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 .
Amendment filed on 04/14/2026 has been entered. Claims 1-29 are pending. Claims 1-3, 9-13, 17, 19-20, 24-25, 28 and 29 are currently amended.
Response to Amendment
The recent filing of the claims contains incorrect status identifiers. MPEP 714, Section C (A) states the current status of all claims in the application must be given. In addition, MPEP, 714(c)(2) states "All claims being currently amended in an amendment paper shall be presented in the claim listing, indicate a status of "currently amended," and be submitted with markings to indicate the changes that have been made relative to the immediate prior version of the claims. Claim 29 of the amendment filed 4/14/2026 has been amended but was not properly labeled. Therefore, this claim has been treated as if the status identifier of "currently amended". If this is not correct, applicants should notify the USPTO with a corrected amendment. Any future correspondence must contain the proper status identifiers.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
Claims 1, 19 and 29 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Regarding independent claims 1, 19 and 29
Step 1 -- whether the claim falls within any statutory category. See MPEP 2106.03
Claim 1 is drawn to a system claim, claim 19 is drawn to a method claim and claim 29 is drawn to computer program product claim. Therefore, each of these claims falls under one of the four categories of statutory subject matter (process/method, machine/product/apparatus, manufacture, or composition of matter).
Step 2A Prong 1 – whether the claim recites a judicial exception. See MPEP 2106.04, subsection II.
Regarding claim 1 is directed to a system for optimized training of a neural network model for data extraction, the system comprising the limitations of: extract words from a structured or semi-structured input document for generating a pre-determined format type of the input document, wherein the extracted word include an entity text and neighboring words associated with the entity text; generate N-grams by analyzing … in a predefined order; label the N-grams with a field name by comparing the generated N-grams with the coordinates corresponding to the words; tokenize each word in the labelled N-gram assigning a token marker. These limitations are directed towards the abstract idea which is performed manually by a human being and/or by paper and pen. These limitations are directed towards the abstract idea of a mathematical relationship, specially organizing information and manipulation information through mathematical correlations.
Independent claim 19 is a method claim reciting similar limitations of claim 1 and is directed towards the abstract idea for similar reasons.
Independent claim 29 is a product claim reciting similar limitations of claim 1 and is directed towards the abstract idea for similar reasons.
Step 2A Prong 2 -- whether the claim as a whole integrates the recited judicial exception into a practical application of the exception or whether the claim is “directed to” the judicial exception. This evaluation is performed by (1) identifying whether there are any additional elements recited in the claim beyond the judicial exception, and (2) evaluating those additional elements individually and in combination to determine whether the claim as a whole integrates the exception into a practical application. See MPEP 2106.04(d).
Regarding independent claim 1, the claim recites additional elements of “memory”, “processor”, “data extraction engine”, “train neural network model” and “the trained neural network model is employed for automatically extracting relevant data from documents.”
These limitations amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). The additional limitations “memory”, “processor”, and “data extraction engine” amount to no more than generally linking the use of the judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). It recites a generic computer or generic computer components that merely act as a tool on which the method operates, and thus, fails to integrate the exception into a practical application. In addition, it recites using neural network (NN) model for extracting data from documents, train the neural network model based on the extracted data and the trained neural network model extract relevant data from documents. Although Applicant recites “trained neural network model is employed for automatically extracting relevant data from document”, a human being can extract data from a document, observe a model with the extracted data, and use the model to extract relevant data from document and thus, fails to integrate the exception into a practical application.
Regarding independent claim 19, this claim is drawn to a method claim reciting similar limitations of claim 1 and is rejected under the same rationale. Claim 19 also recites additional elements of “memory”, “processor”, “data extraction engine” and “train neural network model”.
These limitations amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). The additional limitations “memory”, “processor”, and “data extraction engine” amount to no more than generally linking the use of the judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). It recites a generic computer or generic computer components that merely act as a tool on which the method operates, and thus, fails to integrate the exception into a practical application. In addition, it recites using neural network (NN) model for extracting data from documents and train the neural network model based on the extracted data. Although Applicant recites “trained neural network model is employed for automatically extracting relevant data from document”, a human being can extract data from a document, observe a model with the extracted data, and use the model to extract relevant data from document and thus, fails to integrate the exception into a practical application.
