DETAILED ACTION
This communication is in response to the Amendments and Arguments filed on 2/17/2026.
Claims 1-23 are pending and have been examined.
All previous objections / rejections not mentioned in this Office Action have been withdrawn by the examiner.
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 .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 12/17/2026 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement was considered and attached by the examiner.
Response to Amendments
Applicant has amended independent claims 1, 12, 13, and 14. Hence, the Applicant’s arguments are moot in view of new grounds of rejection. More specifically, the newly added limitation to claim 1, 12, 13, and 24 are “the labeled training data comprising first labels labeling text sentences having a deliberative character and second labels, different than the first labels, labeling text sentences having a deliberative character in context with one or more surrounding text sentences” and “receiving a user input via the interactive user interface, the user input comprising an instruction associated with accepting or rejecting at least one suggested text redaction of the set of suggested text redactions; and updating the display of the plurality of identified text sentences based on the instruction associated with accepting or rejecting at least one suggested text redaction of the set of suggested text redactions.”. The added limitations raises new grounds for rejection. Since Applicant’s arguments are directed towards the new amendment, the arguments are moot in view of new grounds for rejection. Hence, new references have been applied.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 14-23 are rejected under 35 U.S.C. 103 as being unpatentable over Levay et al. (U.S. PG Pub No. 20230281330), hereinafter Levay, in view of Kwok et al. (U.S. PG Pub No. 20240061952), hereinafter Kwok, and in further view of Jaiswal et al. (U.S. PG Pub No. 20120303558).
Regarding claim 14 Levay teaches:
A method for providing suggested text redactions for a document, comprising: (P0002, Systems and methods for redacting information from documents.)
displaying a graphical user interface comprising a text region comprising a visual representation of a document comprising text, and a menu region comprising a first set of suggested text redactions corresponding to the document comprising text and a first set of interactive graphical user interface menu objects configured to receive user inputs corresponding to the first set of suggested text redactions, wherein the first set of suggested text redactions is generated by one or more artificial intelligence models that have been trained on labeled training data, the labeled training data comprising first labels labeling text sentences having a deliberative character and second labels, different than the first labels, labeling text sentences having a deliberative character in context with one or more surrounding text sentences; (P0062, When the user makes a redaction, the redaction is stored (in database) by calling the backend API, which sends an event to the document channel regarding any redaction event associated with this document. The database stores both a table and a history table to track which user performed which redaction action (e.g., add, delete, update), at which time, and to what portion of the document (e.g., using coordinates references to a specific corner of the document). The redaction is provided, via an external message bus service, to the users in real-time through the user interface on the respective frontend.; P0109, As the user provides instructions to effect redactions (through manual, search, image, pattern, and document redaction processes) the File Container updates the display of the document contents in real time. That is, the Redaction API tracks in a Redaction Log desired changes to the cached content and the Redaction API displays changed cached content (i.e., the cached content as modified according to the changes indicated in the Redaction Log) in the File Container so as to appear as the changed cached content would in the document. When the user has completed the redactions, the redaction changes to the content are finalized and the redaction document is made available for download by the user.; P0103, Redaction API effects the selection of a desired redaction methodology for identifying information to be redacted. For example, the Redaction API utilizes a Redaction Wizard. The Redaction Wizard guides a user through the selection and use of the above-described redaction methodologies, and for the document methodology, effects the suggesting to users what information should be redacted from a document based on the type of document (e.g., driver license, bank check, etc.). The Redaction Wizard uses Optical Character Recognition (OCR), Google Vision, Open CV, and Machine Learning (ML) algorithms to automatically detect content and send the content to the Redaction API and the Document Manipulation Engine for redaction.)
receiving a first user input comprising user interaction with a first menu object of the first set of menu objects, wherein the first user input indicates an instruction associated with accepting or rejecting at least one suggested text redaction of the first set of suggested text redactions; and (P0109, As the user provides instructions to effect redactions (through manual, search, image, pattern, and document redaction processes) the File Container updates the display of the document contents in real time. That is, the Redaction API tracks in a Redaction Log desired changes to the cached content and the Redaction API displays changed cached content (i.e., the cached content as modified according to the changes indicated in the Redaction Log) in the File Container so as to appear as the changed cached content would in the document. When the user has completed the redactions, the redaction changes to the content are finalized and the redaction document is made available for download by the user.)
updating display of the text region in accordance with the instruction associated with accepting or rejecting at least one suggested text redaction of the first set of suggested text redactions in response to receiving the first user input. (P0109, As the user provides instructions to effect redactions (through manual, search, image, pattern, and document redaction processes) the File Container updates the display of the document contents in real time.)
