CTNF 18/167,855 CTNF 94442 Notice of Pre-AIA or AIA Status 07-03-aia AIA 15-10-aia The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA. Examiner’s Note Providing supporting paragraph(s) for each limitation of amended/new claim(s) in Remarks is strongly requested for clear and definite claim interpretations by Examiner (e.g., to avoid rejections under 35 U.S.C § 112(a) “Lack of written description”) Applicant can schedule an interview at any stage of the prosecution (e.g., Non-Final, Final, and After-Final ) to discuss any issues related to, for example, rejections under 35 U.S.C § 101 and § 103, for moving toward allowance. For clarification, claim 8 may be amended (e.g., “ non-transitory computer readable storage media”) based on par 75 “A computer readable storage medium , as that term is used in the present disclosure, is not to be construed as storage in the form of transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide, light pulses passing through a fiber optic cable, electrical signals communicated through a wire, and/or other transmission media” to make sure that the claim falls within one of the four statutory categories. Priority Acknowledgment is made of applicant's claim for the present application filed on 02/11/2023 . 07-30-03-h AIA Claim Interpretation 07-30-03 AIA The following is a quotation of 35 U.S.C. 112(f): (f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof. The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph: An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof. 07-30-05 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. 07-30-06 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: Claims 15-20: one or more devices (see figs 2-3 and pars 74-95) 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 07-30-02 AIA 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. 07-34-01 Claim(s) 7, 17 is/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(s) 7 recite(s) the limitation “the action” (line 1). There is insufficient antecedent basis for this limitation in the claim. It is not clear what it is referring to, since it may indicate one of the “actions” (claim 6), or something else. It appears it may need to read “an action”, or something else. For the purposes of examination, “an action” is used. Claim(s) 17 recite(s) the limitation “the actions” (lines 1, 3, 4). There is insufficient antecedent basis for this limitation in the claim. It is not clear what it is referring to, since it may indicate “recommended actions” (claim 1), or something else. It appears it may need to read “actions”, or something else. For the purposes of examination, “actions” is used. Claim(s) 7, 17 each recite(s) limitations that raise issues of indefiniteness as set forth above, and their dependent claims are rejected at least based on their direct and/or indirect dependency from the claims listed above . Appropriate explanation and/or amendment is required. Claim Rejections - 35 USC § 101 07-04-01 AIA 07-04 35 U.S.C. 101 reads as follows: Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title. Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. Regarding claim 1 Step 1 : “Is the claim to a process, machine, manufacture, or composition of matter?” The claim is directed to a method. Therefore, yes . Step 2A Prong 1 : “Does the claim recite an abstract idea, law of nature, or natural phenomenon?” performing a correlation coefficient analysis to identify a subset of labels; (i.e., mental process) performing a risk analytics process on particular first data, of the particular first document, to determine a measure of risk associated with the particular first document, (i.e., mental process) wherein the measure of risk is determined based on the one or more forecasted second documents; (i.e., mental process) evaluating the particular first data dynamically …, (i.e., mental process) wherein the particular first data is evaluated based on the measure of risk; and (i.e., mental process) performing one or more recommended actions based on evaluating the particular first data (i.e., mental process) The claim is directed to an abstract idea. Therefore, yes. Step 2A Prong 2 : “Does the claim recite additional elements that integrate the judicial exception into a practical application?” The following elements are directed to additional elements: receiving, via a network, first data regarding a plurality of first documents and second data regarding a plurality of second documents, (insignificant extra-solution activity of receiving data, see MPEP 2106.05(g), well-understood, routine, and conventional generic computer and/or model, see MPEP 2106.05(f)) wherein the first data and the second data are received from different devices; (insignificant extra-solution activity of receiving data, see MPEP 2106.05(g)) training a neural network model, based on the subset of labels, to determine a mapping between the plurality of first documents and the plurality of second documents, (adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, see MPEP 2106.05(f)) wherein the mapping indicates that one or more second documents, of the plurality of second documents, are associated with a particular first document of the plurality of first documents; (a particular type or source of model/data, Field of Use and Technological Environment, see MPEP 2106.05(h)) training a time-series forecasting model to predict one or more forecasted second documents for the particular first document; (adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, see MPEP 2106.05(f)) … using a reinforcement learning model (well-understood, routine, and conventional generic computer and/or model, see MPEP 2106.05(f)) Therefore, no. Step 2B : “Does the claim recite additional elements that amount to significantly more than the judicial exception?” The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. Specifically, the claimed inventions simply append well-understood, routine and conventional activities previously known to the industry, both when viewed independently and as an ordered combination, specified at a high level of generality, to the judicial exception, (e.g., a claim to an abstract idea requiring no more than a generic computer to perform generic computer functions that are well-understood, routine and conventional activities previously known to the industry). Therefore, no . Regarding claim 2 Step 2A Prong 1 : “Does the claim recite an abstract idea, law of nature, or natural phenomenon?” parsing the first data and the second data, …, to obtain parsed first data and parsed second data; (i.e., mental process) analyzing the parsed first data and the parsed second data to determine matches between first document labels of the particular first document and second document labels of one or more second documents of the plurality of second documents; (i.e., mental process) assigning values based on analyzing the parsed first data and the parsed second data to determine the matches, (i.e., mental process) wherein a first value is assigned to indicate a match between a first document label of the particular first document and a second document label of a second document, and (i.e., mental process) wherein a second value is assigned to indicate no match between a second document label of the particular first document and a second document label of the second document; and (i.e., mental process) generating a data structure that includes multiple entries including the first value and the second value. (i.e., mental process) The claim is directed to an abstract idea. Therefore, yes. Step 2A Prong 2 : “Does the claim recite additional elements that integrate the judicial exception into a practical application?” The following elements are directed to additional elements: … using a parsing algorithm … (well-understood, routine, and conventional generic computer and/or model, see MPEP 2106.05(f)) Therefore, no. Step 2B : “Does the claim recite additional elements that amount to significantly more than the judicial exception?” The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. Therefore, no . Regarding claim 3 Step 2A Prong 1 : “Does the claim recite an abstract idea, law of nature, or natural phenomenon?” performing the correlation coefficient analysis to identify, as a subset of most correlated labels, a subset of the first document labels and a subset of the second document labels that are most correlated out of the first document labels and the second document labels; and (i.e., mental process) The claim is directed to an abstract idea. Therefore, yes. Step 2A Prong 2 : “Does the claim recite additional elements that integrate the judicial exception into a practical application?” The following elements are directed to additional elements: training the neural network model to determine matches between the first documents and the second documents, (adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, see MPEP 2106.05(f)) wherein the neural network model is trained using the identified subset of most correlated labels (adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, see MPEP 2106.05(f)) Therefore, no. Step 2B : “Does the claim recite additional elements that amount to significantly more than the judicial exception?” The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. Therefore, no . Regarding claim 4 Step 2A Prong 1 : “Does the claim recite an abstract idea, law of nature, or natural phenomenon?” The claim recites the abstract idea identified above regarding claim 1. Therefore, yes. Step 2A Prong 2 : “Does the claim recite additional elements that integrate the judicial exception into a practical application?” The following elements are directed to additional elements: training the time-series forecasting model to predict one or more expected amounts of the one or more forecasted second documents for one or more upcoming billing cycles (adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, see MPEP 2106.05(f)) Therefore, no. Step 2B : “Does the claim recite additional elements that amount to significantly more than the judicial exception?” The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. Therefore, no . Regarding claim 5 Step 2A Prong 1 : “Does the claim recite an abstract idea, law of nature, or natural phenomenon?” determining the measure of risk of the particular first document based on an expiration time of the particular first document and a remaining amount of the particular first document (i.e., mental process) The claim is directed to an abstract idea. Therefore, yes. The claim does not add any additional elements (Step 2A Prong 2) or significantly more (Step 2B). Regarding claim 6 Step 2A Prong 1 : “Does the claim recite an abstract idea, law of nature, or natural phenomenon?” determining a performance of the particular first document based on the measure of risk, information regarding an entity associated with the particular first document, and historical performance of first documents associated with the entity; and (i.e., mental process) determining whether actions, identified for the particular first document, improve the performance of the particular first document (i.e., mental process) The claim is directed to an abstract idea. Therefore, yes. The claim does not add any additional elements (Step 2A Prong 2) or significantly more (Step 2B). Regarding claim 7 Step 2A Prong 1 : “Does the claim recite an abstract idea, law of nature, or natural phenomenon?” determining that the actions improve the performance of the particular first document; and (i.e., mental process) The claim is directed to an abstract idea. Therefore, yes. Step 2A Prong 2 : “Does the claim recite additional elements that integrate the judicial exception into a practical application?” The following elements are directed to additional elements: providing the actions as recommendations to improve the performance of the particular first document (insignificant extra-solution activity of receiving data, see MPEP 2106.05(g)) Therefore, no. Step 2B : “Does the claim recite additional elements that amount to significantly more than the judicial exception?” The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. Therefore, no . Regarding claim 8 Step 1 : “Is the claim to a process, machine, manufacture, or composition of matter?” The claim is directed to a composition of matter. Therefore, yes . Step 2A Prong 1 : “Does the claim recite an abstract idea, law of nature, or natural phenomenon?” … perform a correlation coefficient analysis to identify a subset of labels; (i.e., mental process) … perform a risk analytics process on particular PO data, of the particular PO, to determine a measure of risk associated with the particular PO, (i.e., mental process) wherein the measure of risk is determined based on the one or more forecasted invoices; (i.e., mental process) … evaluate the particular PO data dynamically …, (i.e., mental process) wherein the particular PO data is evaluated based on the measure of risk; and (i.e., mental process) … perform an action based on evaluating the particular PO data (i.e., mental process) The claim is directed to an abstract idea. Therefore, yes. Step 2A Prong 2 : “Does the claim recite additional elements that integrate the judicial exception into a practical application?” The following elements are directed to additional elements: one or more computer readable storage media, and program instructions collectively stored on the one or more computer readable storage media, the program instructions comprising: (well-understood, routine, and conventional generic computer and/or model, see MPEP 2106.05(f)) program instructions to ... (well-understood, routine, and conventional generic computer and/or model, see MPEP 2106.05(f)) receive purchase order (PO) data regarding one or more purchase orders (POs) and invoice data regarding a plurality of invoices; (insignificant extra-solution activity of receiving data, see MPEP 2106.05(g)) … train a neural network model based on the subset of labels to determine a mapping between the plurality of invoices and the one or more POs, (adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, see MPEP 2106.05(f)) wherein one or more invoices, of the plurality of invoices, are associated with a particular PO of the one or more POs; (a particular type or source of model/data, Field of Use and Technological Environment, see MPEP 2106.05(h)) … generate a time-series forecasting model to predict one or more forecasted invoices for the particular PO; (adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, see MPEP 2106.05(f)) … using a reinforcement learning model (well-understood, routine, and conventional generic computer and/or model, see MPEP 2106.05(f)) Therefore, no. Step 2B : “Does the claim recite additional elements that amount to significantly more than the judicial exception?” The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. Specifically, the claimed inventions simply append well-understood, routine and conventional activities previously known to the industry, both when viewed independently and as an ordered combination, specified at a high level of generality, to the judicial exception, (e.g., a claim to an abstract idea requiring no more than a generic computer to perform generic computer functions that are well-understood, routine and conventional activities previously known to the industry). Therefore, no . Regarding claim 9 Step 2A Prong 1 : “Does the claim recite an abstract idea, law of nature, or natural phenomenon?” … analyze the PO data to identify one or more first labels with missing values; (i.e., mental process) … analyze the invoice data to identify one or more second labels with missing values; (i.e., mental process) … analyze the PO data, … , to identify one or more first values for the one or more first labels; and (i.e., mental process) … analyze the invoice data, …, to identify one or more second values for the one or more second labels (i.e., mental process) The claim is directed to an abstract idea. Therefore, yes. Step 2A Prong 2 : “Does the claim recite additional elements that integrate the judicial exception into a practical application?” The following elements are directed to additional elements: program instructions to ... (well-understood, routine, and conventional generic computer and/or model, see MPEP 2106.05(f)) … using a natural language processing algorithm (well-understood, routine, and conventional generic computer and/or model, see MPEP 2106.05(f)) Therefore, no. Step 2B : “Does the claim recite additional elements that amount to significantly more than the judicial exception?” The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. Therefore, no . Regarding claim 10 Step 2A Prong 1 : “Does the claim recite an abstract idea, law of nature, or natural phenomenon?” … analyze the PO data and the invoice data to determine matches between PO labels of the particular PO and invoice labels of one or more invoices of the plurality of invoices; (i.e., mental process) … assign values based on analyzing the PO data and the invoice data to determine the matches, (i.e., mental process) wherein a first value is assigned to indicate a match between a first PO label of the particular PO and a first invoice label of an invoice, and (i.e., mental process) wherein a second value is assigned to indicate a match between a second PO label of the particular PO and a second invoice label of the invoice; and (i.e., mental process) … generate a data structure that includes multiple entries with the first value and the second value (i.e., mental process) The claim is directed to an abstract idea. Therefore, yes. Step 2A Prong 2 : “Does the claim recite additional elements that integrate the judicial exception into a practical application?” The following elements are directed to additional elements: program instructions to ... (well-understood, routine, and conventional generic computer and/or model, see MPEP 2106.05(f)) Therefore, no. Step 2B : “Does the claim recite additional elements that amount to significantly more than the judicial exception?” The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. Therefore, no . Regarding claim 11 Step 2A Prong 1 : “Does the claim recite an abstract idea, law of nature, or natural phenomenon?” … perform the correlation coefficient analysis to identify, as a subset of most correlated labels, a subset of the PO labels and a subset of the invoice labels that are most correlated out of the PO labels and the invoice labels; and (i.e., mental process) The claim is directed to an abstract idea. Therefore, yes. Step 2A Prong 2 : “Does the claim recite additional elements that integrate the judicial exception into a practical application?” The following elements are directed to additional elements: program instructions to ... (well-understood, routine, and conventional generic computer and/or model, see MPEP 2106.05(f)) … train a neural network model to determine matches between POs and invoices, (adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, see MPEP 2106.05(f)) wherein the neural network model is trained using the identified subset of most correlated labels (adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, see MPEP 2106.05(f)) Therefore, no. Step 2B : “Does the claim