Regarding independent claim 29, this claim is drawn to a product claim reciting similar limitations of claim 1 and is rejected under the same rationale. Claim 29 also recites additional elements of “memory”, “processor”, “data extraction engine” and “train neural network model”.
These limitations amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). The additional limitations “memory”, “processor”, and “data extraction engine” amount to no more than generally linking the use of the judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). It recites a generic computer or generic computer components that merely act as a tool on which the method operates, and thus, fails to integrate the exception into a practical application. In addition, it recites using neural network (NN) model for extracting data from documents and train the neural network model based on the extracted data. Although Applicant recites “trained neural network model is employed for automatically extracting relevant data from document”, a human being can extract data from a document, observe a model with the extracted data, and use the model to extract relevant data from document and thus, fails to integrate the exception into a practical application.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-10, 12, 14-29 are rejected under 35 U.S.C. 103 as being unpatentable over Appalaraju et al. (Appalaraju), US Patent No. US 11,893,012 B1, and further in view of Zuev et al. (Zuev), US Patent Application Publication No. US 2019/0385054 A1, and further in view of Booth, US Patent Application Publication No. US 2022/0382983 A1.
comprising:
As to independent claim 1, Appalaraju discloses a system for optimized training of a neural network model for data extraction, the system comprising:
a memory storing program instructions (Figure 10 and col. 22, line 46 – col. 23, line 4: system memory may be configured to store instructions);
a processor executing instructions stored in the memory (Figure 10 and col. 22, line 46 – col. 23, line 4: system memory may be configured to store instructions and data accessible by processor(s)); and
a data extraction engine executed by the processor and configured to:
extract words from a structured or semi-structured input document for generating a pre-determined format type of the input document and store the extracted words in a database along with spatial coordinates corresponding to each extracted word, wherein the extracted words include an entity text and neighboring words associated with the entity text (col. 2, line 45 – col. 3, line 7: extract content about related entities such as key-value pairs from images or other representations of target data objects, using machine learning techniques for layout-aware comparisons between a set of pre-obtained reference objects and the target data objects, wherein many data objects whose content is to be extracted in an automated manner, and wherein type of information that has to be included in an income statement, a tax-related form, a medical claim, etc., may be defined by regulations or customs of the organizations involved in preparing or generating the document. To extract the information from previously unseen target documents of the more frequently encountered categories with high accuracy if (a) properties of the different types of documents expected to be analyzed frequently can be collected and stored in advance, and (b) a reasonable probabilistic prediction can be made using the collected properties about the particular document type to which a target document is likely to belong; col. 8, line 53 – col. 9, line 27: Figure 3 and col. 12, line 54 – col. 13, line 8: preliminary content extraction may first be performed on a scanned/photographed target document, the target document may be provided as input to one or more machine learning models such as optical character recognition (OCR) models, and output comprising detected characters, words, numbers and the like, along with their respective bounding boxes such as approximate location/position information expressed using bounding box coordinates; col. 11, line 55 and col. 12, line 35 and Figure 2: the extracted content elements contain keys such as name, date of birth, contact information and these are considered as entity text, wherein some keys such as “contact information” may comprise several lower-level keys (neighboring words) such as Street , City, state, and the like);
generate N-grams by analyzing the neighboring words associated with the entity text present in the predetermined format type of the document based on a threshold distance measurement criterion and combine the extracted neighboring words in a pre-defined order (col. 11, lines 18-41: the corresponding content elements may be identified using n-gram analysis in some implementations, in which n-grams of a chosen token length, e.g, 2 words or tokens, or 3 words/tokens, or n-gram occurs such as nearby words may be performed);
label the N-grams with a filed name by comparing the generated N-grams with the coordinates corresponding to the words (col. 11, lines 18-41: the objects being analyzed are text-containing documents, the structural comparison may comprise a preliminary step of identifying corresponding content elements, the corresponding content elements may be identified using n-gram analysis in which n-grams of a chosen token length that are present within both the target document and the reference document are found), wherein if the entity text is present in more than one place within the document, the coordinates identify the entity text and label the corresponding N-gram with the correct field name without dependence of field values (col. 4, line 46 – col. 5, line 7: a particular key-value descriptor (entity text) may comprise information about at least a text key within an image of a reference document, wherein a key-value descriptor may include text tokens or words that makeup a key, and a bounding box (expressed using a pair of x, y coordinates representing opposite corners of a rectangle) within which the text tokens/words were identified. In addition, key-value-group descriptors may be stored for reference documents, representing scenarios in which multiple values (words) related to a single key may be expected to be present in a given document, as such, a key-value descriptor (entity text) may represent one example of a key-value-group descriptor);
tokenize each word in the labelled N-gram in accordance with a location of each of the words relative to a named entity (NE) (col. 19, lines 33-55: verifying the n-grams representing keys (field name) match the n-grams in the document, and the number of verified matching keys may be used as the key location similarity metric to select the best matched of the reference document); and
train a neural network model based on the tokenized words in the N-gram, wherein the trained neural network model is employed for automatically extracting relevant data from documents (col. 9, line 63 – col. 10, line 25: the preliminary content extraction tools/models may include one or more machine learning models trained to perform optical character recognition (OCR) and the related-content group descriptor (RGD) extracted from the target document image may comprise text tokens, e.g., characters or words and their bounding boxes (coordinates).