Levay does not specifically teach:
displaying a graphical user interface comprising a text region comprising a visual representation of a document comprising text, and a menu region comprising a first set of suggested text redactions corresponding to the document comprising text and a first set of interactive graphical user interface menu objects configured to receive user inputs corresponding to the first set of suggested text redactions, wherein the first set of suggested text redactions is generated by one or more artificial intelligence models that have been trained on labeled training data, the labeled training data comprising first labels labeling text sentences having a deliberative character and second labels, different than the first labels, labeling text sentences having a deliberative character in context with one or more surrounding text sentences;
Kwok, however, teaches:
displaying a graphical user interface comprising a text region comprising a visual representation of a document comprising text, and a menu region comprising a first set of suggested text redactions corresponding to the document comprising text and a first set of interactive graphical user interface menu objects configured to receive user inputs corresponding to the first set of suggested text redactions, wherein the first set of suggested text redactions is generated by one or more artificial intelligence models that have been trained on labeled training data, the labeled training data comprising first labels labeling text sentences having a deliberative character and second labels, different than the first labels, labeling text sentences having a deliberative character in context with one or more surrounding text sentences; (P0024, The sensitive information is input into a machine learning model as part of a dataset, a string of data, or another form of data. Generally, a sensitive information detection model uses a two-fold solution to leverage redacted data as part of training data. First, the machine learning model is used to understand the context around sensitive data (e.g., sensitive information). Second, regular expression logic is used to understand and analyze a regular dataset that may contain sensitive data and represent an understanding of sensitive data inner patterns. By combining these two solutions and training a final machine learning model on a previously, at least partially, human-labeled dataset, a more fully understanding model is created that makes better predictions of what is or is not sensitive data. In some configurations, the machine learning model may use metadata surrounding the possible sensitive information when determining if the data is actually sensitive information.)
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to train an artificial intelligence model to obtain text redaction suggestion trained with labeled sentence and labeled sentence with context. It would have been obvious to combine the references because identifying context data surrounding redacted personal data in the electronic form field, and retraining the machine learning model with the context data and the type of the personal data that improves predictive accuracy of detecting the personal data. (Kwok P0005).
Levay in view of Kwok does not specifically teach:
displaying a graphical user interface comprising a text region comprising a visual representation of a document comprising text, and a menu region comprising a first set of suggested text redactions corresponding to the document comprising text and a first set of interactive graphical user interface menu objects configured to receive user inputs corresponding to the first set of suggested text redactions, wherein the first set of suggested text redactions is generated by one or more artificial intelligence models that have been trained on labeled training data, the labeled training data comprising first labels labeling text sentences having a deliberative character and second labels, different than the first labels, labeling text sentences having a deliberative character in context with one or more surrounding text sentences;
Jaiswal, however, teaches:
displaying a graphical user interface comprising a text region comprising a visual representation of a document comprising text, and a menu region comprising a first set of suggested text redactions corresponding to the document comprising text and a first set of interactive graphical user interface menu objects configured to receive user inputs corresponding to the first set of suggested text redactions, wherein the first set of suggested text redactions is generated by one or more artificial intelligence models that have been trained on labeled training data, the labeled training data comprising first labels labeling text sentences having a deliberative character and second labels, different than the first labels, labeling text sentences having a deliberative character in context with one or more surrounding text sentences; (P0009, Redacting portions from the item of data that contain specific categories of sensitive information that the entity is not authorized to access.; P0041, As illustrated in FIG. 3, at step 302 the systems described herein may identify a plurality of specific categories of sensitive information that are to be protected by a DLP system.; P0043, The term “category of sensitive information,” as used herein, may refer to … sensitive marketing information (e.g., marketing plans, product launch timelines, etc.).)
