DETAILED ACTION
This communication is in response to the Application No. 18/426,510 filed on January 30, 2024
in which Claims 1-15 are presented for examination
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 .
Specification
The disclosure is objected to because of the following informalities:
The specification's paragraph numbers aren't consecutive, [0001] through [0005] each appear twice.
Paragraphs numbered [0001] and [0002] reappear after [0022] (the FIG. 1 discussion “The system 100 includes a processing subsystem 102..." and "Further, in another embodiment, the network 106...").
A paragraph numbered [0003] appears after [0032] ("the marked differences are reviewed in the second verification level...").
Two paragraphs numbered [0004] and [0005] appear after [0034] (the "data points may include...dates, numbers, long text" paragraph and the "user Y inputs a legal document" example)
Appropriate correction is required.
Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is invoked.
As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph:
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable
presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action.
This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because the claim limitation(s) uses a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitation(s) is/are:
Such claim limitation(s) is/are:
“Data extraction module” in Claims 1 - 11
“Verification module” in Claims 1 - 11
“Comparison module” in Claims 1 - 11
“Quality check module” in Claims 1 - 11
Because this/these claim limitation(s) is/are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, it/they is/are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof.
If applicant does not intend to have this/these limitation(s) interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, applicant may: (1) amend the claim limitation(s) to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitation(s) recite(s) sufficient structure to perform the claimed function so as to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1–11 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim limitations “data extraction module” in Claim 1, “verification module” in Claim 1, “comparison module” in Claim 1, and “quality check module” in Claim 1 invoke 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. However, the written description fails to disclose the corresponding structure, material, or acts for performing the entire claimed function and to clearly link the structure, material, or acts to the function. The aforementioned claim limitations are simply recited in Applicant’s specification (see, for example, instant specification Par. [0024]-[0035]) without providing sufficient structure for performing the claimed functions. Therefore, the claims are indefinite and rejected under 35 U.S.C. 112(b) or pre-AIA 35 U.S.C. 112, second paragraph.
Applicant may:
(a) Amend the claim so that the claim limitation will no longer be interpreted as a limitation under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph;
(b) Amend the written description of the specification such that it expressly recites what structure, material, or acts perform the entire claimed function, without introducing any new matter (35 U.S.C. 132(a)); or
(c) Amend the written description of the specification such that it clearly links the structure, material, or acts disclosed therein to the function recited in the claim, without introducing any new matter (35 U.S.C. 132(a)).
If applicant is of the opinion that the written description of the specification already implicitly or inherently discloses the corresponding structure, material, or acts and clearly links them to the function so that one of ordinary skill in the art would recognize what structure, material, or acts perform the claimed function, applicant should clarify the record by either:
(a) Amending the written description of the specification such that it expressly recites the corresponding structure, material, or acts for performing the claimed function and clearly links or associates the structure, material, or acts to the claimed function, without introducing any new matter (35 U.S.C. 132(a)); or
(b) Stating on the record what the corresponding structure, material, or acts, which are implicitly or inherently set forth in the written description of the specification, perform the claimed function.
For more information, see 37 CFR 1.75(d) and MPEP §§ 608.01(o) and 2181.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claim 1-20 are rejected under 35 U.S.C. 101 because these claimed inventions are directed to an abstract idea without significantly more.
Regarding Claim 1:
Step 1: Claim 1 is a system type claim. Therefore, Claims 1-11 fall within one of the four statutory
categories (i.e., process, machine, manufacture, or composition of matter).
2A Prong 1: If a claim limitation, under its broadest reasonable interpretation, covers performance
of the limitation in the mind but for the recitation of generic computer components, then it falls within
the “Mental Processes” grouping of abstract ideas. If a claim limitation, under its broadest reasonable
interpretation, covers performance of the limitation by mathematical calculation but for the recitation
of generic computer components, then it falls within the “Mathematical Concepts” grouping of abstract
ideas.
parse one or more documents to extract a plurality of data points […] to generate a first transaction (mental process – parsing one or more documents to extract a plurality of data points and generate a first transaction may be performed mentally or using pen and paper by a user reading/analyzing the documents, identifying and extracting relevant data points, and accordingly using judgment/evaluation to organize the extracted data points into a first transaction based on said analysis)
and generate a plurality of sub-transactions from the first transaction wherein each of the plurality of sub-transactions comprises the plurality of data points (mental process – generating a plurality of sub-transactions from the first transaction may be performed mentally or using pen and paper by a user reading/analyzing the first transaction, separating and organizing the transaction into multiple sub-transactions, and accordingly using judgment/evaluation to include the relevant data points in each sub-transaction based on said analysis)
[…] to verify each of the plurality of sub-transactions individually by a first user and a second user respectively thereby resulting in a first level of verification (mental process - verifying each of the plurality of sub-transactions individually by a first user and a second user may be performed mentally or using pen and paper by each user reviewing/analyzing the data points in each sub-transaction, determining whether the data is accurate, and accordingly using judgment/evaluation to record their respective verification results, thereby resulting in a first level of verification)
compare the verified results of the plurality of sub-transactions to identify a plurality of differences (mental process - comparing the verified results of the plurality of sub-transactions to identify a plurality of differences may be performed mentally or using pen and paper by a user reviewing/analyzing the respective verification results, comparing the results with one another, and accordingly using judgment/evaluation to identify and record differences between the verified results)
and filter and mark the plurality of differences in the verified results (mental process - filtering and marking the plurality of differences in the verified results may be performed mentally or using pen and paper by a user reviewing/analyzing the identified differences, determining which differences satisfy selected criteria, and accordingly using judgment/evaluation to separate and mark those differences based on said analysis)
generate a second transaction […] wherein the second transaction is based on the marked differences of the verified results(mental process - generating a second transaction based on the marked differences of the verified results may be performed mentally or using pen and paper by a user reviewing/analyzing the marked differences, determining the information associated with those differences, and accordingly using judgment/evaluation to organize said information into a second transaction based on said analysis)
assign the second transaction with the marked differences to a third user for a subsequent review thereby resulting in a second verification level (mental process - assigning the second transaction with the marked differences to a third user for subsequent review may be performed mentally or using pen and paper by a user identifying the second transaction and associated marked differences, selecting a third user to review the information, and accordingly using judgment/evaluation to direct the transaction to the third user for further review, thereby resulting in a second verification level)
and update the plurality of datapoints in response to the review made by the third user thereby enabling quality check of the one or more documents over multiple levels of verification (mental process - updating the plurality of datapoints in response to the review made by the third user may be performed mentally or using pen and paper by a user reviewing/analyzing the third user’s verification results, determining which datapoints require modification, and accordingly using judgment/evaluation to revise the datapoints based on said review, thereby enabling quality checking over multiple levels of verification)
Step 2A Prong 2: This judicial exception is not integrated into a practical application.
a hardware processor; a memory coupled to the hardware processor […](recited at a high-level of generality (i.e., a generic processor, computer-readable storage medium, a communication interface, a user interface and memory) such that it amounts to no more than mere instructions to apply the exception using generic computer components)
a data extraction module (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by using machine learning model […] (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of using a machine learning model without significantly more)
a verification module operatively coupled to the data extraction module (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
a comparison module operatively coupled to the verification module (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
and a quality check module operatively coupled to the comparison module (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by using machine learning model […] (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of using a machine learning model without significantly more)
Step 2B: The claim does not include additional elements considered individually and in combination that
are sufficient to amount to significantly more than the judicial exception.
a hardware processor; a memory coupled to the hardware processor […] (recited at a high-level of generality (i.e., a generic processor, computer-readable storage medium, a communication interface, a user interface and memory) such that it amounts to no more than mere instructions to apply the exception using generic computer components)
a data extraction module (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by using machine learning model […] (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of using a machine learning model without significantly more)
a verification module operatively coupled to the data extraction module (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
a comparison module operatively coupled to the verification module (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
and a quality check module operatively coupled to the comparison module (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by using machine learning model […] (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of using a machine learning model without significantly more)
For the reasons above, Claim 1 is rejected as being directed to an abstract idea without
significantly more. This rejection applies equally to dependent claims 1 - 11. The additional limitations of the dependent claims are addressed below.
Regarding Claim 2:
Step 2A Prong 1: See the rejection of Claim 1 above, which Claim 2 depends on.
Step 2A Prong 2 & Step 2B:
wherein the first user, the second user and the third user are ignorant of each other identity thereby enabling the double blinded verification of data (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that the first user, the second user and the third user are ignorant of each other identity thereby enabling the double blinded verification of data does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
Accordingly, under Step 2A Prong 2 and Step 2B, this additional element does not integrate the
abstract idea into practical application because it does not impose any meaningful limits on practicing
the abstract idea, as discussed above in the rejection of claim 1. The claim does not include additional
elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
Regarding Claim 3:
Step 2A Prong 1: See the rejection of Claim 1 above, which Claim 3 depends on.
wherein the plurality of sub-transactions are merged to generate the second transaction upon verification of the said plurality of sub-transactions (mental process – merging the plurality of sub-transactions to generate the second transaction upon verification may be performed mentally or using pen and paper by a user reviewing/analyzing the verified sub-transactions, combining the information from the verified sub-transactions, and accordingly using judgment/evaluation to organize the combined information into a second transaction based on said analysis)
Step 2A Prong 2 & Step 2B:
Accordingly, under Step 2A Prong 2 and Step 2B, there are no additional elements that integrate the abstract idea into practical application. The claim does not include additional elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
Regarding Claim 4:
Step 2A Prong 1: See the rejection of Claim 1 above, which Claim 4 depends on.
wherein the verification module is configured to enable the first user and the second user to tag an uncertain data at the time of verification to enable the third user to validate the tagged uncertain data (mental process – tagging uncertain data during verification and subsequently validating the tagged uncertain data may be performed mentally or using pen and paper by the first and second users reviewing/analyzing the data, identifying data considered uncertain, and marking such data, and by the third user reviewing/analyzing the marked data and accordingly using judgment/evaluation to validate the uncertain data)
Step 2A Prong 2 & Step 2B:
Accordingly, under Step 2A Prong 2 and Step 2B, there are no additional elements that integrate the abstract idea into practical application. The claim does not include additional elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
Regarding Claim 5:
Step 2A Prong 1: See the rejection of Claim 1 above, which Claim 5 depends on.
Step 2A Prong 2 & Step 2B:
wherein the third user is allowed to view the extracted, verified, or updated data at the first level of verification and the second level of verification (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that the third user is allowed to view the extracted, verified, or updated data at the first level of verification and the second level of verification does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
Accordingly, under Step 2A Prong 2 and Step 2B, this additional element does not integrate the
abstract idea into practical application because it does not impose any meaningful limits on practicing
the abstract idea, as discussed above in the rejection of claim 1. The claim does not include additional
elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
Regarding Claim 6:
Step 2A Prong 1: See the rejection of Claim 1 above, which Claim 6 depends on.
Step 2A Prong 2 & Step 2B:
wherein the data points comprises dates, numbers, long text, short text, currencies, and percentages (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that the data points comprises dates, numbers, long text, short text, currencies, and percentages does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
Accordingly, under Step 2A Prong 2 and Step 2B, this additional element does not integrate the
abstract idea into practical application because it does not impose any meaningful limits on practicing
the abstract idea, as discussed above in the rejection of claim 1. The claim does not include additional
elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
Regarding Claim 7:
Step 2A Prong 1: See the rejection of Claim 1 above, which Claim 7 depends on.