recite additional elements that amount to significantly more than the judicial exception?” The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. Therefore, no . Regarding claim 12 Step 2A Prong 1 : “Does the claim recite an abstract idea, law of nature, or natural phenomenon?” … determine the measure of risk of the particular PO based on an expiration time of the particular PO and a remaining amount of the particular PO (i.e., mental process) The claim is directed to an abstract idea. Therefore, yes. Step 2A Prong 2 : “Does the claim recite additional elements that integrate the judicial exception into a practical application?” The following elements are directed to additional elements: program instructions to ... (well-understood, routine, and conventional generic computer and/or model, see MPEP 2106.05(f)) Therefore, no. Step 2B : “Does the claim recite additional elements that amount to significantly more than the judicial exception?” The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. Therefore, no . Regarding claim 13 Step 2A Prong 1 : “Does the claim recite an abstract idea, law of nature, or natural phenomenon?” … determine a performance of the particular PO based on the measure of risk, information regarding an entity associated with the particular PO, and historical performance of POs associated with the entity; and (i.e., mental process) … determine whether actions, identified for the particular PO, improve the performance of the particular PO (i.e., mental process) The claim is directed to an abstract idea. Therefore, yes. Step 2A Prong 2 : “Does the claim recite additional elements that integrate the judicial exception into a practical application?” The following elements are directed to additional elements: program instructions to ... (well-understood, routine, and conventional generic computer and/or model, see MPEP 2106.05(f)) Therefore, no. Step 2B : “Does the claim recite additional elements that amount to significantly more than the judicial exception?” The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. Therefore, no . Regarding claim 14 Step 2A Prong 1 : “Does the claim recite an abstract idea, law of nature, or natural phenomenon?” … determine that the actions improve the performance of the particular PO; and (i.e., mental process) The claim is directed to an abstract idea. Therefore, yes. Step 2A Prong 2 : “Does the claim recite additional elements that integrate the judicial exception into a practical application?” The following elements are directed to additional elements: program instructions to ... (well-understood, routine, and conventional generic computer and/or model, see MPEP 2106.05(f)) program instructions to provide the actions as recommendations to improve the performance of the particular PO (well-understood, routine, and conventional generic computer and/or model, see MPEP 2106.05(f), and insignificant extra-solution activity of transmitting data, see MPEP 2106.05(g)) Therefore, no. Step 2B : “Does the claim recite additional elements that amount to significantly more than the judicial exception?” The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. Therefore, no . Regarding claim 15 The claim is rejected for the reasons set forth in the rejection of Claim 8 under 35 U.S.C. 101, mutatis mutandis. Regarding claim 16 The claim is rejected for the reasons set forth in the rejection of Claim 13 under 35 U.S.C. 101, mutatis mutandis. Regarding claim 17 The claim is rejected for the reasons set forth in the rejection of Claim 14 under 35 U.S.C. 101, mutatis mutandis. Regarding claim 18 The claim is rejected for the reasons set forth in the rejection of Claim 9 under 35 U.S.C. 101, mutatis mutandis. In addition, Step 2A Prong 2 : “Does the claim recite additional elements that integrate the judicial exception into a practical application?” The following elements are directed to additional elements: receive PO data regarding a plurality of purchase orders (POs) and invoice data regarding a plurality of invoices (insignificant extra-solution activity of receiving data, see MPEP 2106.05(g)) wherein the PO data and the invoice data are received from different devices associated with different cloud systems (insignificant extra-solution activity of receiving data, see MPEP 2106.05(g)) Therefore, no. Step 2B : “Does the claim recite additional elements that amount to significantly more than the judicial exception?” The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. Therefore, no . Regarding claim 19 The claim is rejected for the reasons set forth in the rejection of Claim 10 under 35 U.S.C. 101, mutatis mutandis. In addition, Step 2A Prong 2 : “Does the claim recite additional elements that integrate the judicial exception into a practical application?” The following elements are directed to additional elements: receive, via a network, PO data regarding the plurality of POs and invoice data regarding the plurality of invoices (insignificant extra-solution activity of receiving data, see MPEP 2106.05(g)) and (well-understood, routine, and conventional generic computer and/or model, see MPEP 2106.05(f)) Therefore, no. Step 2B : “Does the claim recite additional elements that amount to significantly more than the judicial exception?” The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. Therefore, no . Regarding claim 20 The claim is rejected for the reasons set forth in the rejection of Claim 12 under 35 U.S.C. 101, mutatis mutandis. Claim Rejections - 35 USC § 103 07-06 AIA 15-10-15 In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status. 07-20-02-aia AIA This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention. 07-20-aia AIA 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. 07-21-aia AIA Claim (s) 1-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over MEDALION et al. (US 2021/0287261 A1) in view of BURRELL et al. (US 2022/0327502 A1) in view of VASHISHT et al. (US 2022/0261875 A1) Regarding claim 1 MEDALION teaches A computer-implemented method comprising: (MEDALION [fig(s) 10-11]) receiving, via a network, first data regarding a plurality of first documents and second data regarding a plurality of second documents, (MEDALION [fig(s) 1, 10] [par(s) 17-19] “The method may analyze the vendor offerings, identify the associated clusters, and generate a bounded list of purchases for that business. For example, a pre-defined list of purchase orders that a dry-cleaning business may select from within accounting software may include "hanger and accessories" and "pressing accessories." The method of the present disclosure may analyze the dry-cleaning business' purchases and identify associated clusters (e.g., formed by analyzing other dry cleaning services' purchases) with more specialized descriptions such as "plastic bags," "irons," and "press pads."” [par(s) 3] “In addition, it is difficult to summarize the services or products offered by a business, despite databases typically having access to the business' invoices and bank transactions. Obtaining accurate and appropriate lists of business offerings (e.g., services and/or products) requires manual entry or adherence to a pre-defined list when selecting offerings.” [par(s) 32] “Invoice preparation module 118 may be configured to extract text from invoices. In some embodiments, invoice preparation module 118 may be configured to use optical character recognition (OCR) to extract text from invoice files (e.g., PDFs, Word Documents, etc.) or images (e.g., PNG, JPEG, etc.). Invoice files or images may be stored in database 122. In some embodiments, invoice preparation module 118 may be configured to generate a representative vector of a vendor based on a plurality of vectors describing line items of the invoice.” ; ) wherein the first data and the second data are received from different devices; (MEDALION [fig(s) 1, 10] “ User Device”s [par(s) 17-19] “The method may analyze the vendor offerings, identify the associated clusters , and generate a bounded list of purchases for that business. For example, a pre-defined list of purchase orders that a dry-cleaning business may select from within accounting software may include "hanger and accessories" and "pressing accessories." The method of the present disclosure may analyze the dry-cleaning business' purchases and identify associated clusters (e.g., formed by analyzing other dry cleaning services' purchases ) with more specialized descriptions such as "plastic bags," "irons," and "press pads."” [par(s) 3] “In addition, it is difficult to summarize the services or products offered by a business, despite databases typically having access to the business' invoices and bank transactions. Obtaining accurate and appropriate lists of business offerings (e.g., services and/or products) requires manual entry or adherence to a pre-defined list when selecting offerings.” [par(s) 32] “Invoice preparation module 118 may be configured to extract text from invoices. In some embodiments, invoice preparation module 118 may be configured to use optical character recognition (OCR) to extract text from invoice files (e.g., PDFs, Word Documents, etc.) or images (e.g., PNG, JPEG, etc.). Invoice files or images may be stored in database 122. In some embodiments, invoice preparation module 118 may be configured to generate a representative vector of a vendor based on a plurality of vectors describing line items of the invoice.” ; ) performing a correlation coefficient analysis to identify a subset of labels; (MEDALION [par(s) 33] “Distance metric calculator 120 may be configured to calculate a relation metric between two vectors. In some embodiments, relation metric calculator 120 may be con figured to calculate a relation metric between a merchant vector and a vendor vector. A relation metric may represent a degree of relation between the vectors and thus the corresponding businesses. A low relation metric may correspond to a low relation between two vectors. For example, a relation metric calculated between a merchant vector for a tax service business and a vendor vector for a pool cleaning service may be low or near-zero. A relation metric calculated between a merchant vector for a Jacuzzi supplier and a vendor vector for a pool supplier may be higher, due to the improved commonalities between the merchants and vendors. A relation metric calculated between a merchant vector for a vehicle repair service and a vendor vector for an auto body shop may be higher than the previous two examples. In some embodiments, the relation metrics may be used to train a neural network to predict related/unrelated pairs of vendors and merchants. In some embodiments the relation metric may be an inner product between the two vectors. In some embodiments, relation metric calculator 120 may be configured to apply a sigmoid function to confine the relation metric between zero and one. In some embodiments, relation metric calculator 120 may be configured to deter mine services related to a business.” [par(s) 49] “Cluster analysis module 116 may calculate a TFIDF value for each word within the cluster. In some embodiments, the TFIDF value may be calculated in relation to the entire cluster, as opposed to the TFIDF value being calculated in relation to the line item that contains the word. Cluster analysis module 116 may identify the highest scoring words and combine them to generate a representative description of the cluster.” ; ) ( Note : Hereinafter, if a limitation has bold brackets (i.e. [·] ) around claim languages, the bracketed claim languages indicate that they have not been taught yet by the current prior art reference but they will be taught by another prior art reference afterwards.) training a neural network model, based on the subset of labels, to determine a mapping [between] the plurality of first documents and the plurality of second documents, (MEDALION [par(s) 16] “ A method may be used to train a neural network architecture via supervised learning to predict missing categories for businesses based on their given description .” [par(s) 33] “Distance metric calculator 120 may be configured to calculate a relation metric between two vectors. In some embodiments, relation metric calculator 120 may be con figured to calculate a relation metric between a merchant vector and a vendor vector. A relation metric may represent a degree of relation between the vectors and thus the corresponding businesses. A low relation metric may correspond to a low relation between two vectors. For example, a relation metric calculated between a merchant vector for a tax service business and a vendor vector for a pool cleaning service may be low or near-zero. A relation metric calculated between a merchant vector for a Jacuzzi supplier and a vendor vector for a pool supplier may be higher, due to the improved commonalities between the merchants and vendors. A relation metric calculated between a merchant vector for a vehicle repair service and a vendor vector for an auto body shop may be higher than the previous two examples. In some embodiments, the relation metrics may be used to train a neural network to predict related/unrelated pairs of vendors and merchants . In some embodiments the relation metric may be an inner product between the two vectors. In some embodiments, relation metric calculator 120 may be configured to apply a sigmoid function to confine the relation metric between zero and one. In some embodiments, relation metric calculator 120 may be configured to deter mine services related to a business.” ; ) training a time-series forecasting model to predict one or more forecasted second documents for the particular first document; (MEDALION [par(s) 16] “Embodiments of the present disclosure relate to various systems and methods that may predict a business' category based e.g., on a given business description or an associated vendor . For example, in some types of accounting software, only a portion of the users ( e.g., businesses) supply both a category and a description of the business. Many users supply only one of the two, and some users do not supply either. A method may be used to train a neural network architecture via supervised learning to predict missing categories for businesses based on their given description. The network may be trained using existing businesses that have provided both descriptions and categories.” [par(s) 21] “some embodiments may reduce the amount of data required to be stored by consolidating portions of the data having similar meanings .” [par(s) 32] “ Invoice preparation module 118 may combine words (in vector format or in the vector space) within each line item using a long short-term memory (LSTM) layer. … Invoice preparation module 118 may also utilize a convolutional neural network to combine line item vectors ( or invoice line vectors) associated with the same vendor to create a vector representing that vendor . In some embodiments, the convolutional neural network may be configured to add or subtract the plurality of line item vectors with various weights to create a single vector representing the vendor.” ; ) performing a risk analytics process on particular first data, of the particular first document, to determine a measure of risk associated with the particular first document, (MEDALION [par(s) 16] “a method may be used to generate a list of factors associated with a business' success . The factors may be generated by analyzing, aggregating and ranking factors determined to be relevant to a business based on its categorization. The factors associated with the business' success may be related to the products and/or services offered by the business and the format of which those products and/or services are offered by the business . The factors may also be related to the products and/or services purchased by the business from a vendor and the format of which those products and/or services are purchased from the vendor” [par(s) 33] “Distance metric calculator 120 may be configured to calculate a relation metric between two vectors. In some embodiments, relation metric calculator 120 may be con figured to calculate a relation metric between a merchant vector and a vendor vector. A relation metric may represent a degree of relation between the vectors and thus the corresponding businesses. A low relation metric may correspond to a low relation between two vectors. For example, a relation metric calculated between a merchant vector for a tax service business and a vendor vector for a pool cleaning service may be low or near-zero. A relation metric calculated between a merchant vector for a Jacuzzi supplier and a vendor vector for a pool supplier may be higher, due to the improved commonalities between the merchants and vendors. A relation metric calculated between a merchant vector for a vehicle repair service and a vendor vector for an auto body shop may be higher than the previous two examples. In some embodiments, the relation metrics may be used to train a neural network to predict related/unrelated pairs of vendors and merchants . In some embodiments the relation metric may be an inner product between the two vectors. In some embodiments, relation metric calculator 120 may be configured to apply a sigmoid function to confine the relation metric between zero and one. In some embodiments, relation metric calculator 120 may be configured to deter mine services related to a business.” ; ) wherein the measure of risk is determined based on the one or more forecasted second documents; (MEDALION [par(s) 16] “a method may be used to generate a list of factors associated with a business' success . The factors may be generated by analyzing, aggregating and ranking factors determined to be relevant to a business based on its categorization. The factors associated with the business' success may be related to the products and/or services offered by the business and the format of which those products and/or services are offered by the business . The factors may also be related to the products and/or services purchased by the business from a vendor and the format of which those products and/or services are purchased from the vendor” [par(s) 33] “Distance metric calculator 120 may be configured to calculate a relation metric between two vectors. In some embodiments, relation metric calculator 120 may be con figured to calculate a relation metric between a merchant vector and a vendor vector. A relation metric may represent a degree of relation between the vectors and thus the corresponding businesses. A low relation metric may correspond to a low relation between two vectors. For example, a relation metric calculated between a merchant vector for a tax service business and a vendor vector for a pool cleaning service may be low or near-zero. A relation metric calculated between a merchant vector for a Jacuzzi supplier and a vendor vector for a pool supplier may be higher, due to the improved commonalities between the merchants and vendors. A relation metric calculated between a merchant vector for a vehicle repair service and a vendor vector for an auto body shop may be higher than the previous two examples. In some embodiments, the relation metrics may be used to train a neural network to predict related/unrelated pairs of vendors and merchants .” [par(s) 