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to determine deliberative language to redact. It would have been obvious to combine the references because choosing a specific category to redact should be based on whether the text is made up of more or less than a specific percentage of one or more specific categories of sensitive information and input text can have deliberative language that is part of a category of sensitive information for redaction. (Jaiswal P0008).
Regarding claim 15 Levay in view of Kwok and further view of Jaiswal teach claim 14.
Levay further teaches:
wherein the first user input indicates acceptance of the suggested text redaction of the first set of suggested text redactions. (P0103, For the document methodology, effects the suggesting to users what information should be redacted from a document based on the type of document (e.g., driver license, bank check, etc.).; P0108, Analyzes the resulting data to enhance the ability of the Redaction Wizard to automatically detect sensitive content in documents and suggest content for possible redaction.; P0109, As the user provides instructions to effect redactions (through manual, search, image, pattern, and document redaction processes) the File Container updates the display of the document contents in real time. That is, the Redaction API tracks in a Redaction Log desired changes to the cached content and the Redaction API displays changed cached content (i.e., the cached content as modified according to the changes indicated in the Redaction Log) in the File Container so as to appear as the changed cached content would in the document.)
Regarding claim 16 Levay in view of Kwok and further view of Jaiswal teach claim 14.
Levay further teaches:
wherein the first user input indicates rejection of the suggested text redaction of the first set of suggested text redactions. (P0103, For the document methodology, effects the suggesting to users what information should be redacted from a document based on the type of document (e.g., driver license, bank check, etc.).; P0108, Analyzes the resulting data to enhance the ability of the Redaction Wizard to automatically detect sensitive content in documents and suggest content for possible redaction.; P0109, As the user provides instructions to effect redactions (through manual, search, image, pattern, and document redaction processes) the File Container updates the display of the document contents in real time. That is, the Redaction API tracks in a Redaction Log desired changes to the cached content and the Redaction API displays changed cached content (i.e., the cached content as modified according to the changes indicated in the Redaction Log) in the File Container so as to appear as the changed cached content would in the document.)
Regarding claim 17 Levay in view of Kwok and further view of Jaiswal teach claim 14.
Levay further teaches:
wherein the menu region comprises an interactive graphical user interface menu option configured to receive user-specified text redaction patterns. (P0085, With reference to FIG. 11, the redaction methodology selector is configured to select a desired redaction methodology for identifying information to be redacted. As shown in FIG. 11, the desired methodology is selected from a selection set including at least one of manual methodology, search methodology, image methodology, pattern methodology and document methodology.; P0092, If the selected methodology is pattern methodology, the information to be redacted is content in a format, and the information is identified by a user identifying the format, the system searching in the document for any content in the format, and the system finding in the document all content in the format. For example, a user can select, indicate, or otherwise provide a format in which information to be redacted may appear and request that the system find information in the document that appears in the provided format.)
Regarding claim 18 Levay in view of Kwok and further view of Jaiswal teach claim 17.
Levay further teaches:
receiving a second user input comprising user interaction with the menu option, wherein the second user input indicates a user-specified text redaction pattern; and in response to receiving the second user input, generating a second set of suggested text redactions corresponding to the user-specified text redaction pattern. (P0092, If the selected methodology is pattern methodology, the information to be redacted is content in a format, and the information is identified by a user identifying the format, the system searching in the document for any content in the format, and the system finding in the document all content in the format. For example, a user can select, indicate, or otherwise provide a format in which information to be redacted may appear and request that the system find information in the document that appears in the provided format.)
Regarding claim 19 Levay in view of Kwok and further view of Jaiswal teach claim 18.
Levay further teaches:
wherein the menu region comprises a second set of interactive graphical user interface menu objects configured to receive user inputs corresponding to the second set of suggested text redactions. (P0109, As the user provides instructions to effect redactions (through manual, search, image, pattern, and document redaction processes) the File Container updates the display of the document contents in real time. That is, the Redaction API tracks in a Redaction Log desired changes to the cached content and the Redaction API displays changed cached content (i.e., the cached content as modified according to the changes indicated in the Redaction Log) in the File Container so as to appear as the changed cached content would in the document. When the user has completed the redactions, the redaction changes to the content are finalized and the redaction document is made available for download by the user.)
Regarding claim 20 Levay in view of Kwok and further view of Jaiswal teach claim 19.