Step 2A Prong 2 & Step 2B:
wherein the data points are customized by using a plurality of keywords and forms based on the search requirements of the user (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that the data points are customized by using a plurality of keywords and forms based on the search requirements of the user does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
Accordingly, under Step 2A Prong 2 and Step 2B, this additional element does not integrate the
abstract idea into practical application because it does not impose any meaningful limits on practicing
the abstract idea, as discussed above in the rejection of claim 1. The claim does not include additional
elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
Regarding Claim 8:
Step 2A Prong 1: See the rejection of Claim 1 above, which Claim 8 depends on.
Step 2A Prong 2 & Step 2B:
wherein if the verification module finds no differences in the plurality of sub-transactions, an output of the first verification is considered as a final output (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that if the verification module finds no differences in the plurality of sub-transactions, an output of the first verification is considered as a final output does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
Accordingly, under Step 2A Prong 2 and Step 2B, this additional element does not integrate the
abstract idea into practical application because it does not impose any meaningful limits on practicing
the abstract idea, as discussed above in the rejection of claim 1. The claim does not include additional
elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
Regarding Claim 9:
Step 2A Prong 1: See the rejection of Claim 1 above, which Claim 9 depends on.
Step 2A Prong 2 & Step 2B:
a database […] (recited at a high-level of generality (i.e., a generic processor, computer-readable storage medium, a communication interface, a user interface and memory, a database) such that it amounts to no more than mere instructions to apply the exception using generic computer components)
[…] to store the verified results along with the corresponding differences (insignificant extra-solution activity — mere storage of data; adding insignificant extra-solution activity to the judicial exception — see MPEP 2106.05(g))
Accordingly, under Step 2A Prong 2 and Step 2B, this additional element does not integrate the
abstract idea into practical application because it does not impose any meaningful limits on practicing
the abstract idea, as discussed above in the rejection of claim 1. The claim does not include additional
elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
Regarding Claim 10:
Step 2A Prong 1: See the rejection of Claim 1 above, which Claim 10 depends on.
Step 2A Prong 2 & Step 2B:
wherein the one or more documents includes unstructured data (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that the one or more documents includes unstructured data does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
Accordingly, under Step 2A Prong 2 and Step 2B, this additional element does not integrate the
abstract idea into practical application because it does not impose any meaningful limits on practicing
the abstract idea, as discussed above in the rejection of claim 1. The claim does not include additional
elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
Regarding Claim 11:
Step 2A Prong 1: See the rejection of Claim 1 above, which Claim 11 depends on.
Step 2A Prong 2 & Step 2B:
wherein the plurality of differences are marked by highlighting the differences in colour coded format (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that the plurality of differences are marked by highlighting the differences in colour coded format does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
Accordingly, under Step 2A Prong 2 and Step 2B, this additional element does not integrate the
abstract idea into practical application because it does not impose any meaningful limits on practicing
the abstract idea, as discussed above in the rejection of claim 1. The claim does not include additional
elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
Regarding Claim 12:
Step 1: Claim 12 is a method type claim. Therefore, Claims 12-14 fall within one of the four statutory
categories (i.e., process, machine, manufacture, or composition of matter).
2A Prong 1: If a claim limitation, under its broadest reasonable interpretation, covers performance
of the limitation in the mind but for the recitation of generic computer components, then it falls within
the “Mental Processes” grouping of abstract ideas. If a claim limitation, under its broadest reasonable
interpretation, covers performance of the limitation by mathematical calculation but for the recitation
of generic computer components, then it falls within the “Mathematical Concepts” grouping of abstract
ideas.
parsing […] one or more documents to extract a plurality of data points […] to generate a first transaction (mental process – parsing one or more documents to extract a plurality of data points and generate a first transaction may be performed mentally or using pen and paper by a user reading/analyzing the documents, identifying and extracting relevant data points, and accordingly using judgment/evaluation to organize the extracted data points into a first transaction based on said analysis)
generating […] a plurality of sub-transactions from the first transaction wherein each of the plurality of sub-transactions comprises the plurality of data points (mental process – generating a plurality of sub-transactions from the first transaction may be performed mentally or using pen and paper by a user reading/analyzing the first transaction, separating and organizing the transaction into multiple sub-transactions, and accordingly using judgment/evaluation to include the relevant data points in each sub-transaction based on said analysis)
verifying […] each of the plurality of sub-transactions individually by a first user and a second user respectively thereby resulting in a first level of verification (mental process - verifying each of the plurality of sub-transactions individually by a first user and a second user may be performed mentally or using pen and paper by each user reviewing/analyzing the data points in each sub-transaction, determining whether the data is accurate, and accordingly using judgment/evaluation to record their respective verification results, thereby resulting in a first level of verification)
comparing […] the verified results of the plurality of sub-transactions to identify a plurality of differences (mental process - comparing the verified results of the plurality of sub-transactions to identify a plurality of differences may be performed mentally or using pen and paper by a user reviewing/analyzing the respective verification results, comparing the results with one another, and accordingly using judgment/evaluation to identify and record differences between the verified results)
filtering and marking […] the plurality of differences in the verified results (mental process - filtering and marking the plurality of differences in the verified results may be performed mentally or using pen and paper by a user reviewing/analyzing the identified differences, determining which differences satisfy selected criteria, and accordingly using judgment/evaluation to separate and mark those differences based on said analysis)
generating […] a second transaction […] wherein the second transaction is based on the marked differences of the verified results(mental process - generating a second transaction based on the marked differences of the verified results may be performed mentally or using pen and paper by a user reviewing/analyzing the marked differences, determining the information associated with those differences, and accordingly using judgment/evaluation to organize said information into a second transaction based on said analysis)
assigning […] the second transaction with the marked differences to a third user for a subsequent review thereby resulting in a second verification level (mental process - assigning the second transaction with the marked differences to a third user for subsequent review may be performed mentally or using pen and paper by a user identifying the second transaction and associated marked differences, selecting a third user to review the information, and accordingly using judgment/evaluation to direct the transaction to the third user for further review, thereby resulting in a second verification level)
and updating […] the plurality of datapoints in response to the review made by the third user thereby enabling quality check of the one or more documents over multiple levels of verification (mental process - updating the plurality of datapoints in response to the review made by the third user may be performed mentally or using pen and paper by a user reviewing/analyzing the third user’s verification results, determining which datapoints require modification, and accordingly using judgment/evaluation to revise the datapoints based on said review, thereby enabling quality checking over multiple levels of verification)
Step 2A Prong 2: This judicial exception is not integrated into a practical application.
a hardware processor; a memory coupled to the hardware processor […](recited at a high-level of generality (i.e., a generic processor, computer-readable storage medium, a communication interface, a user interface and memory) such that it amounts to no more than mere instructions to apply the exception using generic computer components)
[…] by a data extraction module of a processing subsystem […] (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by using machine learning model […] (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of using a machine learning model without significantly more)
[…] by the data extraction module of the processing subsystem […] (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by a verification module of the processing subsystem […] (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by a comparison module of the processing subsystem […](recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by the comparison module of the processing subsystem […](recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by a quality check module of the processing subsystem […] (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by using machine learning model […] (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of using a machine learning model without significantly more)
[…] by a quality check module of the processing subsystem […](recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by a quality check module of the processing subsystem […](recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
Step 2B: The claim does not include additional elements considered individually and in combination that
are sufficient to amount to significantly more than the judicial exception.
a hardware processor; a memory coupled to the hardware processor […](recited at a high-level of generality (i.e., a generic processor, computer-readable storage medium, a communication interface, a user interface and memory) such that it amounts to no more than mere instructions to apply the exception using generic computer components)
[…] by a data extraction module of a processing subsystem […] (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by using machine learning model […] (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of using a machine learning model without significantly more)
[…] by the data extraction module of the processing subsystem […] (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by a verification module of the processing subsystem […] (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by a comparison module of the processing subsystem […](recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by the comparison module of the processing subsystem […](recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by a quality check module of the processing subsystem […] (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by using machine learning model […] (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of using a machine learning model without significantly more)
[…] by a quality check module of the processing subsystem […]( (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by a quality check module of the processing subsystem […]( (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
For the reasons above, Claim 12 is rejected as being directed to an abstract idea without
significantly more. This rejection applies equally to dependent claims 12 - 14. The additional limitations of the dependent claims are addressed below.
Regarding Claim 13:
Step 2A Prong 1: See the rejection of Claim 12 above, which Claim 13 depends on.
tagging an uncertain data by the first user and the second user at the time of verification to enable the third user to validate the tagged uncertain data (mental process – tagging uncertain data during verification and subsequently validating the tagged uncertain data may be performed mentally or using pen and paper by the first and second users reviewing/analyzing the data, identifying data considered uncertain, and marking such data, and by the third user reviewing/analyzing the marked data and accordingly using judgment/evaluation to validate the uncertain data)
Step 2A Prong 2 & Step 2B:
Accordingly, under Step 2A Prong 2 and Step 2B, there are no additional elements that integrate the abstract idea into practical application. The claim does not include additional elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
Regarding Claim 14:
Step 2A Prong 1: See the rejection of Claim 12 above, which Claim 14 depends on.
Step 2A Prong 2 & Step 2B:
viewing, the extracted datapoints and values of a plurality of attributes of the extracted datapoints, wherein the values are verified or updated at a previous verification level (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that viewing, the extracted datapoints and values of a plurality of attributes of the extracted datapoints, wherein the values are verified or updated at a previous verification level does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
Accordingly, under Step 2A Prong 2 and Step 2B, this additional element does not integrate the
abstract idea into practical application because it does not impose any meaningful limits on practicing
the abstract idea, as discussed above in the rejection of claim 12. The claim does not include additional
elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
Regarding Claim 15:
Step 1: Claim 15 is a non-transitory computer medium type claim. Therefore, Claim 15 falls within one of the four statutory categories (i.e., process, machine, manufacture, or composition of matter).