32] “Invoice preparation module 118 may combine words (in vector format or in the vector space) within each line item using a long short- term memory (LSTM) layer. … Invoice preparation module 118 may also utilize a convolutional neural network to combine line item vectors ( or invoice line vectors) associated with the same vendor to create a vector representing that vendor. In some embodiments, the convolutional neural network may be configured to add or subtract the plurality of line item vectors with various weights to create a single vector representing the vendor.” ; ) evaluating the particular first data dynamically using a [reinforcement] learning model, (MEDALION [fig(s) 5] “Embedding”, “LSTM”s, “CNN” [par(s) 26] “Embedding module 110 may be configured to embed text to vector form within a continuous vector space. In some embodiments, embedding module 110 may convert business-related text into a merchant vector within a continuous vector space.” [par(s) 32] “Invoice preparation module 118 may combine words (in vector format or in the vector space) within each line item using a long short-term memory (LSTM) layer. … Invoice preparation module 118 may also utilize a convolutional neural network to combine line item vectors ( or invoice line vectors) associated with the same vendor to create a vector representing that vendor. In some embodiments, the convolutional neural network may be configured to add or subtract the plurality of line item vectors with various weights to create a single vector representing the vendor.” ; ) wherein the particular first data is evaluated based on the measure of risk; and (MEDALION [fig(s) 5] “Embedding”, “LSTM”s, “CNN” [par(s) 16] “a method may be used to generate a list of factors associated with a business' success . … The factors associated with the business' success may be related to the products and/or services offered by the business and the format of which those products and/or services are offered by the business .” [par(s) 33] “ A relation metric may represent a degree of relation between the vectors and thus the corresponding businesses. A low relation metric may correspond to a low relation between two vectors.” [par(s) 32] “Invoice preparation module 118 may combine words (in vector format or in the vector space) within each line item using a long short-term memory (LSTM) layer. … Invoice preparation module 118 may also utilize a convolutional neural network to combine line item vectors ( or invoice line vectors) associated with the same vendor to create a vector representing that vendor. In some embodiments, the convolutional neural network may be configured to add or subtract the plurality of line item vectors with various weights to create a single vector representing the vendor.” [par(s) 55] “ Back propagation algorithms may include gradient-based learning processes for training multilayer networks .” ; ) performing one or more recommended actions based on evaluating the particular first data. (MEDALION [fig(s) 5] “Embedding”, “LSTM”s, “CNN” [par(s) 22] “FIG. 1 is a block diagram of an example system 100 for aggregating a database of businesses, according to an embodiment of the present disclosure. System 100 may include a plurality of user devices 102a, 100b, ... , 102n (102 generally) and a server device 106, all of which may be communicably coupled via network 104. In some embodiments, system 100 may include any number of user devices. For example, for an organization that manages accounting software and an associated database, there may be an extensive user base with thousands or even millions of users that may connect via respective user devices. Server device 106 may be configured to selectively send a variety of information, such as product or service recommendations or business categorizations, to multiple or single user devices .” ; ) However, MEDALION does not appear to explicitly teach: training a neural network model, based on the subset of labels, to determine a mapping [between] the plurality of first documents and the plurality of second documents, wherein the mapping indicates that one or more second documents, of the plurality of second documents, are associated with a particular first document of the plurality of first documents; evaluating the particular first data dynamically using a [reinforcement] learning model, ( Note : Hereinafter, if a limitation has one or more bold underlines, the one or more underlined claim languages indicate that they are taught by the current prior art reference, while the one or more non-underlined claim languages indicate that they have been taught already by one or more previous art references.) BURRELL teaches training a neural network model, based on the subset of labels, to determine a mapping between the plurality of first documents and the plurality of second documents, (BURRELL [fig(s) 1] [par(s) 52] “Identities 123 may include mappings between different attributes . For example, a purchase order document may include the purchase order number stored as an attribute in attributes 122 that is mapped to an invoice number in an invoice document to form a sale identity . Character recognition module 111 may determine identities 123 based on matching attributes between different documents. For example, purchase order and invoice documents described above can list the same set of products/services requested and provided, resulting in the linking of the purchase order and the invoice attributes and create a sale identity . Identities 123 obtained from mapping attributes of attributes 122 may include combined values of attributes.” [par(s) 85] “In step 430, transaction processing system 100 may group attributes into sets of attributes. Transaction processing system 100 may group attributes based on relationships (e.g., correlations 224 of FIG. 2) defined in transaction processing system 100. In some embodiments, relationships between attributes may be learned from previous iterations of transaction processing by transaction processing system 100. Transaction processing system 100 may include machine learning models that may be trained to learn different types of documents, their structure, and the relationships between various attributes that may be part of documents. Machine learning models used in transaction processing system 100 may include neural networks . In some embodiments, transaction processing system 100 components may determine document types using fuzzy logic systems.” See also [par(s) 68-72] ; ) wherein the mapping indicates that one or more second documents, of the plurality of second documents, are associated with a particular first document of the plurality of first documents; (BURRELL [fig(s) 1] [par(s) 52] “Identities 123 may include mappings between different attributes . For example, a purchase order document may include the purchase order number stored as an attribute in attributes 122 that is mapped to an invoice number in an invoice document to form a sale identity . Character recognition module 111 may determine identities 123 based on matching attributes between different documents. For example, purchase order and invoice documents described above can list the same set of products/services requested and provided, resulting in the linking of the purchase order and the invoice attributes and create a sale identity . Identities 123 obtained from mapping attributes of attributes 122 may include combined values of attributes.” [par(s) 85] “In step 430, transaction processing system 100 may group attributes into sets of attributes. Transaction processing system 100 may group attributes based on relationships (e.g., correlations 224 of FIG. 2) defined in transaction processing system 100. In some embodiments, relationships between attributes may be learned from previous iterations of transaction processing by transaction processing system 100. Transaction processing system 100 may include machine learning models that may be trained to learn different types of documents, their structure, and the relationships between various attributes that may be part of documents. Machine learning models used in transaction processing system 100 may include neural networks . In some embodiments, transaction processing system 100 components may determine document types using fuzzy logic systems.” See also [par(s) 68-72] ; ) Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the system of MEDALION with the mapping between POs and invoices of BURRELL . One of ordinary skill in the art would have been motived to combine in order to improve handling a varied set of transactions without knowing the context of the transactions or the documents needed to process a transaction. (BURRELL [par(s) 6] “Such awareness requirements limit a customer from having different forms of transaction with a varied set of steps and different forms of payments to be processed without prior knowledge. Therefore, there is a need for improved methods of systems for handling a varied set of transactions without knowing the context of the transactions or the documents needed to process a transaction.”) However, the combination of MEDALION, BURRELL does not appear to explicitly teach: evaluating the particular first data dynamically using a [reinforcement] learning model, VASHISHT teaches evaluating the particular first data dynamically using a reinforcement learning model, (VASHISHT [par(s) 80] “The RL agent 222 is configured to initialize Q-value function and learn the best optimal path for a particular type of payment transaction based on a reward function . The reward function depends upon approval and fraud probability scores of the particular type of transaction and the cost of applying authorizing components to the particular type of transaction. In one embodiment, the approval and fraud probability scores of the particular type of transaction are determined based on historical transaction data (i.e., a number of processed transactions that were approved or declined due to fraud). In one embodiment, the cost of applying an authorizing component is a transaction-level cost and stored at the product cost repository 228.” [par(s) 3] “Payment networks enable various types of payment transactions. A " card-not-present (CNP)" transaction is a type of payment transaction in which a consumer buys a product/service without the presence of a physical payment card (e.g., debit card, credit card, prepaid card ). In such transactions ( e.g., online/e-commerce, card-on-file), the payment card information is transmitted from a merchant, along with a flag that the payment transaction is a CNP transaction.” [par(s) 140] “The various data elements may include, but not limited to, transaction identifier, issuer identifier, issuer name, merchant name/identifier, acquirer name/identifier, cross-border transaction flag (e.g., cross border, domestic), transaction channel flag (e.g., e-commerce, POS, recurring payments), payment card type (e.g., credit, debit), card product type ( customer/commercial), card-not-present (CNP) transaction flag, response code flag (approve/decline), decline reason code (in case of declined transaction), product flag vectors (indicating applied authorizing components by the issuers), etc.” ; ) Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the system of MEDALION, BURRELL with the reinforcement learning model of VASHISHT . One of ordinary skill in the art would have been motived to combine in order to provide automated means for finding which authorization decision products are needed to be applied to a particular payment transaction. (VASHISHT [par(s) 8] “Thus, there exists a technological need for a technical solution for finding which authorization decision products are needed to be applied to a particular payment transaction in the card-not-present (CNP) payment transactions using automated means”) Regarding claim 2 The combination of MEDALION, BURRELL, VASHISHT teaches claim 1. BURRELL further teaches parsing the first data and the second data, using a parsing algorithm, to obtain parsed first data and parsed second data; (BURRELL [par(s) 20] “Transaction processing system 100 processing various types of documents may receive scanned images of documents requiring parsing the text present in the document and placement of text in the document . In some embodiments, transaction processing system 100 may parse graphical elements or the placement of graphical elements in the document.” [par(s) 26] “Character recognition module 111 may include a feature to extract text elements from an image of an accessed document of document images 121. Character recognition module 111 may identify relationships between extracted text elements. Character recognition module 111 may include a machine learning mod& to help extract text elements and identify relationships between text elements.” [par(s) 94] “In step 510, transaction processing system 100 may extract text elements in a document image. Transaction processing system 100 may obtain document image based on transaction processing request 150 identifying document images in document images 121 or from the transaction processing request 150 itself. Transaction processing system 100 may employ character recognition module 111 to extract text elements from document image.” [fig(s) 1] [par(s) 52] “Identities 123 may include mappings between different attributes . For example, a purchase order document may include the purchase order number stored as an attribute in attributes 122 that is mapped to an invoice number in an invoice document to form a sale identity . Character recognition module 111 may determine identities 123 based on matching attributes between different documents. For example, purchase order and invoice documents described above can list the same set of products/services requested and provided, resulting in the linking of the purchase order and the invoice attributes and create a sale identity ” ; ) analyzing the parsed first data and the parsed second data to determine matches between first document labels of the particular first document and second document labels of one or more second documents of the plurality of second documents; (BURRELL [fig(s) 1] [par(s) 52] “Identities 123 may include mappings between different attributes . For example, a purchase order document may include the purchase order number stored as an attribute in attributes 122 that is mapped to an invoice number in an invoice document to form a sale identity . Character recognition module 111 may determine identities 123 based on matching attributes between different documents. For example, purchase order and invoice documents described above can list the same set of products/services requested and provided, resulting in the linking of the purchase order and the invoice attributes and create a sale identity . Identities 123 obtained from mapping attributes of attributes 122 may include combined values of attributes.” [par(s) 84] “transaction processing system 100 may use document type information to look for certain keywords in extracted text elements and use them to determine attributes. Transaction processing system 100 may determine attributes with content associated with key word matching text elements as attribute values. For example, transaction processing system 100 upon determining input document image as payment check type document can search for keywords "payable" to determine account and amount attributes from the values typed or written following "payable" keyword.” [par(s) 85] “In step 430, transaction processing system 100 may group attributes into sets of attributes. Transaction processing system 100 may group attributes based on relationships (e.g., correlations 224 of FIG. 2) defined in transaction processing system 100. In some embodiments, relationships between attributes may be learned from previous iterations of transaction processing by transaction processing system 100 ” See also [par(s) 68-72] ; ) assigning values based on analyzing the parsed first data and the parsed second data to determine the matches, wherein a first value is assigned to indicate a match between a first document label of the particular first document and a second document label of a second document, and wherein a second value is assigned to indicate no match between a second document label of the particular first document and a second document label of the second document; and (BURRELL [fig(s) 1] [par(s) 52] “Identities 123 may include mappings between different attributes . For example, a purchase order document may include the purchase order number stored as an attribute in attributes 122 that is mapped to an invoice number in an invoice document to form a sale identity . Character recognition module 111 may determine identities 123 based on matching attributes between different documents. For example, purchase order and invoice documents described above can list the same set of products/services requested and provided, resulting in the linking of the purchase order and the invoice attributes and create a sale identity . Identities 123 obtained from mapping attributes of attributes 122 may include combined values of attributes.” [par(s) 31] “Character recognition module 111 may identify and extract text elements and share with confidence scoring module 112 to generate confidence scores that a particular text element represents a particular attribute. For example, an element of text with 3-5 numbers, followed by at least 5 letters, may have a high confidence score for "address."” [par(s) 90] “In step 480, transaction processing system 100 may check whether confidence levels of the association links between transactions and accounts are above a threshold value. If the answer to the question in step 480 is No, then jump to step 499. If the answer to the question in step 480 is yes, then proceed to step 490.” [par(s) 85] “In step 430, transaction processing system 100 may group attributes into sets of attributes. Transaction processing system 100 may group attributes based on relationships (e.g., correlations 224 of FIG. 2) defined in transaction processing system 100. In some embodiments, relationships between attributes may be learned from previous iterations of transaction processing by transaction processing system 100 ” See also [par(s) 68-72] ; ) generating a data structure that includes multiple entries including the first value and the second value. (BURRELL [fig(s) 1-3] [par(s) 52] “Identities 123 may include mappings between different attributes . For example, a purchase order document may include the purchase order number stored as an attribute in attributes 122 that is mapped to an invoice number in an invoice document to form a sale identity . Character recognition module 111 may determine identities 123 based on matching attributes between different documents. For example, purchase order and invoice documents described above can list the same set of products/services requested and provided, resulting in the linking of the purchase order and the invoice attributes and create a sale identity . Identities 123 obtained from mapping attributes of attributes 122 may include combined values of attributes.” [par(s) 85] “In step 430, transaction processing system 100 may group attributes