Levay further teaches:
receiving a third user input comprising user interaction with a second menu object of the second set of menu objects, wherein the third user input indicates an instruction corresponding to a suggested text redaction of the second set of suggested text redactions; and in response to receiving the third user input, updating display of the text region in accordance with the third user input. (P0109, As the user provides instructions to effect redactions (through manual, search, image, pattern, and document redaction processes) the File Container updates the display of the document contents in real time. That is, the Redaction API tracks in a Redaction Log desired changes to the cached content and the Redaction API displays changed cached content (i.e., the cached content as modified according to the changes indicated in the Redaction Log) in the File Container so as to appear as the changed cached content would in the document. When the user has completed the redactions, the redaction changes to the content are finalized and the redaction document is made available for download by the user.)
Regarding claim 21 Levay in view of Kwok and further view of Jaiswal teach claim 20.
Levay further teaches:
wherein the third user input indicates acceptance of the suggested text redaction of the second set of suggested text redactions. (P0109, As the user provides instructions to effect redactions (through manual, search, image, pattern, and document redaction processes) the File Container updates the display of the document contents in real time. That is, the Redaction API tracks in a Redaction Log desired changes to the cached content and the Redaction API displays changed cached content (i.e., the cached content as modified according to the changes indicated in the Redaction Log) in the File Container so as to appear as the changed cached content would in the document. When the user has completed the redactions, the redaction changes to the content are finalized and the redaction document is made available for download by the user.)
Regarding claim 22 Levay in view of Kwok and further view of Jaiswal teach claim 20.
Levay further teaches:
wherein the third user input indicates rejection of the suggested text redaction of the second set of suggested text redactions. (P0109, As the user provides instructions to effect redactions (through manual, search, image, pattern, and document redaction processes) the File Container updates the display of the document contents in real time. That is, the Redaction API tracks in a Redaction Log desired changes to the cached content and the Redaction API displays changed cached content (i.e., the cached content as modified according to the changes indicated in the Redaction Log) in the File Container so as to appear as the changed cached content would in the document. When the user has completed the redactions, the redaction changes to the content are finalized and the redaction document is made available for download by the user.)
Regarding claim 23 Levay in view of Kwok and further view of Jaiswal teach claim 14.
Levay further teaches:
receiving a fourth user input comprising user interaction with one or more portions of the visual representation of the document, wherein the fourth user input indicates one or more portions of the document to redact; and (P0085, With reference to FIG. 11, the redaction methodology selector is configured to select a desired redaction methodology for identifying information to be redacted. As shown in FIG. 11, the desired methodology is selected from a selection set including at least one of manual methodology, search methodology, image methodology, pattern methodology and document methodology.; P0087, If the selected methodology is manual methodology, the information to be redacted is any content in the document, and the information is identified by a user navigating the document and selecting the content.)
in response to receiving the fourth user input, updating display of the text region to redact the one or more portions of the document corresponding to the fourth user input. (P0109, As the user provides instructions to effect redactions (through manual, search, image, pattern, and document redaction processes) the File Container updates the display of the document contents in real time. That is, the Redaction API tracks in a Redaction Log desired changes to the cached content and the Redaction API displays changed cached content (i.e., the cached content as modified according to the changes indicated in the Redaction Log) in the File Container so as to appear as the changed cached content would in the document. When the user has completed the redactions, the redaction changes to the content are finalized and the redaction document is made available for download by the user.)
Claims 1-13 are rejected under 35 U.S.C. 103 as being unpatentable over Levay in view of Kwok, in view of Jaiswal, and further view of Mane (U.S. PG Pub No. 20230205988).
Regarding claim 1, 12, and 13 Levay teaches:
(Claim 1) A method for providing suggested text redactions for a document, comprising: (P0002, Systems and methods for redacting information from documents.)