2A Prong 1: If a claim limitation, under its broadest reasonable interpretation, covers performance
of the limitation in the mind but for the recitation of generic computer components, then it falls within
the “Mental Processes” grouping of abstract ideas. If a claim limitation, under its broadest reasonable
interpretation, covers performance of the limitation by mathematical calculation but for the recitation
of generic computer components, then it falls within the “Mathematical Concepts” grouping of abstract
ideas.
parsing […] one or more documents to extract a plurality of data points […] to generate a first transaction […](mental process – parsing one or more documents to extract a plurality of data points and generate a first transaction may be performed mentally or using pen and paper by a user reading/analyzing the documents, identifying and extracting relevant data points, and accordingly using judgment/evaluation to organize the extracted data points into a first transaction based on said analysis)
generating […] a plurality of sub-transactions from the first transaction wherein each of the plurality of sub-transactions comprises the plurality of data points (mental process – generating a plurality of sub-transactions from the first transaction may be performed mentally or using pen and paper by a user reading/analyzing the first transaction, separating and organizing the transaction into multiple sub-transactions, and accordingly using judgment/evaluation to include the relevant data points in each sub-transaction based on said analysis)
verifying […] each of the plurality of sub-transactions individually by a first user and a second user respectively thereby resulting in a first level of verification (mental process - verifying each of the plurality of sub-transactions individually by a first user and a second user may be performed mentally or using pen and paper by each user reviewing/analyzing the data points in each sub-transaction, determining whether the data is accurate, and accordingly using judgment/evaluation to record their respective verification results, thereby resulting in a first level of verification)
comparing […] the verified results of the plurality of sub-transactions to identify a plurality of differences (mental process - comparing the verified results of the plurality of sub-transactions to identify a plurality of differences may be performed mentally or using pen and paper by a user reviewing/analyzing the respective verification results, comparing the results with one another, and accordingly using judgment/evaluation to identify and record differences between the verified results)
filtering and marking […] the plurality of differences in the verified results (mental process - filtering and marking the plurality of differences in the verified results may be performed mentally or using pen and paper by a user reviewing/analyzing the identified differences, determining which differences satisfy selected criteria, and accordingly using judgment/evaluation to separate and mark those differences based on said analysis)
generating […] a second transaction […] wherein the second transaction is based on the marked differences of the verified results(mental process - generating a second transaction based on the marked differences of the verified results may be performed mentally or using pen and paper by a user reviewing/analyzing the marked differences, determining the information associated with those differences, and accordingly using judgment/evaluation to organize said information into a second transaction based on said analysis)
assigning […] the second transaction with the marked differences to a third user for a subsequent review thereby resulting in a second verification level (mental process - assigning the second transaction with the marked differences to a third user for subsequent review may be performed mentally or using pen and paper by a user identifying the second transaction and associated marked differences, selecting a third user to review the information, and accordingly using judgment/evaluation to direct the transaction to the third user for further review, thereby resulting in a second verification level)
and updating […] the plurality of datapoints in response to the review made by the third user thereby enabling quality check of the one or more documents over multiple levels of verification (mental process - updating the plurality of datapoints in response to the review made by the third user may be performed mentally or using pen and paper by a user reviewing/analyzing the third user’s verification results, determining which datapoints require modification, and accordingly using judgment/evaluation to revise the datapoints based on said review, thereby enabling quality checking over multiple levels of verification)
Step 2A Prong 2: This judicial exception is not integrated into a practical application.
a hardware processor; a memory coupled to the hardware processor […](recited at a high-level of generality (i.e., a generic processor, computer-readable storage medium, a communication interface, a user interface and memory) such that it amounts to no more than mere instructions to apply the exception using generic computer components)
[…] by a data extraction module of a processing subsystem […] (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by using machine learning model […] (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of using a machine learning model without significantly more)
[…]wherein the one or more documents includes unstructured data (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that the one or more documents includes unstructured data does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
[…] by the data extraction module of the processing subsystem […] (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by a verification module of the processing subsystem […] (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by a comparison module of the processing subsystem […] (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by the comparison module of the processing subsystem […](recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by a quality check module of the processing subsystem […] (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by using machine learning model […] (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of using a machine learning model without significantly more)
[…] by a quality check module of the processing subsystem […] (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by a quality check module of the processing subsystem […] (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
Step 2B: The claim does not include additional elements considered individually and in combination that
are sufficient to amount to significantly more than the judicial exception.
a hardware processor; a memory coupled to the hardware processor […](recited at a high-level of generality (i.e., a generic processor, computer-readable storage medium, a communication interface, a user interface and memory) such that it amounts to no more than mere instructions to apply the exception using generic computer components)
[…] by a data extraction module of a processing subsystem […] (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by using machine learning model […] (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of using a machine learning model without significantly more)
[…]wherein the one or more documents includes unstructured data (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that the one or more documents includes unstructured data does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
[…] by the data extraction module of the processing subsystem […] (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by a verification module of the processing subsystem […] (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by a comparison module of the processing subsystem […](recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by the comparison module of the processing subsystem […](recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by a quality check module of the processing subsystem […](recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by using machine learning model […] (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of using a machine learning model without significantly more)
[…] by a quality check module of the processing subsystem […](recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
[…] by a quality check module of the processing subsystem […] (recited at a high-level of generality as a generic computer/software component such that it amounts to no more than mere instructions to apply the judicial exception using generic computer components)
For the reasons above, Claim 15 is rejected as being directed to an abstract idea without
significantly more.
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 1 and 3-15 are rejected under 35 U.S.C. 103 as being unpatentable over Malaney et al. (hereafter Malaney) (US 20070118391) in view of Bobley et al. (hereinafter Bobley) (US 11087409).
Regarding Claim 1, Malaney teaches:
a data extraction module (Malaney, Par. [0131], “The present invention, in preferred embodiments, utilizes Location Diagram concepts and integrates multiple components including image processing, intelligent collation, feedback learning, a document classifier, a verifier, a versioning engine, an information locator, a data extractor, a data scrubber, and manual collaboration”, thus a data extraction module is disclosed) configured to:
parse one or more documents to extract a plurality of data points […] (Malaney, Par. [0201], “In preferred embodiments of the invention, extraction of information or data from the documents or Dox Package that has been captured using the method or apparatus of the invention, and preferably extraction is first done automatically from readily identifiable fields 221 and image snippets of other fields location are re-submitted to the OCR step with a field specific dictionary before repeating the extraction process 220”, & Par. [0206], “The processes preferred embodiments of the instant invention typically extracts the fields (as they appear in the document) required for various business and/or compliance requirements, then transforms them into facts that can be used further for decision making by an automated rules engine or search engine”, thus parse one or more documents to extract a plurality of data points […] is disclosed because Malaney teaches automatically processing captured documents to extract information from multiple identifiable fields for further processing. Malaney’s documents or Dox Package correspond to one or more documents. Malaney’s automatic extraction of information or data from document fields corresponds to parse one or more documents to extract. Malaney’s multiple extracted fields correspond to a plurality of data points because the system extracts multiple pieces of information from the documents and transforms them into facts for subsequent use)
a verification module operatively coupled to the data extraction module wherein the verification module is configured to verify each of […the plurality of sub-transactions…] individually by a first user and a second user respectively thereby resulting in a first level of verification (Malaney, Par. [0131], “The present invention, in preferred embodiments, utilizes Location Diagram concepts and integrates multiple components including image processing, intelligent collation, feedback learning, a document classifier, a verifier, a versioning engine, an information locator, a data extractor, a data scrubber, and manual collaboration”, & Par. [0204], “Human collaborators can verify and/or change the data and/or extracted information in reference to the Dox Package. In preferred embodiments, each field subjected to MEP is extracted by a minimum of two human collaborators and the system compares the extracted value. In the event of a discrepancy, the value in question can be sent to additional human collaborators”, thus a verification module operatively coupled to the data extraction module wherein the verification module is configured to verify each of […the plurality of sub-transactions…] individually by a first user and a second user respectively thereby resulting in a first level of verification is disclosed because Malaney teaches an integrated system including a data extractor, verifier, and manual collaboration, and further teaches that each individual field is handled by at least two human collaborators who verify or change the extracted data. Malaney’s integrated data extractor and verifier correspond to the verification module being operatively coupled to the data extraction module. Malaney’s individual fields subjected to manual extraction correspond to the plurality of sub-transactions. The minimum of two human collaborators correspond to the first user and second user, respectively. Their independent verification of each field corresponds to verifying each sub-transaction individually, and this two-user verification before any discrepancy is escalated to additional collaborators corresponds to the first level of verification)
a comparison module operatively coupled to the verification module (Malaney, Par. [0131], “The present invention, in preferred embodiments, utilizes Location Diagram concepts and integrates multiple components including image processing, intelligent collation, feedback learning, a document classifier, a verifier, a versioning engine, an information locator, a data extractor, a data scrubber, and manual collaboration”, & Par. [0204], “In preferred embodiments, each field subjected to MEP is extracted by a minimum of two human collaborators and the system compares the extracted value”, thus a comparison module operatively coupled to the verification module is disclosed) wherein the comparison module is configured to:
compare the verified results of […the plurality of sub-transactions…] to identify a plurality of differences (Malaney, Par. [0204], “In preferred embodiments, each field subjected to MEP is extracted by a minimum of two human collaborators and the system compares the extracted value. In the event of a discrepancy, the value in question can be sent to additional human collaborators”, & Par. [0192 – 0197], “In preferred embodiments of the instant invention, the collation process is based on incremental learning and various artificial intelligence ("AI")-based techniques, which may include one or more of the following, such as: (1) the Location Diagram- and Feature Vector-based feature extraction and page mapping; (2) SVM and NLP; (3) an intelligent filter technique taking advantage of header and footer based information; (4) collation by finding common threads within or between pages, documents, or sets of documents; (5) finding disagreements based on affinities”, thus compare the verified results of […the plurality of sub-transactions…] to identify a plurality of differences is disclosed because Malaney teaches comparing values obtained from at least two human collaborators and identifying discrepancies between those values. Malaney’s individual fields subjected to MEP correspond to the plurality of sub-transactions. Malaney’s comparison of the extracted values from the two human collaborators corresponds to comparing the verified results. Malaney’s identified discrepancies correspond to the plurality of differences. Malaney further reinforces this comparison by teaching a collation process that includes finding disagreements based on affinities)
filter […] the plurality of differences in the verified results (Malaney, Par. [0190], “For documents having peculiar properties, such as a specific variation of a class of documents, a filter may optionally be applied. An example is if two documents are very close in format and data, but they differ in a very specific property and because of that they belong to different class. A weighted filter, that is a Location Diagram with primary key set for the distinguishing property or feature, is applied so that those can be classified accurately and rapidly”, & Par. [0232], “The basic method used for classification is different from current methods. Also the manner in which and sequence the instant invention uses various complementary technologies, such as filtering and voting, makes the method of the instant invention more accurate”, thus filter […] the plurality of differences in the verified results is disclosed because Malaney teaches applying filtering techniques to distinguish data or documents based on identified differences or distinguishing properties. Malaney’s differences in a specific property correspond to the plurality of differences in the verified results. Malaney’s weighted filter corresponds to filtering those identified differences so that the differing information can be separated and accurately classified)
a quality check module operatively coupled to the comparison module (Malaney, Par. [0203], “Extracted values are scrubbed to get exact value 227. Scrubbing further transforms the extracted value to a specific data type. The accuracy of the scrubbed value is verified. Thus, the system provides multiple confidence levels for decision-making”, thus a quality check module operatively coupled to the comparison module is disclosed because Malaney teaches verifying the accuracy of processed extracted values. Malaney’s accuracy verification corresponds to the quality check, and the multiple confidence levels correspond to evaluating the processed results) wherein the quality check module is configured to:
[…] wherein […the second transaction…] is based on the marked differences of the verified results (Malaney, Par. [0204], “In preferred embodiments, each field subjected to MEP is extracted by a minimum of two human collaborators and the system compares the extracted value. In the event of a discrepancy, the value in question can be sent to additional human collaborators”, & Par. [0205], “Information related to exceptions may also be used for statistical learning. In preferred embodiments, the human collaborator mouses over the exact value to be extracted. This is referred to herein as a "snippet" or a "text snippet" and the method/apparatus can pull the snippet and subject it to further scrubbing and processing 227. These snippets of required/specific values may also be extracted and used for formation of Knowledge Objects”, thus […] wherein […the second transaction…] is based on the marked differences of the verified results is disclosed because Malaney teaches identifying discrepancies in verified values and using exception related information for further processing and statistical learning. Malaney’s discrepancies and exceptions correspond to the marked differences)
assign […the second transaction…] with the marked differences to a third user for a subsequent review […] (Malaney, Par. [0204], “In preferred embodiments, each field subjected to MEP is extracted by a minimum of two human collaborators and the system compares the extracted value. In the event of a discrepancy, the value in question can be sent to additional human collaborators”, & Par. [0188], “Document pages that fall below the confidence threshold value that may be preset or varied by the user, even after the verifier, are sent to exception handling client (Classification client) (i.e., via escalation) 218. There, human collaborators can verify the class, assign a class, or note that the document cannot be identified”, thus assign […the second transaction…] with the marked differences to a third user for a subsequent review […] is disclosed because Malaney teaches sending discrepant values identified after comparison to additional human collaborators and escalating data for further human verification after an earlier verifier step. Malaney’s additional human collaborator corresponds to the third user, and the escalated discrepancy review corresponds to the subsequent review)
update […the plurality of datapoints…] in response to the review made by the third user thereby enabling quality check of the one or more documents over multiple levels of verification (Malaney, Par. [0204], “Human collaborators can verify and/or change the data and/or extracted information in reference to the Dox Package. In preferred embodiments, each field subjected to MEP is extracted by a minimum of two human collaborators and the system compares the extracted value. In the event of a discrepancy, the value in question can be sent to additional human collaborators”, & Par. [0188], “Document pages that fall below the confidence threshold value that may be preset or varied by the user, even after the verifier, are sent to exception handling client (Classification client) (i.e., via escalation) 218. There, human collaborators can verify the class, assign a class, or note that the document cannot be identified”, & Par. [0189], “Frequently the identification of the page or sheet may be made in reference to the header and footer information, although display parameters, such as position of the various images on the human collaborator's computer screen and zoom capabilities, may be varied by the human collaborator. Escalation may occur before, but preferably occurs after, the verifier step”, & Par. [0203], “Extracted values are scrubbed to get exact value 227. Scrubbing further transforms the extracted value to a specific data type. The accuracy of the scrubbed value is verified. Thus, the system provides multiple confidence levels for decision-making”, thus update […the plurality of datapoints…] in response to the review made by the third user thereby enabling quality check of the one or more documents over multiple levels of verification is disclosed because Malaney teaches that human collaborators may verify or change extracted data, including after discrepancies are escalated for additional review. Malaney’s ability to change the extracted data corresponds to updating the plurality of datapoints. Malaney’s escalation after the verifier step corresponds to multiple levels of verification, and the subsequent accuracy verification and multiple confidence levels correspond to quality checking the documents across those verification levels)
Malaney does not explicitly teach a hardware processor, a memory coupled to the hardware processor […], […] by using machine learning model to generate a first transaction, generate a plurality of sub-transactions from the first transaction wherein each of the plurality of sub-transactions comprises […the plurality of data points…], […] and mark […the plurality of differences in the verified results ...], generate a second transaction by using the machine learning model […], and […] thereby resulting in a second verification level.