into sets of attributes. Transaction processing system 100 may group attributes based on relationships (e.g., correlations 224 of FIG. 2) defined in transaction processing system 100. In some embodiments, relationships between attributes may be learned from previous iterations of transaction processing by transaction processing system 100. ” [par(s) 68-69] “Positive indicators” and “Negative indicators” See also [par(s) 70-72] [par(s) 73] “Image processing engine 110 may generate various different outputs 230 related to a transaction including document types 231 of input document images 121, account assignments 232 associated with documents represented by document images 121, and transaction structures 233 forming transactions (e.g., transactions 124) using contents of document images 121. Various outputs 230 may be linked together.” [par(s) 29] “Character recognition module 111 may deter mine attributes in a document of document images 121 and stores them as attributes 122 in data store 120.” ; ) The combination of MEDALION, BURRELL, VASHISHT is combinable with BURRELL for the same rationale as set forth above with respect to claim 1. Regarding claim 3 The combination of MEDALION, BURRELL, VASHISHT teaches claim 2. wherein performing the correlation coefficient analysis comprises: (See claim 1) MEDALION further teaches performing the correlation coefficient analysis to identify, as a subset of most correlated labels, a subset of the first document labels and a subset of the [second document] labels that are most correlated out of the first document labels and the [second document] labels; and (MEDALION [par(s) 18] “According to another embodiment of the present disclosure, a method may be used to generate a list of factors associated with a business' success. The factors may be generated by analyzing, aggregating and ranking factors determined to be relevant to a business based on its categorization . The factors associated with the business' success may be related to the products and/or services offered by the business and the format of which those products and/or services are offered by the business. The factors may also be related to the products and/or services purchased by the business from a vendor and the format of which those products and/or services are purchased from the vendor.” [par(s) 17-19] “For example, a pre-defined list of purchase orders that a dry-cleaning business may select from within accounting software may include "hanger and accessories" and "pressing accessories."” [par(s) 3] “In addition, it is difficult to summarize the services or products offered by a business, despite databases typically having access to the business' invoices and bank transactions.” [par(s) 33] “ A relation metric may represent a degree of relation between the vectors and thus the corresponding businesses. A low relation metric may correspond to a low relation between two vectors. For example, a relation metric calculated between a merchant vector for a tax service business and a vendor vector for a pool cleaning service may be low or near-zero. A relation metric calculated between a merchant vector for a Jacuzzi supplier and a vendor vector for a pool supplier may be higher, due to the improved commonalities between the merchants and vendors. A relation metric calculated between a merchant vector for a vehicle repair service and a vendor vector for an auto body shop may be higher than the previous two examples. In some embodiments, the relation metrics may be used to train a neural network to predict related/unrelated pairs of vendors and merchants.” ; ) wherein the neural network model is trained using the identified subset of most correlated labels. (MEDALION [par(s) 18] “According to another embodiment of the present disclosure, a method may be used to generate a list of factors associated with a business' success. The factors may be generated by analyzing, aggregating and ranking factors determined to be relevant to a business based on its categorization . The factors associated with the business' success may be related to the products and/or services offered by the business and the format of which those products and/or services are offered by the business. The factors may also be related to the products and/or services purchased by the business from a vendor and the format of which those products and/or services are purchased from the vendor.” [par(s) 33] “ A relation metric may represent a degree of relation between the vectors and thus the corresponding businesses. A low relation metric may correspond to a low relation between two vectors. For example, a relation metric calculated between a merchant vector for a tax service business and a vendor vector for a pool cleaning service may be low or near-zero. A relation metric calculated between a merchant vector for a Jacuzzi supplier and a vendor vector for a pool supplier may be higher, due to the improved commonalities between the merchants and vendors. A relation metric calculated between a merchant vector for a vehicle repair service and a vendor vector for an auto body shop may be higher than the previous two examples. In some embodiments, the relation metrics may be used to train a neural network to predict related/unrelated pairs of vendors and merchants.” ; ) BURRELL further teaches performing the correlation coefficient analysis to identify, as a subset of most correlated labels, a subset of the first document labels and a subset of the second document labels that are most correlated out of the first document labels and the second document labels; and (BURRELL [fig(s) 1] [par(s) 52] “Identities 123 may include mappings between different attributes . For example, a purchase order document may include the purchase order number stored as an attribute in attributes 122 that is mapped to an invoice number in an invoice document to form a sale identity . Character recognition module 111 may determine identities 123 based on matching attributes between different documents. For example, purchase order and invoice documents described above can list the same set of products/services requested and provided, resulting in the linking of the purchase order and the invoice attributes and create a sale identity . Identities 123 obtained from mapping attributes of attributes 122 may include combined values of attributes.” [par(s) 85] “In step 430, transaction processing system 100 may group attributes into sets of attributes. Transaction processing system 100 may group attributes based on relationships (e.g., correlations 224 of FIG. 2) defined in transaction processing system 100. In some embodiments, relationships between attributes may be learned from previous iterations of transaction processing by transaction processing system 100. Transaction processing system 100 may include machine learning models that may be trained to learn different types of documents, their structure, and the relationships between various attributes that may be part of documents. Machine learning models used in transaction processing system 100 may include neural networks . In some embodiments, transaction processing system 100 components may determine document types using fuzzy logic systems.” [par(s) 67-73] “ Rankings 225 may include ordering of different attributes associated with a text element extracted from a document requested for processing using transaction processing request 150 . Rankings 225 may also include ordering different attributes associated with different text elements and different documents. Rankings 225 may be based on confidence scores assigned to attributes by confidence scoring module 112” ; ) training the neural network model to determine matches between the first documents and the second documents, (BURRELL [fig(s) 1] [par(s) 52] “Identities 123 may include mappings between different attributes . For example, a purchase order document may include the purchase order number stored as an attribute in attributes 122 that is mapped to an invoice number in an invoice document to form a sale identity . Character recognition module 111 may determine identities 123 based on matching attributes between different documents. For example, purchase order and invoice documents described above can list the same set of products/services requested and provided, resulting in the linking of the purchase order and the invoice attributes and create a sale identity . Identities 123 obtained from mapping attributes of attributes 122 may include combined values of attributes.” [par(s) 85] “In step 430, transaction processing system 100 may group attributes into sets of attributes. Transaction processing system 100 may group attributes based on relationships (e.g., correlations 224 of FIG. 2) defined in transaction processing system 100. In some embodiments, relationships between attributes may be learned from previous iterations of transaction processing by transaction processing system 100. Transaction processing system 100 may include machine learning models that may be trained to learn different types of documents, their structure, and the relationships between various attributes that may be part of documents. Machine learning models used in transaction processing system 100 may include neural networks . In some embodiments, transaction processing system 100 components may determine document types using fuzzy logic systems.” See also [par(s) 68-72] ; ) The combination of MEDALION, BURRELL, VASHISHT is combinable with BURRELL for the same rationale as set forth above with respect to claim 1. Regarding claim 4 The combination of MEDALION, BURRELL, VASHISHT teaches claim 1. wherein training the time-series forecasting model comprises: (See claim 1) MEDALION further teaches training the time-series forecasting model to predict one or more expected amounts of the one or more forecasted second documents for one or more upcoming billing cycles. (MEDALION [par(s) 16] “Embodiments of the present disclosure relate to various systems and methods that may predict a business' category based e.g., on a given business description or an associated vendor . For example, in some types of accounting software, only a portion of the users ( e.g., businesses) supply both a category and a description of the business. Many users supply only one of the two, and some users do not supply either. A method may be used to train a neural network architecture via supervised learning to predict missing categories for businesses based on their given description. The network may be trained using existing businesses that have provided both descriptions and categories.” [par(s) 21] “some embodiments may reduce the amount of data required to be stored by consolidating portions of the data having similar meanings .” [par(s) 32] “ Invoice preparation module 118 may combine words (in vector format or in the vector space) within each line item using a long short-term memory (LSTM) layer. … Invoice preparation module 118 may also utilize a convolutional neural network to combine line item vectors ( or invoice line vectors) associated with the same vendor to create a vector representing that vendor . In some embodiments, the convolutional neural network may be configured to add or subtract the plurality of line item vectors with various weights to create a single vector representing the vendor.” [par(s) 42-45] “An invoice may have a number of line items ( e.g., a list of numbered transactions , each one including a description of the product/service, a price, a vendor, a location, etc.). … Relational data may include percentiles of transaction amounts, frequencies of different range of transaction amount values, etc. In some embodiments, bank transactions may also be obtained by invoice preparation module 110 and blocks 402 and 404 may apply to the bank transactions. In the case where bank transaction vectors are also created, at block 406, the vector representing the vendor may take into account bank transaction vectors along with invoice vectors.” ; ) Regarding claim 5 The combination of MEDALION, BURRELL, VASHISHT teaches claim 1. wherein performing the risk analytics process comprises: (See claim 1) BURRELL further teaches determining the measure of risk of the particular first document based on an expiration time of the particular first document and a remaining amount of the particular first document. (BURRELL [fig(s) 1] [par(s) 107] “In step 640, transaction processing system 100 may group items based on document type within boundaries of transaction. Transaction processing system 100 may group items to balance expenses and payments . For example, a business transaction related to the purchase of products/services may groups items related to expense amount item listed in invoice type document and payment amount item listed in payment check type document . Grouping items may help in understanding the status of a transaction . For example, a business transaction related to the purchase of products may group items related to requested products in the purchase order type document and location tracking items listed in the shipment receipt type document. Transaction processing system 100, upon completion of step 640, completes (step 699) executing method 600 on distributed computing system 300.” [par(s) 61] “In some embodiments, user device 130 may be a scanner (e.g., a flat-bed scanning device, a sheet-fed scanning device, a camera, or the like). User device 130 may scan a collection of documents in an envelope and send the scanned images of the collection of documents as transaction processing request 150. User device 130 may scan multiple envelopes including multiple collections of documents and may send them as transaction processing request 150. User device 130 may scan for a set time period before sending scanned images of collections of documents as transaction processing request 150 . In some embodiments, user device 130 may comprise a scanner or may otherwise be connected to a scanner.” [par(s) 68-69] “Positive indicators” and “Negative indicators” ; Note that MEDALION teaches “risk” as well. ) The combination of MEDALION, BURRELL, VASHISHT is combinable with BURRELL for the same rationale as set forth above with respect to claim 1. Regarding claim 6 The combination of MEDALION, BURRELL, VASHISHT teaches claim 1. wherein evaluating the particular first data dynamically using the reinforcement learning model comprises: (See claim 1) BURRELL further teaches determining a performance of the particular first document based on the measure of risk, information regarding an entity associated with the particular first document, and historical performance of first documents associated with the entity; and (BURRELL [fig(s) 1] [par(s) 17] “Embodiments of the present disclosure are directed to systems and methods configured for learning and predicting the structure of different transactions, different documents involved in processing transactions, identifying different types of documents, and determining statuses of transactions .” [par(s) 7] “The operations may include receive an input of one or more images of documents, analyze, using an image processing engine, the input to determine one or more attributes associated with the one or more images of the documents, identify an account linked to the one or more attributes, determine a transaction associated with the account” [par(s) 107] “Transaction processing system 100 may group items to balance expenses and payments . For example, a business transaction related to the purchase of products/services may groups items related to expense amount item listed in invoice type document and payment amount item listed in payment check type document . Grouping items may help in understanding the status of a transaction .” [par(s) 85] “In some embodiments, relationships between attributes may be learned from previous iterations of transaction processing by transaction processing system 100.” [par(s) 68-69] “Positive indicators”, “In some embodiments, attributes of attributes 122 with confidence scores higher than a threshold value may be considered positive indicators 222. Threshold values of confidence scores may be average, mean, or median of historical values” and “Negative indicators” ; Note that MEDALION teaches “risk” as well. ) The combination of MEDALION, BURRELL, VASHISHT is combinable with BURRELL for the same rationale as set forth above with respect to claim 1. VASHISHT further teaches determining whether actions, identified for the particular first document, improve the performance of the particular first document. (VASHISHT [par(s) 80] “The RL agent 222 is configured to initialize Q-value function and learn the best optimal path for a particular type of payment transaction based on a reward function . The reward function depends upon approval and fraud probability scores of the particular type of transaction and the cost of applying authorizing components to the particular type of transaction. In one embodiment, the approval and fraud probability scores of the particular type of transaction are determined based on historical transaction data (i.e., a number of processed transactions that were approved or declined due to fraud). In one embodiment, the cost of applying an authorizing component is a transaction-level cost and stored at the product cost repository 228.” [par(s) 3] “Payment networks enable various types of payment transactions. A " card-not-present (CNP)" transaction is a type of payment transaction in which a consumer buys a product/service without the presence of a physical payment card (e.g., debit card, credit card, prepaid card ). In such transactions ( e.g., online/e-commerce, card-on-file), the payment card information is transmitted from a merchant, along with a flag that the payment transaction is a CNP transaction.” [par(s) 140] “The various data elements may include, but not limited to, transaction identifier, issuer identifier, issuer name, merchant name/identifier, acquirer name/identifier, cross-border transaction flag (e.g., cross border, domestic), transaction channel flag (e.g., e-commerce, POS, recurring payments), payment card type (e.g., credit, debit), card product type ( customer/commercial), card-not-present (CNP) transaction flag, response code flag (approve/decline), decline reason code (in case of declined transaction), product flag vectors (indicating applied authorizing components by the issuers), etc.” ; e.g., “reward” read(s) on “improve the performance”. Note that MEDALION teaches “PO” as well. ) The combination of MEDALION, BURRELL, VASHISHT is combinable with VASHISHT for the same rationale as set forth above with respect to claim 1. Regarding claim 7 The combination of MEDALION, BURRELL, VASHISHT teaches claim 6. wherein performing the action comprises: (See claim 1) VASHISHT further teaches determining that the actions improve the performance of the particular first document; and (VASHISHT [par(s) 80] “The RL agent 222 is configured to initialize Q-value function and learn the best optimal path for a particular type of payment transaction based on a reward function . The reward function depends upon approval and fraud probability scores of the particular type of transaction and the cost of applying authorizing components