Appalaraju discloses extracting content from an image of a document. Appalaraju, however, does not disclose the input document is a predetermined format type.
In the same field of endeavor, Zuev discloses providing mechanisms for identification of text fields in electronic documents using neural networks, wherein the mechanisms can automatically detect text fields contained in an electronic document and associates each of the text fields with a field type (paragraph [0017]). Zuev further discloses electronic document may refer to a file comprising one or more digital content items that may be visually rendered to provide a visual representation of the electronic documents, and an electronic document may conform to any suitable file format, such as PDF, DOC, ODT, etc. (paragraph [0018]).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Appalaraju to include an electronic document in any suitable file format, as taught by Zuev for the purpose of providing classifications of any type of documents to train a neural network (Zuev, paragraph [0019]).
Appalaraju and Zuev, however, do not disclose tokenize each word for assigning a token maker.
Booth discloses receiving text and processing the text into natural language processing (NLP) tokens, wherein an NLP token is generally a single word or a single character and is assigned several properties such as an identifier (token maker) (paragraph [0023] and [0037]).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, to modify the systems of Appalaraju and Zuev to include assigning a token marker for tokenized each word, as taught by Booth for the purpose of indicating a position of the token in the input text and identifying toke relationships based on token properties such as dependency label.
As to dependent claim 2, Appalaraju and Zuev disclose wherein the pre-defined format includes a Portable Document Format (PDF), or an image format, and wherein the predetermined format type of the input document is an XML file (Zuevl, paragraph [0018]).
As to dependent claim 3, Appalaraju discloses wherein the data extraction engine comprises an annotation unit executed by the processor and configured to render a Graphical User Interface (GUI) via a input unit for carrying out an annotation operation on the predetermined format type of the document, and wherein the annotation unit generates annotation data by copying text from a relevant field in the predetermined format type or the pre-defined format of the document and selecting a text field corresponding to the relevant field for pasting the copied data by using a rubber band technique, and wherein a rubber band technique is used to determine coordinates corresponding to the text field with the copied data, which is stored by the annotation unit in a database, and wherein the annotation data is used for generation of N-grams (col. 5, line 64 – col. 6, line 54; col. 7, lines 12-18).
As to dependent claim 4, Appalaraju and Zuev disclose wherein the data extraction engine comprises an N-gram generation and labelling unit executed by the processor and configured to determine entity text by analyzing neighboring words corresponding to the entity text from left, top right and bottom of the entity text present in the predetermined format type for generating N-grams, and wherein the neighboring words of the entity text are analyzed by applying a threshold distance measurement criterion from the entity text (Appalaraju, col. 4, lines 46 – col. 5, line 7; Zuevl, paragraph [0018]).
As to dependent claim 5, Appalaraju discloses wherein the N-gram generation and labelling unit changes the threshold distance to value -1, in the event it is determined that the neighboring words associated with the entity text are not available at the threshold distance from the entity text, to avoid blank spaces between the neighboring words and entity text (col. 13, line 45 – col. 14, line 2 and col. 15, lines 4-26).