However, Bobley teaches:
a hardware processor (Bobley, Par. [0134], “The computer 2700 also includes a central processing unit (CPU) 2720, in the form of one or more processors, for executing program instructions”, thus a hardware processor is disclosed)
a memory coupled to the hardware processor (Bobley, Par. [0134], “The computer 2700 also includes a central processing unit (CPU) 2720, in the form of one or more processors, for executing program instructions. The exemplary computer platform includes an internal communication bus 2710, program storage and data storage of different forms, e.g., disk 2770, read only memory (ROM) 2730, or random access memory (RAM) 2740, for various data files to be processed and/or communicated by the computer, as well as possibly program instructions to be executed by the CPU”, thus a memory coupled to the hardware processor is disclosed) wherein the memory comprises a set of instructions in the form of a processing subsystem, configured to be executed by the hardware processor, wherein the processing subsystem is hosted on a server, and configured to execute on a network to control bidirectional communications among a plurality of modules wherein the plurality of modules comprises:
[…] by using machine learning model to generate a first transaction (Bobley, Par. [0129], “Pydatalog is used to perform a machine learning algorithm that improves transaction detection component accuracy over a large training data set. The training data may come from a relational data store that contains both raw OCR output and post-process financial transaction information with a high degree of accuracy”, & Par. [0106], “The present system utilizes the bounding box to identify the geographic location of the information on the scanned document and generates a spatial index linking the X and Y location of the information on the document to the transaction that has been extracted. The information within a bounding box is then treated as a single transaction”, thus […] by using machine learning model to generate a first transaction is disclosed because Bobley teaches using a machine learning algorithm to process document-derived data and improve transaction detection, with extracted information associated with an identified transaction. Bobley’s machine learning algorithm corresponds to using a machine learning model. Bobley’s transaction detection and extraction process corresponds to the initial parsing and processing of document data. Bobley’s treatment of extracted information within a bounding box as a single transaction corresponds to generating a first transaction, because the first transaction represents the initial processing instance in which document data is parsed and organized into transaction information)
and generate a plurality of sub-transactions from the first transaction wherein each of the plurality of sub-transactions comprises […the plurality of data points…] (Bobley, Par. [0150], “In step 3412, after all of the transactions have been split into ‘snippets’, the transactions, i.e, the image within the bounding boxes from transactions are displayed to the user. Below each image are fields for the corresponding date and description. The fields are populated with the respective OCR outputs, and the system and/or the user must rectify any errors in the fields”, & Par. [0149], “In one or more embodiments, the transaction date, description and amount of the highlighted transaction are viewable for the purpose of identification, but are not editable in step 3410. Accordingly, when a missing transaction is added, the fields for date, description and amount will be blank”, thus and generate a plurality of sub-transactions from the first transaction wherein each of the plurality of sub-transactions comprises […the plurality of data points…] is disclosed because Bobley teaches splitting transactions into individual snippets for separate processing and review, where each snippet is associated with transaction information such as date, description, and amount. Bobley’s splitting of transactions into snippets corresponds to generating a plurality of sub-transactions from the first transaction. Bobley’s individual snippets correspond to the sub-transactions because they represent separated portions of the transaction that are independently displayed and processed. Bobley’s date, description, and amount fields associated with each transaction snippet correspond to the plurality of data points contained in each sub-transaction)
[…] and mark […the plurality of differences in the verified results ...] (Bobley, Par. [0016], “In one or more embodiments, the annotating includes at least one of highlighting unique transactions verified as accurate, highlighting unique transactions identified as containing errors, or highlighting unique transactions verified as accurate and unique transactions identified as containing errors. In other embodiments, the unique transactions verified as accurate are highlighted in a first color and the unique transactions identified as containing errors are highlighted in a second color”, thus […] and mark […the plurality of differences in the verified results ...] is disclosed because Bobley teaches annotating transaction data by highlighting transactions according to whether they have been verified as accurate or identified as containing errors. Bobley’s transactions identified as containing errors correspond to the differences in the verified results. Bobley’s highlighting of those transactions corresponds to marking the identified differences, and the use of different highlighting for accurate and erroneous transactions further distinguishes the marked differences from verified data)
generate a second transaction by using the machine learning model […] (Bobley, Par. [0129], “Pydatalog is used to perform a machine learning algorithm that improves transaction detection component accuracy over a large training data set”, & Par. [0130], “A transaction (date, description, amount, account, other_acct) def_init_(self, date, description, amount, account, other_acct): super(Transaction, self)._init_( ) # calls the initialization method of the Mixin class self. date=date self. amount=amount self. description=description @pyDatalog.program( ) # indicates that the following method contains pyDatalog clauses def Transfer(from, to, amount”, & “ # An example how a transfer is detected. acct1=‘001’ acct2=‘002’ acct3=‘003’ trans1=Transaction (‘20150108’, ‘transfer from checking acct into saving acct’, 6800, acct1, acct2) trans2=Transaction (‘20150108’, ‘transfer from checking acct into saving acct’, −6800, acct2, acct1)”, thus generate a second transaction by using the machine learning model […] is disclosed because Bobley teaches using a machine learning algorithm for transaction detection and subsequently representing detected transaction information as multiple transaction objects, including trans1 and trans2. Bobley’s machine learning algorithm corresponds to the machine learning model, and Bobley’s trans2 corresponds to a second transaction produced as part of the transaction-detection processing)
[…] thereby resulting in a second verification level (Bobley, Par. [0143], “In Moderation Flow, the system prompts the user to verify transactional and account data from snippets until reconciliation is achieved. After reconciliation, in step 3214, the edited transaction and account data is compiled into Final SME Review, Final SME Review includes a page-by-page human review of all data from each bank document”, thus […] thereby resulting in a second verification level is disclosed because Bobley teaches an initial Moderation Flow verification followed by a Final SME Review. Bobley’s Moderation Flow corresponds to the earlier verification stage, and the subsequent Final SME Review corresponds to the second verification level)
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Malaney’s system to incorporate Bobley’s machine learning based preprocessing and transaction processing techniques so that extracted data is processed using a machine learning model to generate transactions, the transactions are divided for verification, identified differences or errors are marked, and the resulting transaction data is subjected to subsequent review. Bobley teaches that its preprocessing techniques improve processing times, increase accuracy, and reduce errors, and that user review is performed for purposes of quality control. Therefore, A POSITA would have been motivated to use Bobley’s machine learning and transaction processing techniques in Malaney’s multi-user verification system to improve the accuracy and efficiency of generating, verifying, marking, and reviewing transaction data over multiple levels of verification (Bobley, Par. [0004], “The current system and method allows users to scan and upload bank statements, credit card statements, investment statements, and any other tabulated series of repeated entries into a web based application for processing and to identify errors in the entries, sort search and filter the entries. The system and methods utilizing preprocessing techniques improve processing times, increases accuracy and reduces errors”, & Par. [0141], “The annotations may also be used to denote or represent different, algorithmically-inferred accounts. In one or more embodiments, a user review is implemented for purposes of quality control, which includes review of each page and confirmation that all transactions are properly bounded and shaded”)
Regarding Claim 3, Malaney combined with Bobley teaches all the limitations of claim 1 as cited above and Bobley further teaches:
wherein the plurality of sub-transactions are merged to generate the second transaction upon verification of the said plurality of sub-transactions (Bobley, Par. [0143], “In step 3210, each relevant page is sliced into transaction images, for example, from within bounding boxes. In step 3212, a Moderation Flow process begins. In Moderation Flow, the system prompts the user to verify transactional and account data from snippets until reconciliation is achieved. After reconciliation, in step 3214, the edited transaction and account data is compiled into Final SME Review”, thus wherein the plurality of sub-transactions are merged to generate the second transaction upon verification of the said plurality of sub-transactions is disclosed because Bobley teaches verifying transactional data from multiple snippets until reconciliation is achieved and thereafter compiling the verified transaction and account data. Bobley’s snippets correspond to the plurality of sub-transactions, the reconciliation corresponds to verification of the sub-transactions, and compiling the edited transaction and account data after reconciliation corresponds to merging the verified sub-transactions to generate the second transaction)
Regarding Claim 4, Malaney combined with Bobley teaches all the limitations of claim 1 as cited above and Bobley further teaches:
wherein […the verification module…] is configured to enable […the first user…] and […the second user…] to tag an uncertain data at the time of verification to enable […the third user…] to validate the tagged uncertain data (Bobley, Claim 14, “wherein the annotating includes tagging or flagging at least one of the first transaction or the second transaction verified as accurate, the first transaction or the second transaction identified as containing an error, or the first transaction verified as accurate and the second transaction identified as containing an error”, & Par. [0160], “Red Exclamation Point: This means that the date and/or description and amount was marked as “unreadable” and did not go through the entire process of verification. Transactions with this marker need to have the dates/descriptions and amounts verified”, thus wherein […the verification module…] is configured to enable […the first user…] and […the second user…] to tag an uncertain data at the time of verification to enable […the third user…] to validate the tagged uncertain data is disclosed because Bobley teaches tagging or flagging transaction data according to whether the data is accurate or contains an error and marking unreadable data for further verification. Bobley’s tagging or flagging of transaction data corresponds to tagging uncertain data during verification, and Bobley’s unreadable marker identifying data that still needs verification corresponds to enabling the subsequent user to validate the tagged uncertain data)
Regarding Claim 5, Malaney combined with Bobley teaches all the limitations of claim 1 as cited above and Malaney further teaches:
wherein the third user is allowed to view the extracted, verified, or updated data at the first level of verification and […the second level of verification…] (Malaney, Par. [0204], “Human collaborators can verify and/or change the data and/or extracted information in reference to the Dox Package. In preferred embodiments, each field subjected to MEP is extracted by a minimum of two human collaborators and the system compares the extracted value. In the event of a discrepancy, the value in question can be sent to additional human collaborators”, & Par. [0188], “Document pages that fall below the confidence threshold value that may be preset or varied by the user, even after the verifier, are sent to exception handling client (Classification client) (i.e., via escalation) 218. There, human collaborators can verify the class, assign a class, or note that the document cannot be identified”, thus wherein the third user is allowed to view the extracted, verified, or updated data at the first level of verification and […the second level of verification…] is disclosed because Malaney teaches that extracted data may be verified or changed by human collaborators, compared after verification, and, when a discrepancy is identified, sent to additional human collaborators for further review. Malaney further teaches escalation after the verifier step, where the additional human collaborator is presented with the data for verification)
Regarding Claim 6, Malaney combined with Bobley teaches all the limitations of claim 1 as cited above and Bobley further teaches:
wherein […the data points…] comprises dates, numbers, long text, short text, currencies, and percentages (Bobley, Par. [0149], “the transaction date, description and amount of the highlighted transaction are viewable for the purpose of identification, but are not editable in step 3410. Accordingly, when a missing transaction is added, the fields for date, description and amount will be blank”, & Par. [0163], “the data that is visible for a single transaction is presented in the panel on the right hand side. It includes Transaction Type, Account, Statement Period, Transaction Date, Description and Amount”, & Par. [0089], “Most OCR algorithms produce the read as tree-like structure, for example, starting at a root of the tree, the root node represent the page, which can have multiple child nodes to represent paragraphs, which again may have multiple child nodes to represent a line of text, then a word, and finally, a character”, & Par. [0071], “search menu 1600 includes parameters for a date range 1601 and a closeness in dollar value 1602 (0%-25%). For example, if the anchor transaction 1501's value is $2,000 and closeness in dollar value slider 1602 is set to 0%, the system will only generate exact $2,000 matches. If the slider 1602 is set to 25%, the system will show potential matches ranging from $1,500 to $2,500”, thus wherein […the data points…] comprises dates, numbers, long text, short text, currencies, and percentages is disclosed because Bobley teaches transaction data including transaction dates, descriptions, amounts, account information, and percentage-based search parameters, and further teaches processing textual data at paragraph, line, word, and character levels. Bobley’s transaction dates correspond to dates, account and amount values correspond to numbers, paragraphs and lines correspond to long text, words and characters correspond to short text, dollar amounts correspond to currencies, and percentage based closeness values correspond to percentages)