to the particular type of transaction. In one embodiment, the approval and fraud probability scores of the particular type of transaction are determined based on historical transaction data (i.e., a number of processed transactions that were approved or declined due to fraud). In one embodiment, the cost of applying an authorizing component is a transaction-level cost and stored at the product cost repository 228.” [par(s) 3] “Payment networks enable various types of payment transactions. A " card-not-present (CNP)" transaction is a type of payment transaction in which a consumer buys a product/service without the presence of a physical payment card (e.g., debit card, credit card, prepaid card ).” [par(s) 80] “The reward function depends upon approval and fraud probability scores of the particular type of transaction and the cost of applying authorizing components to the particular type of transaction.” [par(s) 123] “Where Q(S t , A t ) represents the estimated cumulative reward value obtained by executing the action A t in the state S t ; R t+1 represents the immediate reward value obtained in the next state S t+1 after executing the action A t in the state S t ; max a Q(S r+1 , a) represents the estimated optimal value that is obtained under state S t+1 ; and αε(0,1] represents the influence of estimation error, similar to stochastic gradient descent and finally converges to the optimal Q-value .” ; e.g., “reward” read(s) on “improve the performance”. Note that MEDALION teaches “first document” as well. ) providing the actions as recommendations to improve the performance of the particular first document. (VASHISHT [par(s) 80] “The RL agent 222 is configured to initialize Q-value function and learn the best optimal path for a particular type of payment transaction based on a reward function . The reward function depends upon approval and fraud probability scores of the particular type of transaction and the cost of applying authorizing components to the particular type of transaction. In one embodiment, the approval and fraud probability scores of the particular type of transaction are determined based on historical transaction data (i.e., a number of processed transactions that were approved or declined due to fraud).” [par(s) 44] “other embodiments may include the parts of the environment 100 ( or other parts) arranged otherwise depending on, for example, determining an optimal combination of products needed to be applied on a payment transaction, thereby resulting in high approval rates for the payment transaction , etc.” [par(s) 33] “Various example embodiments of the present dis closure provide methods, systems, user devices and computer program products for enhancing approval rates of payment processing requests by recommending application of one or authorizing components to payment transactions to issuers, in real time.” [par(s) 3] “A " card-not-present (CNP)" transaction is a type of payment transaction in which a consumer buys a product/service without the presence of a physical payment card (e.g., debit card, credit card, prepaid card ).” [par(s) 80] “The reward function depends upon approval and fraud probability scores of the particular type of transaction and the cost of applying authorizing components to the particular type of transaction.” [par(s) 123] “Where Q(S t , A t ) represents the estimated cumulative reward value obtained by executing the action A t in the state S t ; R t+1 represents the immediate reward value obtained in the next state S t+1 after executing the action A t in the state S t ; max a Q(S r+1 , a) represents the estimated optimal value that is obtained under state S t+1 ; and αε(0,1] represents the influence of estimation error, similar to stochastic gradient descent and finally converges to the optimal Q-value .” ; e.g., “reward” read(s) on “improve the performance”. Note that MEDALION teaches “first document” as well. ) The combination of MEDALION, BURRELL, VASHISHT is combinable with VASHISHT for the same rationale as set forth above with respect to claim 1. Regarding claim 8 MEDALION teaches A computer program product comprising: one or more computer readable storage media, and program instructions collectively stored on the one or more computer readable storage media, the program instructions comprising: (MEDALION [fig(s) 10-11]) program instructions to receive purchase order (PO) data regarding one or more purchase orders (POs) and invoice data regarding a plurality of invoices; (MEDALION [par(s) 17-19] “The method may analyze the vendor offerings, identify the associated clusters, and generate a bounded list of purchases for that business. For example, a pre-defined list of purchase orders that a dry-cleaning business may select from within accounting software may include "hanger and accessories" and "pressing accessories." The method of the present disclosure may analyze the dry-cleaning business' purchases and identify associated clusters (e.g., formed by analyzing other dry cleaning services' purchases) with more specialized descriptions such as "plastic bags," "irons," and "press pads."” [par(s) 3] “In addition, it is difficult to summarize the services or products offered by a business, despite databases typically having access to the business' invoices and bank transactions. Obtaining accurate and appropriate lists of business offerings (e.g., services and/or products) requires manual entry or adherence to a pre-defined list when selecting offerings.” [par(s) 32] “Invoice preparation module 118 may be configured to extract text from invoices. In some embodiments, invoice preparation module 118 may be configured to use optical character recognition (OCR) to extract text from invoice files (e.g., PDFs, Word Documents, etc.) or images (e.g., PNG, JPEG, etc.). Invoice files or images may be stored in database 122. In some embodiments, invoice preparation module 118 may be configured to generate a representative vector of a vendor based on a plurality of vectors describing line items of the invoice.” ; ) program instructions to perform a correlation coefficient analysis to identify a subset of labels; (MEDALION [par(s) 33] “Distance metric calculator 120 may be configured to calculate a relation metric between two vectors. In some embodiments, relation metric calculator 120 may be con figured to calculate a relation metric between a merchant vector and a vendor vector. A relation metric may represent a degree of relation between the vectors and thus the corresponding businesses. A low relation metric may correspond to a low relation between two vectors. For example, a relation metric calculated between a merchant vector for a tax service business and a vendor vector for a pool cleaning service may be low or near-zero. A relation metric calculated between a merchant vector for a Jacuzzi supplier and a vendor vector for a pool supplier may be higher, due to the improved commonalities between the merchants and vendors. A relation metric calculated between a merchant vector for a vehicle repair service and a vendor vector for an auto body shop may be higher than the previous two examples. In some embodiments, the relation metrics may be used to train a neural network to predict related/unrelated pairs of vendors and merchants. In some embodiments the relation metric may be an inner product between the two vectors. In some embodiments, relation metric calculator 120 may be configured to apply a sigmoid function to confine the relation metric between zero and one. In some embodiments, relation metric calculator 120 may be configured to deter mine services related to a business.” [par(s) 49] “Cluster analysis module 116 may calculate a TFIDF value for each word within the cluster. In some embodiments, the TFIDF value may be calculated in relation to the entire cluster, as opposed to the TFIDF value being calculated in relation to the line item that contains the word. Cluster analysis module 116 may identify the highest scoring words and combine them to generate a representative description of the cluster.” ; ) program instructions to train a neural network model based on the subset of labels to determine a mapping [between] the plurality of invoices and the one or more POs, (MEDALION [par(s) 16] “ A method may be used to train a neural network architecture via supervised learning to predict missing categories for businesses based on their given description .” [par(s) 33] “Distance metric calculator 120 may be configured to calculate a relation metric between two vectors. In some embodiments, relation metric calculator 120 may be con figured to calculate a relation metric between a merchant vector and a vendor vector. A relation metric may represent a degree of relation between the vectors and thus the corresponding businesses. A low relation metric may correspond to a low relation between two vectors. For example, a relation metric calculated between a merchant vector for a tax service business and a vendor vector for a pool cleaning service may be low or near-zero. A relation metric calculated between a merchant vector for a Jacuzzi supplier and a vendor vector for a pool supplier may be higher, due to the improved commonalities between the merchants and vendors. A relation metric calculated between a merchant vector for a vehicle repair service and a vendor vector for an auto body shop may be higher than the previous two examples. In some embodiments, the relation metrics may be used to train a neural network to predict related/unrelated pairs of vendors and merchants . In some embodiments the relation metric may be an inner product between the two vectors. In some embodiments, relation metric calculator 120 may be configured to apply a sigmoid function to confine the relation metric between zero and one. In some embodiments, relation metric calculator 120 may be configured to deter mine services related to a business.” ; ) wherein one or more invoices, of the plurality of invoices, are associated with a particular PO of the one or more POs; (MEDALION [par(s) 17-19] “The method may analyze the vendor offerings, identify the associated clusters, and generate a bounded list of purchases for that business. For example, a pre-defined list of purchase orders that a dry-cleaning business may select from within accounting software may include "hanger and accessories" and "pressing accessories." The method of the present disclosure may analyze the dry-cleaning business' purchases and identify associated clusters (e.g., formed by analyzing other dry cleaning services' purchases) with more specialized descriptions such as "plastic bags," "irons," and "press pads."” [par(s) 3] “In addition, it is difficult to summarize the services or products offered by a business, despite databases typically having access to the business' invoices and bank transactions. Obtaining accurate and appropriate lists of business offerings (e.g., services and/or products) requires manual entry or adherence to a pre-defined list when selecting offerings.” [par(s) 32] “Invoice preparation module 118 may be configured to extract text from invoices. In some embodiments, invoice preparation module 118 may be configured to use optical character recognition (OCR) to extract text from invoice files (e.g., PDFs, Word Documents, etc.) or images (e.g., PNG, JPEG, etc.). Invoice files or images may be stored in database 122. In some embodiments, invoice preparation module 118 may be configured to generate a representative vector of a vendor based on a plurality of vectors describing line items of the invoice.” ; ) program instructions to generate a time-series forecasting model to predict one or more forecasted invoices for the particular PO; (MEDALION [par(s) 16] “Embodiments of the present disclosure relate to various systems and methods that may predict a business' category based e.g., on a given business description or an associated vendor . For example, in some types of accounting software, only a portion of the users ( e.g., businesses) supply both a category and a description of the business. Many users supply only one of the two, and some users do not supply either. A method may be used to train a neural network architecture via supervised learning to predict missing categories for businesses based on their given description. The network may be trained using existing businesses that have provided both descriptions and categories.” [par(s) 21] “some embodiments may reduce the amount of data required to be stored by consolidating portions of the data having similar meanings .” [par(s) 32] “ Invoice preparation module 118 may combine words (in vector format or in the vector space) within each line item using a long short-term memory (LSTM) layer. … Invoice preparation module 118 may also utilize a convolutional neural network to combine line item vectors ( or invoice line vectors) associated with the same vendor to create a vector representing that vendor . In some embodiments, the convolutional neural network may be configured to add or subtract the plurality of line item vectors with various weights to create a single vector representing the vendor.” ; ) program instructions to perform a risk analytics process on particular PO data, of the particular PO, to determine a measure of risk associated with the particular PO, (MEDALION [par(s) 16] “a method may be used to generate a list of factors associated with a business' success . The factors may be generated by analyzing, aggregating and ranking factors determined to be relevant to a business based on its categorization. The factors associated with the business' success may be related to the products and/or services offered by the business and the format of which those products and/or services are offered by the business . The factors may also be related to the products and/or services purchased by the business from a vendor and the format of which those products and/or services are purchased from the vendor” [par(s) 33] “Distance metric calculator 120 may be configured to calculate a relation metric between two vectors. In some embodiments, relation metric calculator 120 may be con figured to calculate a relation metric between a merchant vector and a vendor vector. A relation metric may represent a degree of relation between the vectors and thus the corresponding businesses. A low relation metric may correspond to a low relation between two vectors. For example, a relation metric calculated between a merchant vector for a tax service business and a vendor vector for a pool cleaning service may be low or near-zero. A relation metric calculated between a merchant vector for a Jacuzzi supplier and a vendor vector for a pool supplier may be higher, due to the improved commonalities between the merchants and vendors. A relation metric calculated between a merchant vector for a vehicle repair service and a vendor vector for an auto body shop may be higher than the previous two examples. In some embodiments, the relation metrics may be used to train a neural network to predict related/unrelated pairs of vendors and merchants . In some embodiments the relation metric may be an inner product between the two vectors. In some embodiments, relation metric calculator 120 may be configured to apply a sigmoid function to confine the relation metric between zero and one. In some embodiments, relation metric calculator 120 may be configured to deter mine services related to a business.” ; ) wherein the measure of risk is determined based on the one or more forecasted invoices; (MEDALION [par(s) 16] “a method may be used to generate a list of factors associated with a business' success . The factors may be generated by analyzing, aggregating and ranking factors determined to be relevant to a business based on its categorization. The factors associated with the business' success may be related to the products and/or services offered by the business and the format of which those products and/or services are offered by the business . The factors may also be related to the products and/or services purchased by the business from a vendor and the format of which those products and/or services are purchased from the vendor” [par(s) 33] “Distance metric calculator 120 may be configured to calculate a relation metric between two vectors. In some embodiments, relation metric calculator 120 may be con figured to calculate a relation metric between a merchant vector and a vendor vector. A relation metric may represent a degree of relation between the vectors and thus the corresponding businesses. A low relation metric may correspond to a low relation between two vectors. For example, a relation metric calculated between a merchant vector for a tax service business and a vendor vector for a pool cleaning service may be low or near-zero. A relation metric calculated between a merchant vector for a Jacuzzi supplier and a vendor vector for a pool supplier may be higher, due to the improved commonalities between the merchants and vendors. A relation metric calculated between a merchant vector for a vehicle repair service and a vendor vector for an auto body shop may be higher than the previous two examples. In some embodiments, the relation metrics may be used to train a neural network to predict related/unrelated pairs of vendors and merchants .” [par(s) 32] “Invoice preparation module 118 may combine words (in vector format or in the vector space) within each line item using a long short-term memory (LSTM) layer. … Invoice preparation module 118 may also utilize a convolutional neural network to combine line item vectors ( or invoice line vectors) associated with the same vendor to create a vector representing that vendor. In some embodiments, the convolutional neural network may be configured to add or subtract the plurality of line item vectors with various weights to create a single vector representing the vendor.” ; ) program instructions to evaluate the particular PO data dynamically using a [reinforcement] learning model, (MEDALION [fig(s) 5] “Embedding”, “LSTM”s, “CNN” [par(s) 26] “Embedding module 110 may be configured to embed text to vector form within a continuous vector space. In some embodiments, embedding module 110 may convert business-related text into a merchant vector within a continuous vector space.” [par(s) 32] “Invoice preparation module 118 may combine words (in vector format or in the vector space) within each line item using a long short-term memory (LSTM) layer. … Invoice preparation module 118 may also utilize a convolutional neural network to combine line item vectors ( or invoice line vectors) associated with the same vendor to create a vector representing that vendor. In some embodiments, the convolutional neural network may be configured to add or subtract the plurality of line item vectors with various weights to create a single vector representing the vendor.” ; ) wherein the particular PO data is evaluated based on the measure of risk; and (MEDALION [fig(s) 5] “Embedding”, “LSTM”s, “CNN” [par(s) 16] “a method may be used to generate a list of factors associated with a business' success . … The factors associated with the business' success may be related to the products and/or services offered by the business and the format of which those products