As to dependent claim 6, Appalaraju discloses wherein the N-gram generation and labelling unit extracts five neighboring words from the left of the entity text, three neighboring words from the top of the entity text, two neighboring words from right of the entity text, and three neighboring words from bottom of the entity text (col. 11, lines 18-41).
As to dependent claim 7, Appalaraju and Zuev disclose wherein the N-gram generation and labelling unit is configured to determine one or more entity text features from the pre-determined format type of the documents for generation of N-grams, and wherein the text features include position of the entity text in the predetermined format type of the documents, and format of the entity text (Appalaraju, col. 4, lines 46 – col. 5, line 7; Zuevl, paragraph [0018]).
As to dependent claim 8, Appalaraju discloses wherein the position-based features of the entity text include the entity text present at top-left of the document, the entity text present at top-right of the document, the entity text present at bottom-left of the document or the entity text present at bottom-right of the document (col. 4, line 46 – col. 5, line 7).
As to dependent claim 9, Appalaraju and Zuev disclose wherein an N-gram generation and labelling unit is configured to label the generated N-grams by carrying out a matching operation, and wherein the matching operation is carried out by identifying the N-grams using the field value associated with a particular field in the predetermined format type document and the one or more coordinates along with annotation data, which are stored in the database (Appalaraju, col. 6, lines 20-54; Zuevl, paragraph [0018]).
As to dependent claim 10, Appalaraju discloses wherein based on determination of a match, the N-gram generation and labelling unit labels the N-grams with the field name, and wherein the N-gram generation and labelling unit is configured to label all the unmatched N-grams as 'others' (Appalaraju, col. 6, lines 20-54).
As to dependent claim 12, Appalaraju, Zuev and Booth disclose wherein the data extraction engine comprises a tokenization unit executed by the processor and configured to process the generated and labelled N-grams for carrying out a tokenization operation for tokenizing each N-gram and classifying each token with the token marker (Booth, paragraph [0023] and [0037]).
As to dependent claim 14, Appalaraju discloses wherein the data extraction model training unit is configured to convert the tokenized words in the N-gram into sequences and each tokenized word is assigned an integer, and wherein the sequence is padded such that each tokenized word is of a same length, and wherein the padded sequence of words is used as an input for training the neural network model for data extraction (col. 17, lines 20-29).
As to dependent claim 15, Appalaraju discloses wherein the data extraction engine comprises a model accuracy improvement unit executed by the processor and configured to communicate with a data extraction model training unit for improving accuracy of the neural network model in order to effectively extract data from documents, and wherein the model accuracy improvement unit is configured to receive inputs relating to the extracted data from a data extraction unit for improving accuracy of the neural network model (col. 4, lines 11-45).
As to dependent claim 16, Appalaraju and Zuev disclose wherein the model accuracy improvement unit is configured to generate negative N-grams by carrying out a comparison operation with the annotation data, and wherein the model accuracy improvement unit is configured to extract data fields present in the predetermined format type of the document using the trained neural network model and compare the extracted data fields with the annotated data stored in the database (Appalaraju, col. 4, lines 11-45; Zuevl, paragraph [0018]).
As to dependent claim 17, Appalaraju and Zuev disclose wherein in the event the model accuracy improvement unit determines that field values associated with the data fields do not match with the annotated data, then the N-grams that are generated are determined as negative N-grams and are labelled as 'others', and wherein one or more criteria are employed for determining the match including determining if distance between extracted fields and annotated data is minimal or within a pre-defined threshold, then the N-grams are not labelled as 'others', and if one or more keywords associated with the fields in the document are present in negative N-grams, then such N-grams are not considered as 'others', thereby avoiding any positive N-grams being labelled as 'others', and wherein the model accuracy improvement unit is configured to up-scale the generated N-grams for each field except for N-grams labelled as ‘other’ for eliminating imbalance in the N-grams and increasing accuracy of the neural network model (Appalaraju, col. 17, lines 20-29; Zuev, paragraph [0019]).