Regarding Claim 7, Malaney combined with Bobley teaches all the limitations of claim 1 as cited above and Malaney further teaches:
wherein the data points are customized by using a plurality of keywords and forms based on the search requirements of the user (Malaney, Par. [0202], “Relevant information, as defined by a pre-determined business-specific application or set by a user, is extracted from documents that have been successfully classified”, & Par. [0212], “The Knowledge Objects extracted from various forms like 1003 429, 1004 434, and Note 433 are combined to form a transaction set for underwriting of a loan using a rules engine”, & Par. [0215], “The document samples for the document to be learned, and document-specific dictionaries and generic, as well as domain, dictionaries are loaded in to the Learner's Knowledge Base. 301. The Learner reads the document samples, and if document specific dictionaries are not available, then one is generated from the sample documents. 302. For some specific files, human input such as very specific key phrases and location are provided for learning, if required”, thus wherein the data points are customized by using a plurality of keywords and forms based on the search requirements of the user is disclosed because Malaney teaches extracting user-defined relevant information from various forms and using document-specific dictionaries and specific key phrases for processing particular documents. Malaney’s specific key phrases correspond to the plurality of keywords, its various forms correspond to the plurality of forms, and the relevant information set by a user corresponds to customizing the data points based on the user’s requirements)
Regarding Claim 8, Malaney combined with Bobley teaches all the limitations of claim 1 as cited above and Malaney further teaches:
wherein if the verification module finds no differences in […the plurality of sub-transactions…], an output of the first verification is considered as a final output (Malaney, Par. [0204], “each field subjected to MEP is extracted by a minimum of two human collaborators and the system compares the extracted value. In the event of a discrepancy, the value in question can be sent to additional human collaborators”, & Par. [0203], “The accuracy of the scrubbed value is verified. Thus, the system provides multiple confidence levels for decision-making. The system generates a Knowledge Object from the scrubbed results. The values with very high extraction confidence but very low scrubbing confidence are sent to a human assisted Field Location Process ("Manual FLP") 225. The system generates a field value from scrubbed results that pass the confidence threshold for the overall process”, thus wherein if the verification module finds no differences in […the plurality of sub-transactions…], an output of the first verification is considered as a final output is disclosed because Malaney teaches comparing values obtained from multiple human collaborators and sending a value for additional review when a discrepancy is identified. Malaney further teaches generating a field value from verified results that pass the confidence threshold. When no discrepancy is identified and the verified results satisfy the confidence threshold, the verified field value proceeds as the output of the verification process)
Regarding Claim 9, Malaney combined with Bobley teaches all the limitations of claim 1 as cited above and Bobley further teaches:
a database to store the verified results along with the corresponding differences (Bobley, Par. [0165], “In Final SME Review, modifications are logged, the results are written to the data store, and the data is updated in the database”, & Par. [0167], “After reconciliation has been achieved, or when the automated system determines or the user conducting the Final SME Review believes that the data is sufficiently accurate, the results will be sent to the client. The bank statement data that was validated in the BO tool will be stored in a database”, thus a database to store the verified results along with the corresponding differences is disclosed because Bobley teaches logging modifications made during Final SME Review, writing the resulting data to a data store, updating the data in a database, and storing validated bank statement data in the database. Bobley’s validated bank statement data corresponds to the verified results, and the logged modifications correspond to the corresponding differences associated with those verified results)
Regarding Claim 10, Malaney combined with Bobley teaches all the limitations of claim 1 as cited above and Malaney further teaches:
wherein the one or more documents includes unstructured data (Malaney, Par. [0092], “As used herein, the term "information fields" refers to the content of the blanks on the forms, e.g., in the context of the mortgage field, the price of the property, the amount financed, the address, etc. or specific content from an unstructured document such as stated interest rate in a promissory note”, thus wherein the one or more documents includes unstructured data is disclosed because Malaney teaches extracting specific content from an unstructured document. Malaney’s specific content contained in the unstructured document corresponds to unstructured data included in the one or more documents)
Regarding Claim 11, Malaney combined with Bobley teaches all the limitations of claim 1 as cited above and Bobley further teaches:
wherein […the plurality of differences…] are marked by highlighting the differences in colour coded format (Bobley, Par. [0016], “the annotating includes at least one of highlighting unique transactions verified as accurate, highlighting unique transactions identified as containing errors, or highlighting unique transactions verified as accurate and unique transactions identified as containing errors. In other embodiments, the unique transactions verified as accurate are highlighted in a first color and the unique transactions identified as containing errors are highlighted in a second color”, thus wherein […the plurality of differences…] are marked by highlighting the differences in colour coded format is disclosed because Bobley teaches identifying transaction data containing errors and highlighting that data in a color different from transaction data verified as accurate. Bobley’s transactions identified as containing errors correspond to the plurality of differences, and highlighting those transactions in a second color corresponds to marking the differences in a colour coded format)
Regarding Claim 12, Malaney teaches:
parsing, by a data extraction module of a processing subsystem, one or more documents to extract a plurality of data points […] (Malaney, Par. [0201], “In preferred embodiments of the invention, extraction of information or data from the documents or Dox Package that has been captured using the method or apparatus of the invention, and preferably extraction is first done automatically from readily identifiable fields 221 and image snippets of other fields location are re-submitted to the OCR step with a field specific dictionary before repeating the extraction process 220”, & Par. [0206], “The processes preferred embodiments of the instant invention typically extracts the fields (as they appear in the document) required for various business and/or compliance requirements, then transforms them into facts that can be used further for decision making by an automated rules engine or search engine”, & Par. [0131], “The present invention, in preferred embodiments, utilizes Location Diagram concepts and integrates multiple components including image processing, intelligent collation, feedback learning, a document classifier, a verifier, a versioning engine, an information locator, a data extractor, a data scrubber, and manual collaboration”, thus parsing, by a data extraction module of a processing subsystem, one or more documents to extract a plurality of data points […] is disclosed because Malaney teaches automatically processing captured documents to extract information from multiple identifiable fields for further processing. Malaney’s documents or Dox Package correspond to one or more documents. Malaney’s automatic extraction of information or data from document fields corresponds to parse one or more documents to extract. Malaney’s multiple extracted fields correspond to a plurality of data points because the system extracts multiple pieces of information from the documents and transforms them into facts for subsequent use)
verifying, by a verification module of the processing subsystem, each of […the plurality of sub-transactions…] individually by a first user and a second user respectively thereby resulting in a first level of verification (Malaney, Par. [0131], “The present invention, in preferred embodiments, utilizes Location Diagram concepts and integrates multiple components including image processing, intelligent collation, feedback learning, a document classifier, a verifier, a versioning engine, an information locator, a data extractor, a data scrubber, and manual collaboration”, & Par. [0204], “Human collaborators can verify and/or change the data and/or extracted information in reference to the Dox Package. In preferred embodiments, each field subjected to MEP is extracted by a minimum of two human collaborators and the system compares the extracted value. In the event of a discrepancy, the value in question can be sent to additional human collaborators”, thus verifying, by a verification module of the processing subsystem, each of […the plurality of sub-transactions…] individually by a first user and a second user respectively thereby resulting in a first level of verification is disclosed because Malaney teaches an integrated system including a data extractor, verifier, and manual collaboration, and further teaches that each individual field is handled by at least two human collaborators who verify or change the extracted data. Malaney’s integrated data extractor and verifier correspond to the verification module. Malaney’s individual fields subjected to manual extraction correspond to the plurality of sub-transactions. The minimum of two human collaborators correspond to the first user and second user, respectively. Their independent verification of each field corresponds to verifying each sub-transaction individually, and this two-user verification before any discrepancy is escalated to additional collaborators corresponds to the first level of verification)
comparing, by a comparison module of the processing subsystem, the verified results of […the plurality of sub-transactions…] for identifying a plurality of differences (Malaney, Par. [0204], “In preferred embodiments, each field subjected to MEP is extracted by a minimum of two human collaborators and the system compares the extracted value. In the event of a discrepancy, the value in question can be sent to additional human collaborators”, & Par. [0131], “The present invention, in preferred embodiments, utilizes Location Diagram concepts and integrates multiple components including image processing, intelligent collation, feedback learning, a document classifier, a verifier, a versioning engine, an information locator, a data extractor, a data scrubber, and manual collaboration”, & Par. [0204], “In preferred embodiments, each field subjected to MEP is extracted by a minimum of two human collaborators and the system compares the extracted value”, & Par. [0192 – 0197], “In preferred embodiments of the instant invention, the collation process is based on incremental learning and various artificial intelligence ("AI")-based techniques, which may include one or more of the following, such as: (1) the Location Diagram- and Feature Vector-based feature extraction and page mapping; (2) SVM and NLP; (3) an intelligent filter technique taking advantage of header and footer based information; (4) collation by finding common threads within or between pages, documents, or sets of documents; (5) finding disagreements based on affinities”, thus comparing, by a comparison module of the processing subsystem, the verified results of […the plurality of sub-transactions…] for identifying a plurality of differences is disclosed because Malaney teaches comparing values obtained from at least two human collaborators and identifying discrepancies between those values. Malaney’s individual fields subjected to MEP correspond to the plurality of sub-transactions. Malaney’s comparison of the extracted values from the two human collaborators corresponds to comparing the verified results. Malaney’s identified discrepancies correspond to the plurality of differences. Malaney further reinforces this comparison by teaching a collation process that includes finding disagreements based on affinities)
filtering […], by the comparison module of the processing subsystem the plurality of differences in the verified results (Malaney, Par. [0190], “For documents having peculiar properties, such as a specific variation of a class of documents, a filter may optionally be applied. An example is if two documents are very close in format and data, but they differ in a very specific property and because of that they belong to different class. A weighted filter, that is a Location Diagram with primary key set for the distinguishing property or feature, is applied so that those can be classified accurately and rapidly”, & Par. [0232], “The basic method used for classification is different from current methods. Also the manner in which and sequence the instant invention uses various complementary technologies, such as filtering and voting, makes the method of the instant invention more accurate”, thus filtering […], by the comparison module of the processing subsystem the plurality of differences in the verified results is disclosed because Malaney teaches applying filtering techniques to distinguish data or documents based on identified differences or distinguishing properties. Malaney’s differences in a specific property correspond to the plurality of differences in the verified results. Malaney’s weighted filter corresponds to filtering those identified differences so that the differing information can be separated and accurately classified)