and/or services are offered by the business .” [par(s) 33] “ A relation metric may represent a degree of relation between the vectors and thus the corresponding businesses. A low relation metric may correspond to a low relation between two vectors.” [par(s) 32] “Invoice preparation module 118 may combine words (in vector format or in the vector space) within each line item using a long short-term memory (LSTM) layer. … Invoice preparation module 118 may also utilize a convolutional neural network to combine line item vectors ( or invoice line vectors) associated with the same vendor to create a vector representing that vendor. In some embodiments, the convolutional neural network may be configured to add or subtract the plurality of line item vectors with various weights to create a single vector representing the vendor.” [par(s) 55] “ Back propagation algorithms may include gradient-based learning processes for training multilayer networks .” ; ) program instructions to perform an action based on evaluating the particular PO data. (MEDALION [fig(s) 5] “Embedding”, “LSTM”s, “CNN” [par(s) 22] “FIG. 1 is a block diagram of an example system 100 for aggregating a database of businesses, according to an embodiment of the present disclosure. System 100 may include a plurality of user devices 102a, 100b, ... , 102n (102 generally) and a server device 106, all of which may be communicably coupled via network 104. In some embodiments, system 100 may include any number of user devices. For example, for an organization that manages accounting software and an associated database, there may be an extensive user base with thousands or even millions of users that may connect via respective user devices. Server device 106 may be configured to selectively send a variety of information, such as product or service recommendations or business categorizations, to multiple or single user devices .” ; ) However, MEDALION does not appear to explicitly teach: program instructions to train a neural network model based on the subset of labels to determine a mapping [between] the plurality of invoices and the one or more POs, program instructions to evaluate the particular PO data dynamically using a [reinforcement] learning model, BURRELL teaches program instructions to train a neural network model based on the subset of labels to determine a mapping between the plurality of invoices and the one or more POs, (BURRELL [fig(s) 1] [par(s) 52] “Identities 123 may include mappings between different attributes . For example, a purchase order document may include the purchase order number stored as an attribute in attributes 122 that is mapped to an invoice number in an invoice document to form a sale identity . Character recognition module 111 may determine identities 123 based on matching attributes between different documents. For example, purchase order and invoice documents described above can list the same set of products/services requested and provided, resulting in the linking of the purchase order and the invoice attributes and create a sale identity . Identities 123 obtained from mapping attributes of attributes 122 may include combined values of attributes.” [par(s) 85] “In step 430, transaction processing system 100 may group attributes into sets of attributes. Transaction processing system 100 may group attributes based on relationships (e.g., correlations 224 of FIG. 2) defined in transaction processing system 100. In some embodiments, relationships between attributes may be learned from previous iterations of transaction processing by transaction processing system 100. Transaction processing system 100 may include machine learning models that may be trained to learn different types of documents, their structure, and the relationships between various attributes that may be part of documents. Machine learning models used in transaction processing system 100 may include neural networks . In some embodiments, transaction processing system 100 components may determine document types using fuzzy logic systems.” See also [par(s) 68-72] ; ) The combination of MEDALION, BURRELL, VASHISHT is combinable with BURRELL for the same rationale as set forth above with respect to claim 1. However, the combination of MEDALION, BURRELL does not appear to explicitly teach: program instructions to evaluate the particular PO data dynamically using a [reinforcement] learning model, VASHISHT teaches program instructions to evaluate the particular PO data dynamically using a reinforcement learning model, (VASHISHT [par(s) 80] “The RL agent 222 is configured to initialize Q-value function and learn the best optimal path for a particular type of payment transaction based on a reward function . The reward function depends upon approval and fraud probability scores of the particular type of transaction and the cost of applying authorizing components to the particular type of transaction. In one embodiment, the approval and fraud probability scores of the particular type of transaction are determined based on historical transaction data (i.e., a number of processed transactions that were approved or declined due to fraud). In one embodiment, the cost of applying an authorizing component is a transaction-level cost and stored at the product cost repository 228.” [par(s) 3] “Payment networks enable various types of payment transactions. A " card-not-present (CNP)" transaction is a type of payment transaction in which a consumer buys a product/service without the presence of a physical payment card (e.g., debit card, credit card, prepaid card ). In such transactions ( e.g., online/e-commerce, card-on-file), the payment card information is transmitted from a merchant, along with a flag that the payment transaction is a CNP transaction.” [par(s) 140] “The various data elements may include, but not limited to, transaction identifier, issuer identifier, issuer name, merchant name/identifier, acquirer name/identifier, cross-border transaction flag (e.g., cross border, domestic), transaction channel flag (e.g., e-commerce, POS, recurring payments), payment card type (e.g., credit, debit), card product type ( customer/commercial), card-not-present (CNP) transaction flag, response code flag (approve/decline), decline reason code (in case of declined transaction), product flag vectors (indicating applied authorizing components by the issuers), etc.” ; ) The combination of MEDALION, BURRELL, VASHISHT is combinable with VASHISHT for the same rationale as set forth above with respect to claim 1. Regarding claim 9 The combination of MEDALION, BURRELL, VASHISHT teaches claim 8. BURRELL further teaches program instructions to analyze the PO data to identify one or more first labels with missing values; (BURRELL [fig(s) 1] [par(s) 19] “determining document types and transaction structure. Mapping documents to various parts of a transaction can help determine missing parts of a transaction that help determine the current status of a transaction.” [par(s) 52] “Identities 123 may include mappings between different attributes . For example, a purchase order document may include the purchase order number stored as an attribute in attributes 122 that is mapped to an invoice number in an invoice document to form a sale identity . Character recognition module 111 may determine identities 123 based on matching attributes between different documents. For example, purchase order and invoice documents described above can list the same set of products/services requested and provided, resulting in the linking of the purchase order and the invoice attributes and create a sale identity . Identities 123 obtained from mapping attributes of attributes 122 may include combined values of attributes.” [par(s) 85] “In step 430, transaction processing system 100 may group attributes into sets of attributes. Transaction processing system 100 may group attributes based on relationships (e.g., correlations 224 of FIG. 2) defined in transaction processing system 100. In some embodiments, relationships between attributes may be learned from previous iterations of transaction processing by transaction processing system 100. ” ; Note that MEDALION teaches invoices and POs as well. ) program instructions to analyze the invoice data to identify one or more second labels with missing values; (BURRELL [fig(s) 1] [par(s) 19] “determining document types and transaction structure. Mapping documents to various parts of a transaction can help determine missing parts of a transaction that help determine the current status of a transaction.” [par(s) 101] “In the above example transaction of purchase and payment, a payment amount covering a subset of items listed in invoice type document can result in a transaction boundary that includes only those items in purchase order and invoice documents with the remaining attributes associated with unpaid items included in a second incomplete transaction . Transaction processing system 100, upon completion of step 550, completes (step 599) executing method 500 on distributed com puting system 300” [par(s) 52] “Identities 123 may include mappings between different attributes . For example, a purchase order document may include the purchase order number stored as an attribute in attributes 122 that is mapped to an invoice number in an invoice document to form a sale identity . Character recognition module 111 may determine identities 123 based on matching attributes between different documents. For example, purchase order and invoice documents described above can list the same set of products/services requested and provided, resulting in the linking of the purchase order and the invoice attributes and create a sale identity . Identities 123 obtained from mapping attributes of attributes 122 may include combined values of attributes.” [par(s) 85] “In step 430, transaction processing system 100 may group attributes into sets of attributes. Transaction processing system 100 may group attributes based on relationships (e.g., correlations 224 of FIG. 2) defined in transaction processing system 100. In some embodiments, relationships between attributes may be learned from previous iterations of transaction processing by transaction processing system 100. ” ; Note that MEDALION teaches invoices and POs as well. ) program instructions to analyze the PO data, using a natural language processing algorithm, to identify one or more first values for the one or more first labels; and (BURRELL [fig(s) 1] [par(s) 3] “Automated transaction processing to identify the beginning and end of a business transaction from paper-based documents is a challenging task requiring different technologies, such as image processing, character recognition, and natural language processing .” [par(s) 29-32] “Character recognition module 111 may use extracted text elements to determine attributes present in the accessed image of document. An attribute may be considered a classification of data. For example, attributes may be names, contact information, etc., of individuals involved in a transaction. Character recognition module 111 may determine attributes in a document of document images 121 and stores them as attributes 122 in data store 120. Character recognition module 111 may determine attributes from extracted text elements based on document type information. … Character recognition module 111 may identify and extract text elements and share with confidence scoring module 112 to generate confidence scores that a particular text element represents a particular attribute .” [par(s) 52] “Identities 123 may include mappings between different attributes . For example, a purchase order document may include the purchase order number stored as an attribute in attributes 122 that is mapped to an invoice number in an invoice document to form a sale identity . Character recognition module 111 may determine identities 123 based on matching attributes between different documents. For example, purchase order and invoice documents described above can list the same set of products/services requested and provided, resulting in the linking of the purchase order and the invoice attributes and create a sale identity . Identities 123 obtained from mapping attributes of attributes 122 may include combined values of attributes.” ; Note that MEDALION teaches invoices and POs as well. ) program instructions to analyze the invoice data, using the natural language processing algorithm, to identify one or more second values for the one or more second labels. (BURRELL [fig(s) 1] [par(s) 3] “Automated transaction processing to identify the beginning and end of a business transaction from paper-based documents is a challenging task requiring different technologies, such as image processing, character recognition, and natural language processing .” [par(s) 29-32] “Character recognition module 111 may use extracted text elements to determine attributes present in the accessed image of document. An attribute may be considered a classification of data. For example, attributes may be names, contact information, etc., of individuals involved in a transaction. Character recognition module 111 may determine attributes in a document of document images 121 and stores them as attributes 122 in data store 120. Character recognition module 111 may determine attributes from extracted text elements based on document type information. … Character recognition module 111 may identify and extract text elements and share with confidence scoring module 112 to generate confidence scores that a particular text element represents a particular attribute .” [par(s) 52] “Identities 123 may include mappings between different attributes . For example, a purchase order document may include the purchase order number stored as an attribute in attributes 122 that is mapped to an invoice number in an invoice document to form a sale identity . Character recognition module 111 may determine identities 123 based on matching attributes between different documents. For example, purchase order and invoice documents described above can list the same set of products/services requested and provided, resulting in the linking of the purchase order and the invoice attributes and create a sale identity . Identities 123 obtained from mapping attributes of attributes 122 may include combined values of attributes.” ; Note that MEDALION teaches invoices and POs as well. ) The combination of MEDALION, BURRELL, VASHISHT is combinable with BURRELL for the same rationale as set forth above with respect to claim 1. Regarding claim 10 The combination of MEDALION, BURRELL, VASHISHT teaches claim 8. BURRELL further teaches program instructions to analyze the PO data and the invoice data to determine matches between PO labels of the particular PO and invoice labels of one or more invoices of the plurality of invoices; (BURRELL [fig(s) 1] [par(s) 52] “Identities 123 may include mappings between different attributes . For example, a purchase order document may include the purchase order number stored as an attribute in attributes 122 that is mapped to an invoice number in an invoice document to form a sale identity . Character recognition module 111 may determine identities 123 based on matching attributes between different documents. For example, purchase order and invoice documents described above can list the same set of products/services requested and provided, resulting in the linking of the purchase order and the invoice attributes and create a sale identity . Identities 123 obtained from mapping attributes of attributes 122 may include combined values of attributes.” [par(s) 85] “In step 430, transaction processing system 100 may group attributes into sets of attributes. Transaction processing system 100 may group attributes based on relationships (e.g., correlations 224 of FIG. 2) defined in transaction processing system 100. In some embodiments, relationships between attributes may be learned from previous iterations of transaction processing by transaction processing system 100. Transaction processing system 100 may include machine learning models that may be trained to learn different types of documents, their structure, and the relationships between various attributes that may be part of documents . Machine learning models used in transaction processing system 100 may include neural networks . In some embodiments, transaction processing system 100 components may determine document types using fuzzy logic systems.” See also [par(s) 70-72] ; ) program instructions to assign values based on analyzing the PO data and the invoice data to determine the matches, wherein a first value is assigned to indicate a match between a first PO label of the particular PO and a first invoice label of an invoice, and wherein a second value is assigned to indicate a match between a second PO label of the particular PO and a second invoice label of the invoice; and (BURRELL [fig(s) 1] [par(s) 52] “Identities 123 may include mappings between different attributes . For example, a purchase order document may include the purchase order number stored as an attribute in attributes 122 that is mapped to an invoice number in an invoice document to form a sale identity . Character recognition module 111 may determine identities 123 based on matching attributes between different documents. For example, purchase order and invoice documents described above can list the same set of products/services requested and provided, resulting in the linking of the purchase order and the invoice attributes and create a sale identity . Identities 123 obtained from mapping attributes of attributes 122 may include combined values of attributes.” [par(s) 85] “In step 430, transaction processing system 100 may group attributes into sets of attributes. Transaction processing system 100 may group attributes based on relationships (e.g., correlations 224 of FIG. 2) defined in transaction processing system 100. In some embodiments, relationships between attributes may be learned from previous iterations of transaction processing by transaction processing system 100. Transaction processing system 100 may include machine learning models that may be trained to learn different types of documents, their structure, and the relationships between various attributes that may be part of documents . Machine learning models used in transaction processing system 100 may include neural networks . In some embodiments, transaction processing system 100 components may determine document types using fuzzy logic systems.” [par(s) 68-69] “Positive indicators” and “Negative indicators” See also [par(s) 70-72] ; ) program instructions to generate a data structure that includes multiple entries with the first value and the second value. (BURRELL [fig(s) 1-3] [par(s) 52] “Identities 123 may include mappings between different attributes . For example, a purchase order document may include the purchase order number stored as an attribute in attributes 122 that is mapped to an invoice number in an invoice document to form a sale identity . Character recognition module 111 may determine identities 123 based on matching attributes between different documents. For example, purchase order and invoice documents described above can list the same set of products/services