As to dependent claim 18, Appalaraju discloses wherein the model accuracy improvement unit is configured to determine a confidence score for the field values present in the document based on predictions made by the neural network model, and wherein in the event the neural network model predicts two or more values for a particular field in the document, then the model accuracy improvement unit is configured to filter the values based on the confidence score, and wherein the model accuracy improvement unit considers the values with maximum confidence score (col. 13, line 45 – col. 14, line 54).
Claims 19-28 are method claims that contain similar limitations of claims 1, 2, 4,5, 7, 9, 10, 14, 16 and 17, respectively. Therefore, claims 19-28 are rejected under the same rationale.
Claim 29 is a computer program product that contains similar limitations of claim 1. Therefore, claim 29 is rejected under the same rationale.
Claims 11 and 13 are rejected under 35 U.S.C. 103 as being unpatentable over Appalaraju, Zuev and Booth as discussed in claim 1 above, and further in view of Okamoto, US Patent Application Publication No. US 2021/0125106 A1.
As to dependent claim 11, Appalaraju and Zuev disclose wherein the data extraction engine comprises a post processing unit executed by the processor and configured to process the predetermined format type of the documents for converting all the numeric values present in the document to a machine-readable format (Appalaraju, col. 6, lines 20-54; Zuevl, paragraph [0018]). Appalaraju and Zuev, however, do not disclose converting the numeric values to a constant, and wherein replacing all the numeric values with the constant reduces variation in patterns of numeric values and the neural network model is trained with accuracy.
In the same field of endeavor, Okamoto discloses the accuracy of the model to change when the type of the model trained with various types of features of learning data is different (Okamoto, paragraph [0200]). Okamoto further discloses when the learning data is strings, the index generating unit 42 generates a generation index based on the number of types of the strings included in the learning data. Furthermore, when the learning data is floating-point numbers, the index generating unit 42 generates a conversion index for converting the learning data into the input data to be input to a model, as a model generation index, for example, the index generating unit 42 determines whether to bucketize floating-point numbers, which range of values is to be classified into which bucket and the like. The index generating unit 42 may determine whether to make the range of values corresponding to each bucket constant, whether to make the number of pieces of learning data to be classified into each bucket constant (Okamoto, paragraph [0245]).
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the systems of Appalaraju and Zuev to incorporate converting the numeric values to a constant, and wherein replacing all the numeric values with the constant reduces variation in patterns of numeric values and the neural network model is trained with accuracy, as taught by Okamoto for the purpose of optimizing an artificial intelligence model.
As to dependent claim 13, Appalaraju and Okamoto disclose wherein the data extraction engine comprises a data extraction model training unit executed by the processor and configured to receive the tokenized words from a tokenization unit to train the neural network model (Appalaraju, col. 9, line 63 – col. 10, line 25), and wherein the data extraction model training unit employs a keras library to train a recurrent neural network model based on bi-directional Long Short-term Memory (bi-LSTM) technique (Okamoto, paragraph [0199]).
Response to Arguments
In the Remarks, Applicant argues in substance that
Applicant submits that the amended claim 1 includes sufficient elements to clearly recite that even assuming judicial exception subsists in the claims, in fact, the exception is integrated into a practical application as required by the 2019 PEG.
In reply to this argument, Examiner disagree. Claim 1 is directed to a system for optimized training of a neural network model for data extraction, the system comprising the limitations of: extract words from a structured or semi-structured input document for generating a pre-determined format type of the input document, wherein the extracted word include an entity text and neighboring words associated with the entity text; generate N-grams by analyzing … in a predefined order; label the N-grams with a field name by comparing the generated N-grams with the coordinates corresponding to the words; tokenize each word in the labelled N-gram assigning a token marker. These limitations are directed towards the abstract idea which is performed manually by a human being and/or by paper and pen. These limitations are directed towards the abstract idea of a mathematical relationship, specially organizing information and manipulation information through mathematical correlations.
The claim recites additional elements of “memory”, “processor”, “data extraction engine”, “train neural network model” and “the trained neural network model is employed for automatically extracting relevant data from documents.”