[…] , by a quality check module of the processing subsystem, […] wherein […the second transaction…] is based on the marked differences of the verified results (Malaney, Par. [0203], “Extracted values are scrubbed to get exact value 227. Scrubbing further transforms the extracted value to a specific data type. The accuracy of the scrubbed value is verified. Thus, the system provides multiple confidence levels for decision-making”, & Par. [0204], “In preferred embodiments, each field subjected to MEP is extracted by a minimum of two human collaborators and the system compares the extracted value. In the event of a discrepancy, the value in question can be sent to additional human collaborators”, & Par. [0205], “Information related to exceptions may also be used for statistical learning. In preferred embodiments, the human collaborator mouses over the exact value to be extracted. This is referred to herein as a "snippet" or a "text snippet" and the method/apparatus can pull the snippet and subject it to further scrubbing and processing 227. These snippets of required/specific values may also be extracted and used for formation of Knowledge Objects”, thus […] , by a quality check module of the processing subsystem, […] wherein […the second transaction…] is based on the marked differences of the verified results is disclosed because Malaney teaches identifying discrepancies in verified values and using exception related information for further processing and statistical learning. Malaney’s discrepancies and exceptions correspond to the marked differences)
assigning, by the quality check module of the processing subsystem, […the second transaction…] with the marked differences to a third user for a subsequent review […] (Malaney, Par. [0204], “In preferred embodiments, each field subjected to MEP is extracted by a minimum of two human collaborators and the system compares the extracted value. In the event of a discrepancy, the value in question can be sent to additional human collaborators”, & Par. [0188], “Document pages that fall below the confidence threshold value that may be preset or varied by the user, even after the verifier, are sent to exception handling client (Classification client) (i.e., via escalation) 218. There, human collaborators can verify the class, assign a class, or note that the document cannot be identified”, thus assigning, by the quality check module of the processing subsystem, […the second transaction…] with the marked differences to a third user for a subsequent review […]is disclosed because Malaney teaches sending discrepant values identified after comparison to additional human collaborators and escalating data for further human verification after an earlier verifier step. Malaney’s additional human collaborator corresponds to the third user, and the escalated discrepancy review corresponds to the subsequent review)
updating, by the quality check module of the processing subsystem, the plurality of datapoint in response to the review made by a third user thereby enabling quality check of the one or more documents over multiple levels of verification (Malaney, Par. [0204], “Human collaborators can verify and/or change the data and/or extracted information in reference to the Dox Package. In preferred embodiments, each field subjected to MEP is extracted by a minimum of two human collaborators and the system compares the extracted value. In the event of a discrepancy, the value in question can be sent to additional human collaborators”, & Par. [0188], “Document pages that fall below the confidence threshold value that may be preset or varied by the user, even after the verifier, are sent to exception handling client (Classification client) (i.e., via escalation) 218. There, human collaborators can verify the class, assign a class, or note that the document cannot be identified”, & Par. [0189], “Frequently the identification of the page or sheet may be made in reference to the header and footer information, although display parameters, such as position of the various images on the human collaborator's computer screen and zoom capabilities, may be varied by the human collaborator. Escalation may occur before, but preferably occurs after, the verifier step”, & Par. [0203], “Extracted values are scrubbed to get exact value 227. Scrubbing further transforms the extracted value to a specific data type. The accuracy of the scrubbed value is verified. Thus, the system provides multiple confidence levels for decision-making”, thus updating, by the quality check module of the processing subsystem, the plurality of datapoints in response to the review made by a third user thereby enabling quality check of the one or more documents over multiple levels of verification is disclosed because Malaney teaches that human collaborators may verify or change extracted data, including after discrepancies are escalated for additional review. Malaney’s ability to change the extracted data corresponds to updating the plurality of datapoints. Malaney’s escalation after the verifier step corresponds to multiple levels of verification, and the subsequent accuracy verification and multiple confidence levels correspond to quality checking the documents across those verification levels)
Malaney does not explicitly teach […] by using machine learning model to generate a first transaction, generating […] a plurality of sub-transactions from the first transaction wherein each of the plurality of sub-transactions comprises […the plurality of data points…], […] and marking […the plurality of differences in the verified results ...], generating […] a second transaction by using the machine learning model […], and […] thereby resulting in a second verification level.
However, Bobley teaches:
[…] by using machine learning model to generate a first transaction (Bobley, Par. [0129], “Pydatalog is used to perform a machine learning algorithm that improves transaction detection component accuracy over a large training data set. The training data may come from a relational data store that contains both raw OCR output and post-process financial transaction information with a high degree of accuracy”, & Par. [0106], “The present system utilizes the bounding box to identify the geographic location of the information on the scanned document and generates a spatial index linking the X and Y location of the information on the document to the transaction that has been extracted. The information within a bounding box is then treated as a single transaction”, thus […] by using machine learning model to generate a first transaction is disclosed because Bobley teaches using a machine learning algorithm to process document-derived data and improve transaction detection, with extracted information associated with an identified transaction. Bobley’s machine learning algorithm corresponds to using a machine learning model. Bobley’s transaction detection and extraction process corresponds to the initial parsing and processing of document data. Bobley’s treatment of extracted information within a bounding box as a single transaction corresponds to generating a first transaction, because the first transaction represents the initial processing instance in which document data is parsed and organized into transaction information)
generating […] a plurality of sub-transactions from the first transaction wherein each of the plurality of sub-transactions comprises […the plurality of data points…] (Bobley, Par. [0150], “In step 3412, after all of the transactions have been split into ‘snippets’, the transactions, i.e, the image within the bounding boxes from transactions are displayed to the user. Below each image are fields for the corresponding date and description. The fields are populated with the respective OCR outputs, and the system and/or the user must rectify any errors in the fields”, & Par. [0149], “In one or more embodiments, the transaction date, description and amount of the highlighted transaction are viewable for the purpose of identification, but are not editable in step 3410. Accordingly, when a missing transaction is added, the fields for date, description and amount will be blank”, thus generating […] a plurality of sub-transactions from the first transaction wherein each of the plurality of sub-transactions comprises […the plurality of data points…] is disclosed because Bobley teaches splitting transactions into individual snippets for separate processing and review, where each snippet is associated with transaction information such as date, description, and amount. Bobley’s splitting of transactions into snippets corresponds to generating a plurality of sub-transactions from the first transaction. Bobley’s individual snippets correspond to the sub-transactions because they represent separated portions of the transaction that are independently displayed and processed. Bobley’s date, description, and amount fields associated with each transaction snippet correspond to the plurality of data points contained in each sub-transaction)
[…] and marking […the plurality of differences in the verified results ...] (Bobley, Par. [0016], “In one or more embodiments, the annotating includes at least one of highlighting unique transactions verified as accurate, highlighting unique transactions identified as containing errors, or highlighting unique transactions verified as accurate and unique transactions identified as containing errors. In other embodiments, the unique transactions verified as accurate are highlighted in a first color and the unique transactions identified as containing errors are highlighted in a second color”, thus […] and marking […the plurality of differences in the verified results ...] is disclosed because Bobley teaches annotating transaction data by highlighting transactions according to whether they have been verified as accurate or identified as containing errors. Bobley’s transactions identified as containing errors correspond to the differences in the verified results. Bobley’s highlighting of those transactions corresponds to marking the identified differences, and the use of different highlighting for accurate and erroneous transactions further distinguishes the marked differences from verified data)
generating […] a second transaction by using the machine learning model […] (Bobley, Par. [0129], “Pydatalog is used to perform a machine learning algorithm that improves transaction detection component accuracy over a large training data set”, & Par. [0130], “A transaction (date, description, amount, account, other_acct) def_init_(self, date, description, amount, account, other_acct): super(Transaction, self)._init_( ) # calls the initialization method of the Mixin class self. date=date self. amount=amount self. description=description @pyDatalog.program( ) # indicates that the following method contains pyDatalog clauses def Transfer(from, to, amount”, & “ # An example how a transfer is detected. acct1=‘001’ acct2=‘002’ acct3=‘003’ trans1=Transaction (‘20150108’, ‘transfer from checking acct into saving acct’, 6800, acct1, acct2) trans2=Transaction (‘20150108’, ‘transfer from checking acct into saving acct’, −6800, acct2, acct1)”, thus generating […] a second transaction by using the machine learning model […] is disclosed because Bobley teaches using a machine learning algorithm for transaction detection and subsequently representing detected transaction information as multiple transaction objects, including trans1 and trans2. Bobley’s machine learning algorithm corresponds to the machine learning model, and Bobley’s trans2 corresponds to a second transaction produced as part of the transaction-detection processing)
[…] thereby resulting in a second verification level (Bobley, Par. [0143], “In Moderation Flow, the system prompts the user to verify transactional and account data from snippets until reconciliation is achieved. After reconciliation, in step 3214, the edited transaction and account data is compiled into Final SME Review, Final SME Review includes a page-by-page human review of all data from each bank document”, thus […] thereby resulting in a second verification level is disclosed because Bobley teaches an initial Moderation Flow verification followed by a Final SME Review. Bobley’s Moderation Flow corresponds to the earlier verification stage, and the subsequent Final SME Review corresponds to the second verification level)
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Malaney’s system to incorporate Bobley’s machine learning based preprocessing and transaction processing techniques so that extracted data is processed using a machine learning model to generate transactions, the transactions are divided for verification, identified differences or errors are marked, and the resulting transaction data is subjected to subsequent review. Bobley teaches that its preprocessing techniques improve processing times, increase accuracy, and reduce errors, and that user review is performed for purposes of quality control. Therefore, A POSITA would have been motivated to use Bobley’s machine learning and transaction processing techniques in Malaney’s multi-user verification system to improve the accuracy and efficiency of generating, verifying, marking, and reviewing transaction data over multiple levels of verification (Bobley, Par. [0004], “The current system and method allows users to scan and upload bank statements, credit card statements, investment statements, and any other tabulated series of repeated entries into a web based application for processing and to identify errors in the entries, sort search and filter the entries. The system and methods utilizing preprocessing techniques improve processing times, increases accuracy and reduces errors”, & Par. [0141], “The annotations may also be used to denote or represent different, algorithmically-inferred accounts. In one or more embodiments, a user review is implemented for purposes of quality control, which includes review of each page and confirmation that all transactions are properly bounded and shaded”)
Regarding Claim 13, Malaney combined with Bobley teaches all the limitations of claim 12 as cited above and Bobley further teaches:
tagging an uncertain data by […the first user…] and […the second user…] at the time of verification to enable […the third user…] to validate the tagged uncertain data (Bobley, Claim 14, “wherein the annotating includes tagging or flagging at least one of the first transaction or the second transaction verified as accurate, the first transaction or the second transaction identified as containing an error, or the first transaction verified as accurate and the second transaction identified as containing an error”, & Par. [0160], “Red Exclamation Point: This means that the date and/or description and amount was marked as “unreadable” and did not go through the entire process of verification. Transactions with this marker need to have the dates/descriptions and amounts verified”, thus tagging an uncertain data by […the first user…] and […the second user…] at the time of verification to enable […the third user…] to validate the tagged uncertain data is disclosed because Bobley teaches tagging or flagging transaction data according to whether the data is accurate or contains an error and marking unreadable data for further verification. Bobley’s tagging or flagging of transaction data corresponds to tagging uncertain data during verification, and Bobley’s unreadable marker identifying data that still needs verification corresponds to enabling the subsequent user to validate the tagged uncertain data)
Regarding Claim 14, Malaney combined with Bobley teaches all the limitations of claim 12 as cited above and Malaney further teaches:
viewing, the extracted datapoints and values of a plurality of attributes of the extracted datapoints, wherein the values are verified or updated at a previous verification level (Malaney, Par. [0204], “Human collaborators can verify and/or change the data and/or extracted information in reference to the Dox Package. In preferred embodiments, each field subjected to MEP is extracted by a minimum of two human collaborators and the system compares the extracted value. In the event of a discrepancy, the value in question can be sent to additional human collaborators”, & Par. [0188], “Document pages that fall below the confidence threshold value that may be preset or varied by the user, even after the verifier, are sent to exception handling client (Classification client) (i.e., via escalation) 218. There, human collaborators can verify the class, assign a class, or note that the document cannot be identified”, thus viewing, the extracted datapoints and values of a plurality of attributes of the extracted datapoints, wherein the values are verified or updated at a previous verification level is disclosed because Malaney teaches that extracted data may be verified or changed by human collaborators, compared after verification, and, when a discrepancy is identified, sent to additional human collaborators for further review. Malaney further teaches escalation after the verifier step, where the additional human collaborator is presented with the data for verification)
Regarding Claim 15, Malaney teaches:
parsing, by a data extraction module of a processing subsystem, one or more documents to extract a plurality of data points […] wherein the one or more documents includes unstructured data (Malaney, Par. [0201], “In preferred embodiments of the invention, extraction of information or data from the documents or Dox Package that has been captured using the method or apparatus of the invention, and preferably extraction is first done automatically from readily identifiable fields 221 and image snippets of other fields location are re-submitted to the OCR step with a field specific dictionary before repeating the extraction process 220”, & Par. [0206], “The processes preferred embodiments of the instant invention typically extracts the fields (as they appear in the document) required for various business and/or compliance requirements, then transforms them into facts that can be used further for decision making by an automated rules engine or search engine”, & Par. [0131], “The present invention, in preferred embodiments, utilizes Location Diagram concepts and integrates multiple components including image processing, intelligent collation, feedback learning, a document classifier, a verifier, a versioning engine, an information locator, a data extractor, a data scrubber, and manual collaboration”, & Par. [0092], “As used herein, the term "information fields" refers to the content of the blanks on the forms, e.g., in the context of the mortgage field, the price of the property, the amount financed, the address, etc. or specific content from an unstructured document such as stated interest rate in a promissory note”, thus parsing, by a data extraction module of a processing subsystem, one or more documents to extract a plurality of data points […] wherein the one or more documents includes unstructured data is disclosed because Malaney teaches automatically processing captured documents to extract information from multiple identifiable fields for further processing. Malaney’s documents or Dox Package correspond to one or more documents. Malaney’s automatic extraction of information or data from document fields corresponds to parse one or more documents to extract. Malaney’s multiple extracted fields correspond to a plurality of data points because the system extracts multiple pieces of information from the documents and transforms them into facts for subsequent use)
verifying, by a verification module of the processing subsystem, each of […the plurality of sub-transactions…] individually by a first user and a second user respectively thereby resulting in a first level of verification (Malaney, Par. [0131], “The present invention, in preferred embodiments, utilizes Location Diagram concepts and integrates multiple components including image processing, intelligent collation, feedback learning, a document classifier, a verifier, a versioning engine, an information locator, a data extractor, a data scrubber, and manual collaboration”, & Par. [0204], “Human collaborators can verify and/or change the data and/or extracted information in reference to the Dox Package. In preferred embodiments, each field subjected to MEP is extracted by a minimum of two human collaborators and the system compares the extracted value. In the event of a discrepancy, the value in question can be sent to additional human collaborators”, thus verifying, by a verification module of the processing subsystem, each of […the plurality of sub-transactions…] individually by a first user and a second user respectively thereby resulting in a first level of verification is disclosed because Malaney teaches an integrated system including a data extractor, verifier, and manual collaboration, and further teaches that each individual field is handled by at least two human collaborators who verify or change the extracted data. Malaney’s integrated data extractor and verifier correspond to the verification module. Malaney’s individual fields subjected to manual extraction correspond to the plurality of sub-transactions. The minimum of two human collaborators correspond to the first user and second user, respectively. Their independent verification of each field corresponds to verifying each sub-transaction individually, and this two-user verification before any discrepancy is escalated to additional collaborators corresponds to the first level of verification)
comparing, by a comparison module of the processing subsystem, the verified results of […the plurality of sub-transactions…] for identifying a plurality of differences (Malaney, Par. [0204], “In preferred embodiments, each field subjected to MEP is extracted by a minimum of two human collaborators and the system compares the extracted value. In the event of a discrepancy, the value in question can be sent to additional human collaborators”, & Par. [0131], “The present invention, in preferred embodiments, utilizes Location Diagram concepts and integrates multiple components including image processing, intelligent collation, feedback learning, a document classifier, a verifier, a versioning engine, an information locator, a data extractor, a data scrubber, and manual collaboration”, & Par. [0204], “In preferred embodiments, each field subjected to MEP is extracted by a minimum of two human collaborators and the system compares the extracted value”, & Par. [0192 – 0197], “In preferred embodiments of the instant invention, the collation process is based on incremental learning and various artificial intelligence ("AI")-based techniques, which may include one or more of the following, such as: (1) the Location Diagram- and Feature Vector-based feature extraction and page mapping; (2) SVM and NLP; (3) an intelligent filter technique taking advantage of header and footer based information; (4) collation by finding common threads within or between pages, documents, or sets of documents; (5) finding disagreements based on affinities”, thus comparing, by a comparison module of the processing subsystem, the verified results of […the plurality of sub-transactions…] for identifying a plurality of differences is disclosed because Malaney teaches comparing values obtained from at least two human collaborators and identifying discrepancies between those values. Malaney’s individual fields subjected to MEP correspond to the plurality of sub-transactions. Malaney’s comparison of the extracted values from the two human collaborators corresponds to comparing the verified results. Malaney’s identified discrepancies correspond to the plurality of differences. Malaney further reinforces this comparison by teaching a collation process that includes finding disagreements based on affinities)
filtering […], by the comparison module of the processing subsystem the plurality of differences in the verified results (Malaney, Par. [0190], “For documents having peculiar properties, such as a specific variation of a class of documents, a filter may optionally be applied. An example is if two documents are very close in format and data, but they differ in a very specific property and because of that they belong to different class. A weighted filter, that is a Location Diagram with primary key set for the distinguishing property or feature, is applied so that those can be classified accurately and rapidly”, & Par. [0232], “The basic method used for classification is different from current methods. Also the manner in which and sequence the instant invention uses various complementary technologies, such as filtering and voting, makes the method of the instant invention more accurate”, thus filtering […], by the comparison module of the processing subsystem the plurality of differences in the verified results is disclosed because Malaney teaches applying filtering techniques to distinguish data or documents based on identified differences or distinguishing properties. Malaney’s differences in a specific property correspond to the plurality of differences in the verified results. Malaney’s weighted filter corresponds to filtering those identified differences so that the differing information can be separated and accurately classified)
[…] , by a quality check module of the processing subsystem, […] wherein […the second transaction…] is based on the marked differences of the verified results (Malaney, Par. [0203], “Extracted values are scrubbed to get exact value 227. Scrubbing further transforms the extracted value to a specific data type. The accuracy of the scrubbed value is verified. Thus, the system provides multiple confidence levels for decision-making”, & Par. [0204], “In preferred embodiments, each field subjected to MEP is extracted by a minimum of two human collaborators and the system compares the extracted value. In the event of a discrepancy, the value in question can be sent to additional human collaborators”, & Par. [0205], “Information related to exceptions may also be used for statistical learning. In preferred embodiments, the human collaborator mouses over the exact value to be extracted. This is referred to herein as a "snippet" or a "text snippet" and the method/apparatus can pull the snippet and subject it to further scrubbing and processing 227. These snippets of required/specific values may also be extracted and used for formation of Knowledge Objects”, thus […] , by a quality check module of the processing subsystem, […] wherein […the second transaction…] is based on the marked differences of the verified results is disclosed because Malaney teaches identifying discrepancies in verified values and using exception related information for further processing and statistical learning. Malaney’s discrepancies and exceptions correspond to the marked differences)
assigning, by the quality check module of the processing subsystem, […the second transaction…] with the marked differences to a third user for a subsequent review […] (Malaney, Par. [0204], “In preferred embodiments, each field subjected to MEP is extracted by a minimum of two human collaborators and the system compares the extracted value. In the event of a discrepancy, the value in question can be sent to additional human collaborators”, & Par. [0188], “Document pages that fall below the confidence threshold value that may be preset or varied by the user, even after the verifier, are sent to exception handling client (Classification client) (i.e., via escalation) 218. There, human collaborators can verify the class, assign a class, or note that the document cannot be identified”, thus assigning, by the quality check module of the processing subsystem, […the second transaction…] with the marked differences to a third user for a subsequent review […]is disclosed because Malaney teaches sending discrepant values identified after comparison to additional human collaborators and escalating data for further human verification after an earlier verifier step. Malaney’s additional human collaborator corresponds to the third user, and the escalated discrepancy review corresponds to the subsequent review)
updating, by the quality check module of the processing subsystem, the plurality of datapoint in response to the review made by a third user thereby enabling quality check of the one or more documents over multiple levels of verification (Malaney, Par. [0204], “Human collaborators can verify and/or change the data and/or extracted information in reference to the Dox Package. In preferred embodiments, each field subjected to MEP is extracted by a minimum of two human collaborators and the system compares the extracted value. In the event of a discrepancy, the value in question can be sent to additional human collaborators”, & Par. [0188], “Document pages that fall below the confidence threshold value that may be preset or varied by the user, even after the verifier, are sent to exception handling client (Classification client) (i.e., via escalation) 218. There, human collaborators can verify the class, assign a class, or note that the document cannot be identified”, & Par. [0189], “Frequently the identification of the page or sheet may be made in reference to the header and footer information, although display parameters, such as position of the various images on the human collaborator's computer screen and zoom capabilities, may be varied by the human collaborator. Escalation may occur before, but preferably occurs after, the verifier step”, & Par. [0203], “Extracted values are scrubbed to get exact value 227. Scrubbing further transforms the extracted value to a specific data type. The accuracy of the scrubbed value is verified. Thus, the system provides multiple confidence levels for decision-making”, thus updating, by the quality check module of the processing subsystem, the plurality of datapoints in response to the review made by a third user thereby enabling quality check of the one or more documents over multiple levels of verification is disclosed because Malaney teaches that human collaborators may verify or change extracted data, including after discrepancies are escalated for additional review. Malaney’s ability to change the extracted data corresponds to updating the plurality of datapoints. Malaney’s escalation after the verifier step corresponds to multiple levels of verification, and the subsequent accuracy verification and multiple confidence levels correspond to quality checking the documents across those verification levels)
Malaney does not explicitly teach […] by using machine learning model to generate a first transaction, generating […] a plurality of sub-transactions from the first transaction wherein each of the plurality of sub-transactions comprises […the plurality of data points…], […] and marking […the plurality of differences in the verified results ...], generating […] a second transaction by using the machine learning model […], and […] thereby resulting in a second verification level.