requested and provided, resulting in the linking of the purchase order and the invoice attributes and create a sale identity . Identities 123 obtained from mapping attributes of attributes 122 may include combined values of attributes.” [par(s) 85] “In step 430, transaction processing system 100 may group attributes into sets of attributes. Transaction processing system 100 may group attributes based on relationships (e.g., correlations 224 of FIG. 2) defined in transaction processing system 100. In some embodiments, relationships between attributes may be learned from previous iterations of transaction processing by transaction processing system 100. ” [par(s) 68-69] “Positive indicators” and “Negative indicators” See also [par(s) 70-72] [par(s) 73] “Image processing engine 110 may generate various different outputs 230 related to a transaction including document types 231 of input document images 121, account assignments 232 associated with documents represented by document images 121, and transaction structures 233 forming transactions (e.g., transactions 124) using contents of document images 121. Various outputs 230 may be linked together.” ; ) The combination of MEDALION, BURRELL, VASHISHT is combinable with BURRELL for the same rationale as set forth above with respect to claim 1. Regarding claim 11 The combination of MEDALION, BURRELL, VASHISHT teaches claim 10. wherein the program instructions to perform the correlation coefficient analysis comprise: (See claim 8) MEDALION further teaches program instructions to perform the correlation coefficient analysis to identify, as a subset of most correlated labels, a subset of the PO labels and a subset of the [invoice] labels that are most correlated out of the PO labels and the [invoice] labels; and (MEDALION [par(s) 18] “According to another embodiment of the present disclosure, a method may be used to generate a list of factors associated with a business' success. The factors may be generated by analyzing, aggregating and ranking factors determined to be relevant to a business based on its categorization . The factors associated with the business' success may be related to the products and/or services offered by the business and the format of which those products and/or services are offered by the business. The factors may also be related to the products and/or services purchased by the business from a vendor and the format of which those products and/or services are purchased from the vendor.” [par(s) 17-19] “For example, a pre-defined list of purchase orders that a dry-cleaning business may select from within accounting software may include "hanger and accessories" and "pressing accessories."” [par(s) 3] “In addition, it is difficult to summarize the services or products offered by a business, despite databases typically having access to the business' invoices and bank transactions.” [par(s) 33] “ A relation metric may represent a degree of relation between the vectors and thus the corresponding businesses. A low relation metric may correspond to a low relation between two vectors. For example, a relation metric calculated between a merchant vector for a tax service business and a vendor vector for a pool cleaning service may be low or near-zero. A relation metric calculated between a merchant vector for a Jacuzzi supplier and a vendor vector for a pool supplier may be higher, due to the improved commonalities between the merchants and vendors. A relation metric calculated between a merchant vector for a vehicle repair service and a vendor vector for an auto body shop may be higher than the previous two examples. In some embodiments, the relation metrics may be used to train a neural network to predict related/unrelated pairs of vendors and merchants.” ; ) wherein the neural network model is trained using the identified subset of most correlated labels. (MEDALION [par(s) 18] “According to another embodiment of the present disclosure, a method may be used to generate a list of factors associated with a business' success. The factors may be generated by analyzing, aggregating and ranking factors determined to be relevant to a business based on its categorization . The factors associated with the business' success may be related to the products and/or services offered by the business and the format of which those products and/or services are offered by the business. The factors may also be related to the products and/or services purchased by the business from a vendor and the format of which those products and/or services are purchased from the vendor.” [par(s) 33] “ A relation metric may represent a degree of relation between the vectors and thus the corresponding businesses. A low relation metric may correspond to a low relation between two vectors. For example, a relation metric calculated between a merchant vector for a tax service business and a vendor vector for a pool cleaning service may be low or near-zero. A relation metric calculated between a merchant vector for a Jacuzzi supplier and a vendor vector for a pool supplier may be higher, due to the improved commonalities between the merchants and vendors. A relation metric calculated between a merchant vector for a vehicle repair service and a vendor vector for an auto body shop may be higher than the previous two examples. In some embodiments, the relation metrics may be used to train a neural network to predict related/unrelated pairs of vendors and merchants.” ; ) BURRELL further teaches program instructions to perform the correlation coefficient analysis to identify, as a subset of most correlated labels, a subset of the PO labels and a subset of the invoice labels that are most correlated out of the PO labels and the invoice labels; and (BURRELL [fig(s) 1] [par(s) 52] “Identities 123 may include mappings between different attributes . For example, a purchase order document may include the purchase order number stored as an attribute in attributes 122 that is mapped to an invoice number in an invoice document to form a sale identity . Character recognition module 111 may determine identities 123 based on matching attributes between different documents. For example, purchase order and invoice documents described above can list the same set of products/services requested and provided, resulting in the linking of the purchase order and the invoice attributes and create a sale identity . Identities 123 obtained from mapping attributes of attributes 122 may include combined values of attributes.” [par(s) 85] “In step 430, transaction processing system 100 may group attributes into sets of attributes. Transaction processing system 100 may group attributes based on relationships (e.g., correlations 224 of FIG. 2) defined in transaction processing system 100. In some embodiments, relationships between attributes may be learned from previous iterations of transaction processing by transaction processing system 100. Transaction processing system 100 may include machine learning models that may be trained to learn different types of documents, their structure, and the relationships between various attributes that may be part of documents. Machine learning models used in transaction processing system 100 may include neural networks . In some embodiments, transaction processing system 100 components may determine document types using fuzzy logic systems.” [par(s) 67-73] “ Rankings 225 may include ordering of different attributes associated with a text element extracted from a document requested for processing using transaction processing request 150 . Rankings 225 may also include ordering different attributes associated with different text elements and different documents. Rankings 225 may be based on confidence scores assigned to attributes by confidence scoring module 112” ; ) program instructions to train a neural network model to determine matches between POs and invoices, (BURRELL [fig(s) 1] [par(s) 52] “Identities 123 may include mappings between different attributes . For example, a purchase order document may include the purchase order number stored as an attribute in attributes 122 that is mapped to an invoice number in an invoice document to form a sale identity . Character recognition module 111 may determine identities 123 based on matching attributes between different documents. For example, purchase order and invoice documents described above can list the same set of products/services requested and provided, resulting in the linking of the purchase order and the invoice attributes and create a sale identity . Identities 123 obtained from mapping attributes of attributes 122 may include combined values of attributes.” [par(s) 85] “In step 430, transaction processing system 100 may group attributes into sets of attributes. Transaction processing system 100 may group attributes based on relationships (e.g., correlations 224 of FIG. 2) defined in transaction processing system 100. In some embodiments, relationships between attributes may be learned from previous iterations of transaction processing by transaction processing system 100. Transaction processing system 100 may include machine learning models that may be trained to learn different types of documents, their structure, and the relationships between various attributes that may be part of documents. Machine learning models used in transaction processing system 100 may include neural networks . In some embodiments, transaction processing system 100 components may determine document types using fuzzy logic systems.” See also [par(s) 70-72] ; ) The combination of MEDALION, BURRELL, VASHISHT is combinable with BURRELL for the same rationale as set forth above with respect to claim 1. Regarding claim 12 The combination of MEDALION, BURRELL, VASHISHT teaches claim 8. wherein the program instructions to perform the risk analytics process comprise: (See claim 8) program instructions to determine the measure of risk of the particular PO based on an expiration time of the particular PO and a remaining amount of the particular PO. (BURRELL [fig(s) 1] [par(s) 107] “In step 640, transaction processing system 100 may group items based on document type within boundaries of transaction. Transaction processing system 100 may group items to balance expenses and payments . For example, a business transaction related to the purchase of products/services may groups items related to expense amount item listed in invoice type document and payment amount item listed in payment check type document . Grouping items may help in understanding the status of a transaction . For example, a business transaction related to the purchase of products may group items related to requested products in the purchase order type document and location tracking items listed in the shipment receipt type document. Transaction processing system 100, upon completion of step 640, completes (step 699) executing method 600 on distributed computing system 300.” [par(s) 61] “In some embodiments, user device 130 may be a scanner (e.g., a flat-bed scanning device, a sheet-fed scanning device, a camera, or the like). User device 130 may scan a collection of documents in an envelope and send the scanned images of the collection of documents as transaction processing request 150. User device 130 may scan multiple envelopes including multiple collections of documents and may send them as transaction processing request 150. User device 130 may scan for a set time period before sending scanned images of collections of documents as transaction processing request 150 . In some embodiments, user device 130 may comprise a scanner or may otherwise be connected to a scanner.” [par(s) 68-69] “Positive indicators” and “Negative indicators” ; Note that MEDALION teaches “risk” as well. ) The combination of MEDALION, BURRELL, VASHISHT is combinable with BURRELL for the same rationale as set forth above with respect to claim 1. Regarding claim 13 The combination of MEDALION, BURRELL, VASHISHT teaches claim 8. wherein the program instructions to evaluate the particular PO data dynamically using the reinforcement learning model comprise: (See claim 8) BURRELL further teaches program instructions to determine a performance of the particular PO based on the measure of risk, information regarding an entity associated with the particular PO, and historical performance of POs associated with the entity; and (BURRELL [fig(s) 1] [par(s) 17] “Embodiments of the present disclosure are directed to systems and methods configured for learning and predicting the structure of different transactions, different documents involved in processing transactions, identifying different types of documents, and determining statuses of transactions .” [par(s) 7] “The operations may include receive an input of one or more images of documents, analyze, using an image processing engine, the input to determine one or more attributes associated with the one or more images of the documents, identify an account linked to the one or more attributes, determine a transaction associated with the account” [par(s) 107] “Transaction processing system 100 may group items to balance expenses and payments . For example, a business transaction related to the purchase of products/services may groups items related to expense amount item listed in invoice type document and payment amount item listed in payment check type document . Grouping items may help in understanding the status of a transaction .” [par(s) 85] “In some embodiments, relationships between attributes may be learned from previous iterations of transaction processing by transaction processing system 100.” [par(s) 68-69] “Positive indicators”, “In some embodiments, attributes of attributes 122 with confidence scores higher than a threshold value may be considered positive indicators 222. Threshold values of confidence scores may be average, mean, or median of historical values” and “Negative indicators” ; Note that MEDALION teaches “risk” as well. ) The combination of MEDALION, BURRELL, VASHISHT is combinable with BURRELL for the same rationale as set forth above with respect to claim 1. VASHISHT further teaches program instructions to determine whether actions, identified for the particular PO, improve the performance of the particular PO. (VASHISHT [par(s) 80] “The RL agent 222 is configured to initialize Q-value function and learn the best optimal path for a particular type of payment transaction based on a reward function . The reward function depends upon approval and fraud probability scores of the particular type of transaction and the cost of applying authorizing components to the particular type of transaction. In one embodiment, the approval and fraud probability scores of the particular type of transaction are determined based on historical transaction data (i.e., a number of processed transactions that were approved or declined due to fraud). In one embodiment, the cost of applying an authorizing component is a transaction-level cost and stored at the product cost repository 228.” [par(s) 3] “Payment networks enable various types of payment transactions. A " card-not-present (CNP)" transaction is a type of payment transaction in which a consumer buys a product/service without the presence of a physical payment card (e.g., debit card, credit card, prepaid card ). In such transactions ( e.g., online/e-commerce, card-on-file), the payment card information is transmitted from a merchant, along with a flag that the payment transaction is a CNP transaction.” [par(s) 140] “The various data elements may include, but not limited to, transaction identifier, issuer identifier, issuer name, merchant name/identifier, acquirer name/identifier, cross-border transaction flag (e.g., cross border, domestic), transaction channel flag (e.g., e-commerce, POS, recurring payments), payment card type (e.g., credit, debit), card product type ( customer/commercial), card-not-present (CNP) transaction flag, response code flag (approve/decline), decline reason code (in case of declined transaction), product flag vectors (indicating applied authorizing components by the issuers), etc.” ; e.g., “reward” read(s) on “improve the performance”. Note that MEDALION teaches “PO” as well. ) The combination of MEDALION, BURRELL, VASHISHT is combinable with VASHISHT for the same rationale as set forth above with respect to claim 1. Regarding claim 14 The combination of MEDALION, BURRELL, VASHISHT teaches claim 13. VASHISHT further teaches program instructions to determine that the actions improve the performance of the particular PO; and (VASHISHT [par(s) 80] “The RL agent 222 is configured to initialize Q-value function and learn the best optimal path for a particular type of payment transaction based on a reward function . The reward function depends upon approval and fraud probability scores of the particular type of transaction and the cost of applying authorizing components to the particular type of transaction. In one embodiment, the approval and fraud probability scores of the particular type of transaction are determined based on historical transaction data (i.e., a number of processed transactions that were approved or declined due to fraud). In one embodiment, the cost of applying an authorizing component is a transaction-level cost and stored at the product cost repository 228.” [par(s) 3] “Payment networks enable various types of payment transactions. A " card-not-present (CNP)" transaction is a type of payment transaction in which a consumer buys a product/service without the presence of a physical payment card (e.g., debit card, credit card, prepaid card ).” [par(s) 80] “The reward function depends upon approval and fraud probability scores of the particular type of transaction and the cost of applying authorizing components to the particular type of transaction.” [par(s) 123] “Where Q(S t , A t ) represents the estimated cumulative reward value obtained by executing the action A t in the state S t ; R t+1 represents the immediate reward value obtained in the next state S t+1 after executing the action A t in the state S t ; max a Q(S r+1 , a) represents the estimated optimal value that is obtained under state S t+1 ; and αε(0,1] represents the influence of estimation error, similar to stochastic gradient descent and finally converges to the optimal Q-value .” ; e.g., “reward” read(s) on “improve the performance”. Note that MEDALION teaches “PO” as well. ) program instructions to provide the actions as recommendations to improve the performance of the particular PO. (VASHISHT [par(s) 80] “The RL agent 222 is configured to initialize Q-value function and learn the best optimal path for a particular type of payment transaction based on a reward function . The reward function depends upon approval and fraud probability scores of the particular type of transaction and the cost of applying authorizing components to the particular type of transaction. In one embodiment, the approval and fraud probability scores of the particular type of transaction are determined based on historical transaction data (i.e., a number of processed transactions that were approved or declined due to fraud).” [par(s) 44] “other embodiments may include the parts of the environment 100 ( or other parts) arranged otherwise depending