our method achieves 97% overall accuracy on a large corpus of real- world CSV files and improves the accuracy on messy CSV files by almost 22% compared to existing approaches).
Regarding Claim 16, the combined teachings of Saurav and Gerrit disclose the one or more tangible processor-readable storage media of claim 15.
Saurav further teaches wherein the unknown structure properties include a structure dialect of the structured data content ([0052]: Commonly, many tabular data files use double quotes and commas as delimiters. Therefore, the first group of training files includes files that employ these standard delimiters. The second group of training files includes files that have been modified to employ a delimiter pair randomly selected from all possible combinations of non-alphabetic, non-numeric characters found in the file).
Regarding Claim 17, the combined teachings of Saurav and Gerrit disclose the one or more tangible processor-readable storage media of claim 15.
Saurav further teaches wherein the unknown structure properties include a low-level data type of the structured data content associated with a data parser of a programming language designated to parse the structured data content (Fig. 1; [0046]: the file parser 132 parses (or partitions) the tabular data file… when the system is deployed and applied to files with unknown delimiters; [0055]: By applying machine learning to issues pertaining to automating the data wrangling process, the techniques detailed herein create a robust solution that can work with arbitrary character sets and delimiter choices; See also para [0075]; [This is non-functional descriptive material describing the data, and is not functionally involved in the steps recited. None of the claimed steps are depending on any of the information being described. The inputting, predicting, and extracting steps would be performed the same to achieve a same outcome regardless of the data type]).
Regarding Claim 18, the combined teachings of Saurav and Gerrit disclose the one or more tangible processor-readable storage media of claim 15.
Saurav further teaches wherein the unknown structure properties include a high-level data type of the structured data content associated with a machine learning framework (Fig. 1; [0046]: the file parser 132 parses (or partitions) the tabular data file… when the system is deployed and applied to files with unknown delimiters; [0055]: By applying machine learning to issues pertaining to automating the data wrangling process, the techniques detailed herein create a robust solution that can work with arbitrary character sets and delimiter choices; See also para [0075]; [This is non-functional descriptive material describing the data, and is not functionally involved in the steps recited. None of the claimed steps are depending on any of the information being described. The inputting, predicting, and extracting steps would be performed the same to achieve a same outcome regardless of the data type]).
Regarding Claim 20, the combined teachings of Saurav and Gerrit disclose the one or more tangible processor-readable storage media of claim 15.
Saurav further teaches wherein the inputting operation includes inputting a visual image sample of the structured data content ([0023]: Once the model is trained, a file for which the correct delimiters are unknown can be parsed with various candidate delimiters; [0072]: The present invention may be a system, a method, and/or a computer program product at any possible technical detail level of integration; [This is non-functional descriptive material describing the input data, and is not functionally involved in the steps recited. None of the claimed steps are depending on any of the information being described. The inputting, predicting, and extracting steps would be performed the same to achieve a same outcome regardless of the data.]),
wherein the trained machine learning model is further trained by visual image training samples of the structured training data content with the labeled structure properties corresponding to the unknown structure properties ([0046]: The purpose of parsing the file is… to train the linear regression model (during training on files with known delimiters), or to score a candidate delimiter pair (when the system is deployed and applied to files with unknown delimiters), and
wherein the predicting operation further predicts the unknown structure properties of the structured data content using the trained machine learning model based on the visual image sample (Fig. 2; [0060]: Step 206 includes outputting, to at least one user, the candidate column delimiter-candidate string delimiter pairing having the highest likelihood of validity).
Regarding Claim 21, the combined teachings of Saurav and Gerrit disclose the one or more tangible processor-readable storage media of claim 15.
Gerrit further teaches wherein the structured data content includes a structured datastore having the unknown structure properties ([Page 4]: we receive a text file from an unknown source, created with an unknown formatter using an unknown dialect, that contains unknown data, and are asked to choose the parameters that allow a faithful reconstruction of the original data).
Regarding Claim 22, the combined teachings of Saurav and Gerrit disclose the computing system of claim 9.
Gerrit further teaches wherein the structured data content includes a structured datastore having the unknown structure properties ([Page 4]: we receive a text file from an unknown source, created with an unknown formatter using an unknown dialect, that contains unknown data, and are asked to choose the parameters that allow a faithful reconstruction of the original data).
Regarding Claim 23, the combined teachings of Saurav and Gerrit disclose the method of claim 1.
Saurav further teaches, further comprising inputting the parsed elements to a machine learning framework, wherein the parsed elements are accurately interpretable by the machine learning framework based on the known structure properties (Fig. 1; [0046]: the file parser 132 parses (or partitions) the tabular data file… when the system is deployed and applied to files with unknown delimiters; [0055]: By applying machine learning to issues pertaining to automating the data wrangling process, the techniques detailed herein create a robust solution that can work with arbitrary character sets and delimiter choices; See also para [0075]).
Regarding Claim 24, the combined teachings of Saurav and Gerrit disclose the method of claim 1.
Gerrit further teaches wherein the unknown structure properties include: a low-level data type of the structured data content associated with a data parser of a programming language designated to parse the structured data content; and a high-level data type of the structured data content associated with a machine learning framework ([Pages 19-20]: Data Type Detection: As mentioned in the main text, we use a regular expression based type detection engine. Below is a brief overview of the different types we consider and the detection method we use for that type).
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). 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 extension fee 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 SHIRLEY D. HICKS whose telephone number is (571)272-3304. The examiner can normally be reached Mon - Fri 7:30 - 4:00.
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, Charles Rones can be reached on (571) 272-4085. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/S D H/Examiner, Art Unit 2168
/CHARLES RONES/Supervisory Patent Examiner, Art Unit 2168