(Claim 12) A system for providing suggested text redactions for a document, comprising one or more processors and a memory coupled to the processors comprising instructions executable by the processors, the processors being operable when executing the instructions to cause the system to perform a method comprising: (A system for providing suggested text redactions for a document, comprising one or more processors and a memory coupled to the processors comprising instructions executable by the processors, the processors being operable when executing the instructions to cause the system to perform a method comprising:)
(Claim 13) A non-transitory computer-readable storage medium storing instructions that, when executed by one or more processors of an electronic device, cause the device to perform a method comprising: (P0184, Implementations of at least some of the subject matter described in this specification can be implemented as one or more computer program products, e.g., one or more modules of computer program instructions encoded on a tangible and non-transitory computer readable medium for execution by, or to control the operation of, data processing apparatus.; P0186, The processes and logic flows described in this specification can be performed by one or more programmable processors executing one or more computer programs to perform functions by operating on input data and generating output.)
receiving, from a user, the document comprising text; (P0005, The system includes a redaction container that includes a pre-processing module configured to receive a plurality of documents.)
extracting the text from the document; (P0063, FIG. 9 illustrates an example process flow for optical character recognize (OCR). When a user uploads a file (or, equivalently, a document), the R-API sends a request to the OCR service to determine whether that file requires processing using the OCR service.; P0064, Performing a “fast OCR check” by checking whether the document has any text.; P0067, The load balancer handles all the steps in the OCR process, which include page image extraction from the file, document layout analysis, image page correction, text line detection, text word detection, text recognition, and building the output file.)
displaying the plurality of identified text sentences and the set of suggested text redactions for the plurality of identified text sentences in an interactive user interface. (P0109, The Redaction API further displays the cached content in the File Container so as to appear as the content would in the document.)
receiving a user input via the interactive user interface, the user input comprising an instruction associated with accepting or rejecting at least one suggested text redaction of the set of suggested text redactions; and (P0109, As the user provides instructions to effect redactions (through manual, search, image, pattern, and document redaction processes) the File Container updates the display of the document contents in real time. That is, the Redaction API tracks in a Redaction Log desired changes to the cached content and the Redaction API displays changed cached content (i.e., the cached content as modified according to the changes indicated in the Redaction Log) in the File Container so as to appear as the changed cached content would in the document. When the user has completed the redactions, the redaction changes to the content are finalized and the redaction document is made available for download by the user.)
updating the display of the plurality of identified text sentences based on the instruction associated with accepting or rejecting at least one suggested text redaction of the set of suggested text redactions. (P0109, As the user provides instructions to effect redactions (through manual, search, image, pattern, and document redaction processes) the File Container updates the display of the document contents in real time.)
Levay does not specifically teach:
parsing the extracted text into a plurality of identified text sentences;
inputting the plurality of identified text sentences into one or more trained artificial intelligence models that have been trained on labeled text sentences to generate a set of suggested text redactions for the plurality of identified text sentences, the labeled training data comprising first labels labeling text sentences having a deliberative character and second labels, different than the first labels, labeling text sentences having a deliberative character in context with one or more surrounding text sentences;
Kwok, however, teaches:
inputting the plurality of identified text sentences into one or more trained artificial intelligence models that have been trained on labeled text sentences to generate a set of suggested text redactions for the plurality of identified text sentences, the labeled training data comprising first labels labeling text sentences having a deliberative character and second labels, different than the first labels, labeling text sentences having a deliberative character in context with one or more surrounding text sentences; (P0030, A primary input to the machine learning model and the NLP logic is the dataset itself, which may or may not contains sensitive information. The dataset is input to the sensitive information protection system and then into the machine learning model together with the NLP logic.; P0024, The sensitive information is input into a machine learning model as part of a dataset, a string of data, or another form of data. Generally, a sensitive information detection model uses a two-fold solution to leverage redacted data as part of training data. First, the machine learning model is used to understand the context around sensitive data (e.g., sensitive information). Second, regular expression logic is used to understand and analyze a regular dataset that may contain sensitive data and represent an understanding of sensitive data inner patterns. By combining these two solutions and training a final machine learning model on a previously, at least partially, human-labeled dataset, a more fully understanding model is created that makes better predictions of what is or is not sensitive data. In some configurations, the machine learning model may use metadata surrounding the possible sensitive information when determining if the data is actually sensitive information.)
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to train an artificial intelligence model to obtain text redaction suggestion trained with labeled sentence and labeled sentence with context. It would have been obvious to combine the references because identifying context data surrounding redacted personal data in the electronic form field, and retraining the machine learning model with the context data and the type of the personal data that improves predictive accuracy of detecting the personal data. (Kwok P0005).