However, Bobley teaches:
[…] by using machine learning model to generate a first transaction (Bobley, Par. [0129], “Pydatalog is used to perform a machine learning algorithm that improves transaction detection component accuracy over a large training data set. The training data may come from a relational data store that contains both raw OCR output and post-process financial transaction information with a high degree of accuracy”, & Par. [0106], “The present system utilizes the bounding box to identify the geographic location of the information on the scanned document and generates a spatial index linking the X and Y location of the information on the document to the transaction that has been extracted. The information within a bounding box is then treated as a single transaction”, thus […] by using machine learning model to generate a first transaction is disclosed because Bobley teaches using a machine learning algorithm to process document-derived data and improve transaction detection, with extracted information associated with an identified transaction. Bobley’s machine learning algorithm corresponds to using a machine learning model. Bobley’s transaction detection and extraction process corresponds to the initial parsing and processing of document data. Bobley’s treatment of extracted information within a bounding box as a single transaction corresponds to generating a first transaction, because the first transaction represents the initial processing instance in which document data is parsed and organized into transaction information)
generating […] a plurality of sub-transactions from the first transaction wherein each of the plurality of sub-transactions comprises […the plurality of data points…] (Bobley, Par. [0150], “In step 3412, after all of the transactions have been split into ‘snippets’, the transactions, i.e, the image within the bounding boxes from transactions are displayed to the user. Below each image are fields for the corresponding date and description. The fields are populated with the respective OCR outputs, and the system and/or the user must rectify any errors in the fields”, & Par. [0149], “In one or more embodiments, the transaction date, description and amount of the highlighted transaction are viewable for the purpose of identification, but are not editable in step 3410. Accordingly, when a missing transaction is added, the fields for date, description and amount will be blank”, thus generating […] a plurality of sub-transactions from the first transaction wherein each of the plurality of sub-transactions comprises […the plurality of data points…] is disclosed because Bobley teaches splitting transactions into individual snippets for separate processing and review, where each snippet is associated with transaction information such as date, description, and amount. Bobley’s splitting of transactions into snippets corresponds to generating a plurality of sub-transactions from the first transaction. Bobley’s individual snippets correspond to the sub-transactions because they represent separated portions of the transaction that are independently displayed and processed. Bobley’s date, description, and amount fields associated with each transaction snippet correspond to the plurality of data points contained in each sub-transaction)
[…] and marking […the plurality of differences in the verified results ...] (Bobley, Par. [0016], “In one or more embodiments, the annotating includes at least one of highlighting unique transactions verified as accurate, highlighting unique transactions identified as containing errors, or highlighting unique transactions verified as accurate and unique transactions identified as containing errors. In other embodiments, the unique transactions verified as accurate are highlighted in a first color and the unique transactions identified as containing errors are highlighted in a second color”, thus […] and marking […the plurality of differences in the verified results ...] is disclosed because Bobley teaches annotating transaction data by highlighting transactions according to whether they have been verified as accurate or identified as containing errors. Bobley’s transactions identified as containing errors correspond to the differences in the verified results. Bobley’s highlighting of those transactions corresponds to marking the identified differences, and the use of different highlighting for accurate and erroneous transactions further distinguishes the marked differences from verified data)
generating […] a second transaction by using the machine learning model […] (Bobley, Par. [0129], “Pydatalog is used to perform a machine learning algorithm that improves transaction detection component accuracy over a large training data set”, & Par. [0130], “A transaction (date, description, amount, account, other_acct) def_init_(self, date, description, amount, account, other_acct): super(Transaction, self)._init_( ) # calls the initialization method of the Mixin class self. date=date self. amount=amount self. description=description @pyDatalog.program( ) # indicates that the following method contains pyDatalog clauses def Transfer(from, to, amount”, & “ # An example how a transfer is detected. acct1=‘001’ acct2=‘002’ acct3=‘003’ trans1=Transaction (‘20150108’, ‘transfer from checking acct into saving acct’, 6800, acct1, acct2) trans2=Transaction (‘20150108’, ‘transfer from checking acct into saving acct’, −6800, acct2, acct1)”, thus generating […] a second transaction by using the machine learning model […] is disclosed because Bobley teaches using a machine learning algorithm for transaction detection and subsequently representing detected transaction information as multiple transaction objects, including trans1 and trans2. Bobley’s machine learning algorithm corresponds to the machine learning model, and Bobley’s trans2 corresponds to a second transaction produced as part of the transaction-detection processing)
[…] thereby resulting in a second verification level (Bobley, Par. [0143], “In Moderation Flow, the system prompts the user to verify transactional and account data from snippets until reconciliation is achieved. After reconciliation, in step 3214, the edited transaction and account data is compiled into Final SME Review, Final SME Review includes a page-by-page human review of all data from each bank document”, thus […] thereby resulting in a second verification level is disclosed because Bobley teaches an initial Moderation Flow verification followed by a Final SME Review. Bobley’s Moderation Flow corresponds to the earlier verification stage, and the subsequent Final SME Review corresponds to the second verification level)
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Malaney’s system to incorporate Bobley’s machine learning based preprocessing and transaction processing techniques so that extracted data is processed using a machine learning model to generate transactions, the transactions are divided for verification, identified differences or errors are marked, and the resulting transaction data is subjected to subsequent review. Bobley teaches that its preprocessing techniques improve processing times, increase accuracy, and reduce errors, and that user review is performed for purposes of quality control. Therefore, A POSITA would have been motivated to use Bobley’s machine learning and transaction processing techniques in Malaney’s multi-user verification system to improve the accuracy and efficiency of generating, verifying, marking, and reviewing transaction data over multiple levels of verification (Bobley, Par. [0004], “The current system and method allows users to scan and upload bank statements, credit card statements, investment statements, and any other tabulated series of repeated entries into a web based application for processing and to identify errors in the entries, sort search and filter the entries. The system and methods utilizing preprocessing techniques improve processing times, increases accuracy and reduces errors”, & Par. [0141], “The annotations may also be used to denote or represent different, algorithmically-inferred accounts. In one or more embodiments, a user review is implemented for purposes of quality control, which includes review of each page and confirmation that all transactions are properly bounded and shaded”)
Claim 2 is rejected under 35 U.S.C. 103 as being unpatentable over Malaney et al. (hereafter Malaney) (US 20070118391) in view of Bobley et al. (hereinafter Bobley) (US 11087409), and in further view of Islamaj et al. (hereinafter Islamaj, a non-patent literature reference titled “NLM-Chem-BC7: manually annotated full-text resources for chemical entity annotation and indexing in biomedical articles”).
Regarding Claim 2, Malaney combined with Bobley teaches all the limitations of claim 1 as cited above:
Malaney combined with Bobley does not explicitly teach wherein […the first user…], […the second user…] and […the third user…] are ignorant of each other identity thereby enabling the double blinded verification of data.
However, Islamaj teaches:
wherein […the first user…], […the second user…] and […the third user…] are ignorant of each other identity thereby enabling the double blinded verification of data (Islamaj, Page 3 – Methods, “he first round of manual annotations consisted of each annotator working on and completing the annotations of the assigned articles independently. At this stage, the annotators did not know the identities of their partners”, & Page 4 – Methods, “The second round of annotations consisted of each annotator working independently in their own annotation space, again without knowing the identities of their partner annotators”, thus wherein […the first user…] and […the second user…] are ignorant of each other identity thereby enabling the double blinded verification of data is disclosed because Islamaj teaches that paired annotators perform their assigned annotations independently without knowing the identity of their annotation partner during multiple rounds of verification. Islamaj’s paired annotators correspond to the first user and second user, and their independent work without knowledge of each other’s identities corresponds to double blinded verification of data)
It would have been obvious to one of ordinary skill in the art before the effective filing date to further modify the combined system of Malaney and Bobley according to Islamaj by having the users perform verification independently without knowing the identities of the other users. Malaney already teaches verification by multiple human collaborators and resolution of discrepancies, while Islamaj teaches identity-blinded independent review in which differences and discrepancies are identified and subsequently resolved. Islamaj further reports that inter-annotator agreement increased from 68% after the first round to 84% after the second round and ultimately reached complete consensus. Therefore, a POSITA would have been motivated to use Islamaj’s identity-blinded verification technique in the combined system to maintain independent verification and facilitate identification and resolution of differences between users, thereby improving the reliability of the verification results (Islamaj, Page 3 – Methods, “After completion, these annotations were reviewed by the technical team to identify differences and discrepancies. Inter-annotator agreement was measured to be 68%”, & Page 4 – Methods, “After completion, the annotations were again reviewed, inter-annotator agreement was computed to be 84% and the remaining differences and discrepancies were analyzed”, & “n the third and final round of annotations, the annotation partners for each document were revealed, and every pair of annotators collaboratively reviewed and discussed any remaining differences and finalized the shared document annotation reaching 100% complete consensus”)
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant’s disclosure. US 20220391994 is pertinent because it teaches automatically processing financial documents, using machine learning and/or artificial intelligence to extract data from the documents, identifying missing or incorrect extracted data, and enabling a user to correct the data through an annotator interface. Because applicant’s disclosure similarly concerns automated document processing, data extraction, and verification or correction of extracted data, the reference is relevant to the invention but is not relied upon in the rejection.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MAHLIET ADMASU whose telephone number is (571)272-0034. The examiner can normally be reached Mon-Fri, 8am-5pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Alexey Shmatov can be reached at (571)270-3428. 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.
/M.T.A./Examiner, Art Unit 2123
/ALEXEY SHMATOV/Supervisory Patent Examiner, Art Unit 2123
Prosecution Insights