These limitations amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). The additional limitations “memory”, “processor”, and “data extraction engine” amount to no more than generally linking the use of the judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). It recites a generic computer or generic computer components that merely act as a tool on which the method operates, and thus, fails to integrate the exception into a practical application. In addition, it recites using neural network (NN) model for extracting data from documents, train the neural network model based on the extracted data and the trained neural network model extract relevant data from documents. Although Applicant recites “trained neural network model is employed for automatically extracting relevant data from document”, a human being can extract data from a document, observe a model with the extracted data, and use the model to extract relevant data from document and thus, fails to integrate the exception into a practical application.
The cited prior art does not disclose “extract words from a structured or semi-structured input document for generating a pre-determined format type of the input document and store the extracted words in a database along with spatial coordinates corresponding to each extracted word, wherein the extracted words include an entity text and neighboring words associated with the entity text”.
In reply to this argument, Appalaraju discloses in col. 2, line 45 – col. 3, line 7: extract content about related entities such as key-value pairs from images or other representations of target data objects, using machine learning techniques for layout-aware comparisons between a set of pre-obtained reference objects and the target data objects, wherein many data objects whose content is to be extracted in an automated manner, and wherein type of information that has to be included in an income statement, a tax-related form, a medical claim, etc., may be defined by regulations or customs of the organizations involved in preparing or generating the document. To extract the information from previously unseen target documents of the more frequently encountered categories with high accuracy if (a) properties of the different types of documents expected to be analyzed frequently can be collected and stored in advance, and (b) a reasonable probabilistic prediction can be made using the collected properties about the particular document type to which a target document is likely to belong. Appalaraju further discloses in col. 8, line 53 – col. 9, line 27: Figure 3 and col. 12, line 54 – col. 13, line 8: preliminary content extraction may first be performed on a scanned/photographed target document, the target document may be provided as input to one or more machine learning models such as optical character recognition (OCR) models, and output comprising detected characters, words, numbers and the like, along with their respective bounding boxes such as approximate location/position information expressed using bounding box coordinates. Appalaraju further discloses in col. 11, line 55 and col. 12, line 35 and Figure 2: the extracted content elements contain keys such as name, date of birth, contact information and these are considered as entity text, wherein some keys such as “contact information” may comprise several lower-level keys (neighboring words) such as Street , City, state, and the like. Thus, Appalaraju discloses “extract words from a structured or semi-structured input document for generating a pre-determined format type of the input document and store the extracted words in a database along with spatial coordinates corresponding to each extracted word, wherein the extracted words include an entity text and neighboring words associated with the entity text”.
The cited prior art does not disclose “label the N-grams with a filed name by comparing the generated N-grams with the coordinates corresponding to the words, wherein if the entity text is present in more than one place within the document, the coordinates identify the entity text and label the corresponding N-gram with the correct field name without dependence of field values”.
In reply to this argument, Appalaraju discloses in col. 11, lines 18-41: the objects being analyzed are text-containing documents, the structural comparison may comprise a preliminary step of identifying corresponding content elements, the corresponding content elements may be identified using n-gram analysis in which n-grams of a chosen token length that are present within both the target document and the reference document are found). Appalaraju further discloses in col. 4, line 46 – col. 5, line 7: a particular key-value descriptor (entity text) may comprise information about at least a text key within an image of a reference document, wherein a key-value descriptor may include text tokens or words that makeup a key, and a bounding box (expressed using a pair of x, y coordinates representing opposite corners of a rectangle) within which the text tokens/words were identified. In addition, key-value-group descriptors may be stored for reference documents, representing scenarios in which multiple values (words) related to a single key may be expected to be present in a given document, as such, a key-value descriptor (entity text) may represent one example of a key-value-group descriptor). Thus, Appalaraju discloses “label the N-grams with a filed name by comparing the generated N-grams with the coordinates corresponding to the words, wherein if the entity text is present in more than one place within the document, the coordinates identify the entity text and label the corresponding N-gram with the correct field name without dependence of field values”.
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37CFR 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 CHAU T NGUYEN whose telephone number is (571)272-4092. The examiner can normally be reached on Monday-Friday from 8am to 5pm.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Cesar Paula, can be reached at telephone number 5712724128. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
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) Form at https://www.uspto.gov/patents/uspto-automated-interview-request-air-form.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
/CHAU T NGUYEN/Primary Examiner, Art Unit 2145
Prosecution Insights