Levay in view of Kwok does not specifically teach:
inputting the plurality of identified text sentences into one or more trained artificial intelligence models that have been trained on labeled text sentences to generate a set of suggested text redactions for the plurality of identified text sentences, the labeled training data comprising first labels labeling text sentences having a deliberative character and second labels, different than the first labels, labeling text sentences having a deliberative character in context with one or more surrounding text sentences;
Jaiswal, however, teaches:
inputting the plurality of identified text sentences into one or more trained artificial intelligence models that have been trained on labeled text sentences to generate a set of suggested text redactions for the plurality of identified text sentences, the labeled training data comprising first labels labeling text sentences having a deliberative character and second labels, different than the first labels, labeling text sentences having a deliberative character in context with one or more surrounding text sentences; (P0009, Redacting portions from the item of data that contain specific categories of sensitive information that the entity is not authorized to access.; P0041, As illustrated in FIG. 3, at step 302 the systems described herein may identify a plurality of specific categories of sensitive information that are to be protected by a DLP system.; P0043, The term “category of sensitive information,” as used herein, may refer to … sensitive marketing information (e.g., marketing plans, product launch timelines, etc.).)
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to determine deliberative language to redact. It would have been obvious to combine the references because choosing a specific category to redact should be based on whether the text is made up of more or less than a specific percentage of one or more specific categories of sensitive information and input text can have deliberative language that is part of a category of sensitive information for redaction. (Jaiswal P0008).
Levay in view of Kwok and further view of Jaiswal does not specifically teach:
parsing the extracted text into a plurality of identified text sentences;
Mane, however, teaches:
parsing the extracted text into a plurality of identified text sentences; (P0005, System comprises a parser that analyzes documents to identify structured, semi-structured, and unstructured data from a document.; P0058, Candidates generator breaks down unstructured content/data into sentences and then each sentence is subjected to a POS (Part of Speech) tagger. The candidates for redaction are generated based on NLP metadata.)
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to parse text into sentences. It would have been obvious to combine the references because parsing text into sentences is a known technique to yield a predictable result of identifying sentences to determine the different parts of speech for potential redactions.
Regarding claim 2 Levay, in view of Kwok, in view of Jaiswal, and further view of Mane teach claim 1.
Levay further teaches:
wherein the document is a Portable Document Format (PDF) document, a plain text (TXT) document, a Joint Photographic Experts Group (JPEG) document, or a Portable Network Graphics (PNG) document. (P0099, The document is provided by the user to the system in a file type, and the redacted version of the document is saved in the file type. … At least one of the file types is the Portable Document Format (PDF).)
Regarding claim 3 Levay, in view of Kwok, in view of Jaiswal, and further view of Mane teach claim 1.
Levay further teaches:
wherein extracting the text comprises identifying one or more text-based sections of the document from a plurality of sections of the document. (P0067, The load balancer handles all the steps in the OCR process, which include page image extraction from the file, document layout analysis, image page correction, text line detection, text word detection, text recognition, and building the output file.)
Regarding claim 4 Levay, in view of Kwok, in view of Jaiswal, and further view of Mane teach teach claim 1.
Levay further teaches:
wherein extracting the text comprises computing a visual position and size for a plurality of text characters of the text. (P0067, The load balancer handles all the steps in the OCR process, which include page image extraction from the file, document layout analysis, image page correction, text line detection, text word detection, text recognition, and building the output file.)
Regarding claim 5 Levay, in view of Kwok, in view of Jaiswal, and further view of Mane teach teach claim 1.
Levay further teaches:
wherein parsing the extracted text comprises identifying visual boundaries for a plurality of graphic representations of text characters and assembling the plurality of graphic representations of text characters into one or more groups. (P0067, The load balancer handles all the steps in the OCR process, which include page image extraction from the file, document layout analysis, image page correction, text line detection, text word detection, text recognition, and building the output file.)
Regarding claim 6 Levay, in view of Kwok, in view of Jaiswal, and further view of Mane teach claim 1.
Levay in view of Kwok and further view of Jaiswal does not specifically teach:
wherein parsing the extracted text comprises grouping the extracted text into the plurality of identified text sentences.
Mane, however, teaches:
wherein parsing the extracted text comprises grouping the extracted text into the plurality of identified text sentences. (P0005, System comprises a parser that analyzes documents to identify structured, semi-structured, and unstructured data from a document.; P0058, Candidates generator breaks down unstructured content/data into sentences and then each sentence is subjected to a POS (Part of Speech) tagger. The candidates for redaction are generated based on NLP metadata.)
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to parse text into sentences. It would have been obvious to combine the references because parsing text into sentences is a known technique to yield a predictable result of identifying sentences to determine the different parts of speech for potential redactions.
Regarding claim 7 Levay, in view of Kwok, in view of Jaiswal, and further view of Mane teach claim 1.
Levay further teaches:
wherein the one or more trained artificial intelligence models comprise a trained language model. (P0044, Using natural language processing (NLP) that uses machine learning (ML) to uncover information in unstructured data and text within.)
Regarding claim 8 Levay, in view of Kwok, in view of Jaiswal, and further view of Mane teach claim 1.
Levay further teaches:
wherein the set of suggested text redactions is displayed on a representation of the document. (P0109, As the user provides instructions to effect redactions (through manual, search, image, pattern, and document redaction processes) the File Container updates the display of the document contents in real time.)
Regarding claim 9 Levay, in view of Kwok, in view of Jaiswal, and further view of Mane teach claim 1.
Levay further teaches:
wherein the set of suggested text redactions corresponds to whether each of the plurality of identified sentences is associated with one or more predefined categories of information for redaction. (P0103, For the document methodology, effects the suggesting to users what information should be redacted from a document based on the type of document (e.g., driver license, bank check, etc.).; P0108, Analyzes the resulting data to enhance the ability of the Redaction Wizard to automatically detect sensitive content in documents and suggest content for possible redaction.)
Regarding claim 10 Levay, in view of Kwok, in view of Jaiswal, and further view of Mane teach claim 1.
Levay in view of Kwok does not specifically teach:
wherein the one or more predefined categories of information comprises deliberative language.
Jaiswal, however, teaches:
wherein the one or more predefined categories of information comprises deliberative language. (P0009, Redacting portions from the item of data that contain specific categories of sensitive information that the entity is not authorized to access.; P0041, As illustrated in FIG. 3, at step 302 the systems described herein may identify a plurality of specific categories of sensitive information that are to be protected by a DLP system.; P0043, The term “category of sensitive information,” as used herein, may refer to … sensitive marketing information (e.g., marketing plans, product launch timelines, etc.).)
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to determine a category as deliberative language. It would have been obvious to combine the references because choosing a specific category to redact should be based on whether the text is made up of more or less than a specific percentage of one or more specific categories of sensitive information and input text can have deliberative language that is part of a category of sensitive information for redaction. (Jaiswal P0008).
Regarding claim 11 Levay, in view of Kwok, in view of Jaiswal, and further view of Mane teach claim 1.
Levay does not specifically teach:
prior to inputting the plurality of identified text sentences into the one or more trained artificial intelligence models, determining a set of features associated with the plurality of identified text sentences, wherein the set of features are inputted into the one or more trained artificial intelligence models.
Kwok, however, teaches:
prior to inputting the plurality of identified text sentences into the one or more trained artificial intelligence models, determining a set of features associated with the plurality of identified text sentences, wherein the set of features are inputted into the one or more trained artificial intelligence models. (P0035, Neural networks can also be implemented within the NLP logic 216. Some techniques include the use of “word embedding” to capture semantic properties of words, and an increase in end-to-end learning of a higher-level task (e.g., question answering) instead of relying on a pipeline of separate intermediate tasks (e.g., part-of-speech and dependency parsing).)
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to determine a set of features associated with identified text sentences. It would have been obvious to combine the references because obtaining features for a sentence is a known technique that yields a predictable result of preserving semantic relationship between items when providing input into artificial intelligence models. (Kwok P0035)
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 nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DANIEL WONSUK CHUNG whose telephone number is (571)272-1345. The examiner can normally be reached Monday - Friday (7am-4pm)[PT].
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, PIERRE-LOUIS DESIR can be reached at (571)272-7799. 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.
/DANIEL W CHUNG/Examiner, Art Unit 2659
/PIERRE LOUIS DESIR/Supervisory Patent Examiner, Art Unit 2659
Prosecution Insights