on, for example, determining an optimal combination of products needed to be applied on a payment transaction, thereby resulting in high approval rates for the payment transaction , etc.” [par(s) 33] “Various example embodiments of the present dis closure provide methods, systems, user devices and computer program products for enhancing approval rates of payment processing requests by recommending application of one or authorizing components to payment transactions to issuers, in real time.” [par(s) 3] “A " card-not-present (CNP)" transaction is a type of payment transaction in which a consumer buys a product/service without the presence of a physical payment card (e.g., debit card, credit card, prepaid card ).” [par(s) 80] “The reward function depends upon approval and fraud probability scores of the particular type of transaction and the cost of applying authorizing components to the particular type of transaction.” [par(s) 123] “Where Q(S t , A t ) represents the estimated cumulative reward value obtained by executing the action A t in the state S t ; R t+1 represents the immediate reward value obtained in the next state S t+1 after executing the action A t in the state S t ; max a Q(S r+1 , a) represents the estimated optimal value that is obtained under state S t+1 ; and αε(0,1] represents the influence of estimation error, similar to stochastic gradient descent and finally converges to the optimal Q-value .” ; e.g., “reward” read(s) on “improve the performance”. Note that MEDALION teaches “PO” as well. ) The combination of MEDALION, BURRELL, VASHISHT is combinable with VASHISHT for the same rationale as set forth above with respect to claim 1. Regarding claim 15 MEDALION teaches A system comprising: one or more devices configured to: (MEDALION [fig(s) 10-11]) perform a correlation coefficient analysis to identify a subset of most correlated labels out of labels of a plurality of [invoices] and labels of a plurality of purchase orders (POs); (MEDALION [par(s) 18] “According to another embodiment of the present disclosure, a method may be used to generate a list of factors associated with a business' success. The factors may be generated by analyzing, aggregating and ranking factors determined to be relevant to a business based on its categorization . The factors associated with the business' success may be related to the products and/or services offered by the business and the format of which those products and/or services are offered by the business. The factors may also be related to the products and/or services purchased by the business from a vendor and the format of which those products and/or services are purchased from the vendor.” [par(s) 17-19] “For example, a pre-defined list of purchase orders that a dry-cleaning business may select from within accounting software may include "hanger and accessories" and "pressing accessories."” [par(s) 3] “In addition, it is difficult to summarize the services or products offered by a business, despite databases typically having access to the business' invoices and bank transactions.” [par(s) 33] “ A relation metric may represent a degree of relation between the vectors and thus the corresponding businesses. A low relation metric may correspond to a low relation between two vectors. For example, a relation metric calculated between a merchant vector for a tax service business and a vendor vector for a pool cleaning service may be low or near-zero. A relation metric calculated between a merchant vector for a Jacuzzi supplier and a vendor vector for a pool supplier may be higher, due to the improved commonalities between the merchants and vendors. A relation metric calculated between a merchant vector for a vehicle repair service and a vendor vector for an auto body shop may be higher than the previous two examples. In some embodiments, the relation metrics may be used to train a neural network to predict related/unrelated pairs of vendors and merchants.” ; ) train a neural network model, based on the subset of most correlated labels, to determine a mapping [between] the plurality of invoices and the plurality of POs, (MEDALION [par(s) 18] “According to another embodiment of the present disclosure, a method may be used to generate a list of factors associated with a business' success. The factors may be generated by analyzing, aggregating and ranking factors determined to be relevant to a business based on its categorization . The factors associated with the business' success may be related to the products and/or services offered by the business and the format of which those products and/or services are offered by the business. The factors may also be related to the products and/or services purchased by the business from a vendor and the format of which those products and/or services are purchased from the vendor.” [par(s) 33] “ A relation metric may represent a degree of relation between the vectors and thus the corresponding businesses. A low relation metric may correspond to a low relation between two vectors. For example, a relation metric calculated between a merchant vector for a tax service business and a vendor vector for a pool cleaning service may be low or near-zero. A relation metric calculated between a merchant vector for a Jacuzzi supplier and a vendor vector for a pool supplier may be higher, due to the improved commonalities between the merchants and vendors. A relation metric calculated between a merchant vector for a vehicle repair service and a vendor vector for an auto body shop may be higher than the previous two examples. In some embodiments, the relation metrics may be used to train a neural network to predict related/unrelated pairs of vendors and merchants.” ; ) train a time-series forecasting model to predict one or more forecasted invoices for the particular PO; (MEDALION [par(s) 16] “Embodiments of the present disclosure relate to various systems and methods that may predict a business' category based e.g., on a given business description or an associated vendor . For example, in some types of accounting software, only a portion of the users ( e.g., businesses) supply both a category and a description of the business. Many users supply only one of the two, and some users do not supply either. A method may be used to train a neural network architecture via supervised learning to predict missing categories for businesses based on their given description. The network may be trained using existing businesses that have provided both descriptions and categories.” [par(s) 21] “some embodiments may reduce the amount of data required to be stored by consolidating portions of the data having similar meanings .” [par(s) 32] “ Invoice preparation module 118 may combine words (in vector format or in the vector space) within each line item using a long short-term memory (LSTM) layer. … Invoice preparation module 118 may also utilize a convolutional neural network to combine line item vectors ( or invoice line vectors) associated with the same vendor to create a vector representing that vendor . In some embodiments, the convolutional neural network may be configured to add or subtract the plurality of line item vectors with various weights to create a single vector representing the vendor.” ; ) determine a measure of risk associated with the particular PO, (MEDALION [par(s) 16] “a method may be used to generate a list of factors associated with a business' success . The factors may be generated by analyzing, aggregating and ranking factors determined to be relevant to a business based on its categorization. The factors associated with the business' success may be related to the products and/or services offered by the business and the format of which those products and/or services are offered by the business . The factors may also be related to the products and/or services purchased by the business from a vendor and the format of which those products and/or services are purchased from the vendor” [par(s) 33] “Distance metric calculator 120 may be configured to calculate a relation metric between two vectors. In some embodiments, relation metric calculator 120 may be con figured to calculate a relation metric between a merchant vector and a vendor vector. A relation metric may represent a degree of relation between the vectors and thus the corresponding businesses. A low relation metric may correspond to a low relation between two vectors. … In some embodiments, the relation metrics may be used to train a neural network to predict related/unrelated pairs of vendors and merchants . In some embodiments the relation metric may be an inner product between the two vectors. In some embodiments, relation metric calculator 120 may be configured to apply a sigmoid function to confine the relation metric between zero and one. In some embodiments, relation metric calculator 120 may be configured to deter mine services related to a business.” ; ) wherein the measure of risk is determined based on the one or more forecasted invoices; (MEDALION [par(s) 16] “a method may be used to generate a list of factors associated with a business' success . The factors may be generated by analyzing, aggregating and ranking factors determined to be relevant to a business based on its categorization. The factors associated with the business' success may be related to the products and/or services offered by the business and the format of which those products and/or services are offered by the business . The factors may also be related to the products and/or services purchased by the business from a vendor and the format of which those products and/or services are purchased from the vendor” [par(s) 33] “Distance metric calculator 120 may be configured to calculate a relation metric between two vectors. In some embodiments, relation metric calculator 120 may be con figured to calculate a relation metric between a merchant vector and a vendor vector. A relation metric may represent a degree of relation between the vectors and thus the corresponding businesses. A low relation metric may correspond to a low relation between two vectors. … In some embodiments, the relation metrics may be used to train a neural network to predict related/unrelated pairs of vendors and merchants .” [par(s) 32] “Invoice preparation module 118 may combine words (in vector format or in the vector space) within each line item using a long short-term memory (LSTM) layer. … Invoice preparation module 118 may also utilize a convolutional neural network to combine line item vectors ( or invoice line vectors) associated with the same vendor to create a vector representing that vendor. In some embodiments, the convolutional neural network may be configured to add or subtract the plurality of line item vectors with various weights to create a single vector representing the vendor.” ; ) evaluate particular PO data, of the particular PO, dynamically using a [reinforcement] learning model, (MEDALION [fig(s) 5] “Embedding”, “LSTM”s, “CNN” [par(s) 26] “Embedding module 110 may be configured to embed text to vector form within a continuous vector space. In some embodiments, embedding module 110 may convert business-related text into a merchant vector within a continuous vector space.” [par(s) 32] “Invoice preparation module 118 may combine words (in vector format or in the vector space) within each line item using a long short-term memory (LSTM) layer. … Invoice preparation module 118 may also utilize a convolutional neural network to combine line item vectors ( or invoice line vectors) associated with the same vendor to create a vector representing that vendor. In some embodiments, the convolutional neural network may be configured to add or subtract the plurality of line item vectors with various weights to create a single vector representing the vendor.” ; ) wherein the particular PO data is evaluated based on the measure of risk; and (MEDALION [fig(s) 5] “Embedding”, “LSTM”s, “CNN” [par(s) 16] “a method may be used to generate a list of factors associated with a business' success . … The factors associated with the business' success may be related to the products and/or services offered by the business and the format of which those products and/or services are offered by the business .” [par(s) 33] “ A relation metric may represent a degree of relation between the vectors and thus the corresponding businesses. A low relation metric may correspond to a low relation between two vectors.” [par(s) 32] “Invoice preparation module 118 may combine words (in vector format or in the vector space) within each line item using a long short-term memory (LSTM) layer. … Invoice preparation module 118 may also utilize a convolutional neural network to combine line item vectors ( or invoice line vectors) associated with the same vendor to create a vector representing that vendor. In some embodiments, the convolutional neural network may be configured to add or subtract the plurality of line item vectors with various weights to create a single vector representing the vendor.” [par(s) 55] “ Back propagation algorithms may include gradient-based learning processes for training multilayer networks .” ; ) perform one or more recommended actions based on evaluating the particular PO data. (MEDALION [fig(s) 5] “Embedding”, “LSTM”s, “CNN” [par(s) 22] “FIG. 1 is a block diagram of an example system 100 for aggregating a database of businesses, according to an embodiment of the present disclosure. System 100 may include a plurality of user devices 102a, 100b, ... , 102n (102 generally) and a server device 106, all of which may be communicably coupled via network 104. In some embodiments, system 100 may include any number of user devices. For example, for an organization that manages accounting software and an associated database, there may be an extensive user base with thousands or even millions of users that may connect via respective user devices. Server device 106 may be configured to selectively send a variety of information, such as product or service recommendations or business categorizations, to multiple or single user devices .” ; ) However, MEDALION does not appear to explicitly teach: perform a correlation coefficient analysis to identify a subset of most correlated labels out of labels of a plurality of [invoices] and labels of a plurality of purchase orders (POs); train a neural network model, based on the subset of most correlated labels, to determine a mapping [between] the plurality of invoices and the plurality of POs, wherein the mapping indicates that one or more invoices, of the plurality of invoices, are associated with a particular purchase order (PO) of the plurality of POs; evaluate particular PO data, of the particular PO, dynamically using a [reinforcement] learning model, BURRELL teaches perform a correlation coefficient analysis to identify a subset of most correlated labels out of labels of a plurality of invoices and labels of a plurality of purchase orders (POs); (BURRELL [fig(s) 1] [par(s) 52] “Identities 123 may include mappings between different attributes . For example, a purchase order document may include the purchase order number stored as an attribute in attributes 122 that is mapped to an invoice number in an invoice document to form a sale identity . Character recognition module 111 may determine identities 123 based on matching attributes between different documents. For example, purchase order and invoice documents described above can list the same set of products/services requested and provided, resulting in the linking of the purchase order and the invoice attributes and create a sale identity . Identities 123 obtained from mapping attributes of attributes 122 may include combined values of attributes.” [par(s) 85] “In step 430, transaction processing system 100 may group attributes into sets of attributes. Transaction processing system 100 may group attributes based on relationships (e.g., correlations 224 of FIG. 2) defined in transaction processing system 100. In some embodiments, relationships between attributes may be learned from previous iterations of transaction processing by transaction processing system 100. Transaction processing system 100 may include machine learning models that may be trained to learn different types of documents, their structure, and the relationships between various attributes that may be part of documents. Machine learning models used in transaction processing system 100 may include neural networks . In some embodiments, transaction processing system 100 components may determine document types using fuzzy logic systems.” See also [par(s) 68-72] ; ) train a neural network model, based on the subset of most correlated labels, to determine a mapping between the plurality of invoices and the plurality of POs, (BURRELL [fig(s) 1] [par(s) 52] “Identities 123 may include mappings between different attributes . For example, a purchase order document may include the purchase order number stored as an attribute in attributes 122 that is mapped to an invoice number in an invoice document to form a sale identity . Character recognition module 111 may determine identities 123 based on matching attributes between different documents. For example, purchase order and invoice documents described above can list the same set of products/services requested and provided, resulting in the linking of the purchase order and the invoice attributes and create a sale identity . Identities 123 obtained from mapping attributes of attributes 122 may include combined values of attributes.” [par(s) 85] “In step 430, transaction processing system 100 may group attributes into sets of attributes. Transaction processing system 100 may group attributes based on relationships (e.g., correlations 224 of FIG. 2) defined in transaction processing system 100. In some embodiments, relationships between attributes may be learned from previous iterations of transaction processing by transaction processing system 100. Transaction processing system 100 may include machine learning models that may be trained to learn different types of documents, their structure, and the relationships between various attributes that may be part of documents. Machine learning models used in transaction processing system 100 may include neural networks . In some embodiments, transaction processing system 100 components may determine document types using fuzzy logic systems.” See also [par(s) 68-72] ; ) wherein the mapping indicates that one or more invoices, of the plurality of invoices, are associated with a particular purchase order (PO) of the plurality of POs; (BURRELL [fig(s) 1] [par(s) 52] “Identities 123 may include mappings between different attributes . For example, a purchase order document may include the purchase order number stored as an attribute in attributes 122 that is mapped to an invoice number in an invoice document to form a sale identity . Character recognition module 111 may determine identities 123 based on matching attributes between different documents. For example, purchase order and invoice documents described above can list the same set of products/services requested and provided, resulting in the linking of the purchase order and the invoice attributes and create a sale identity . Identities 123 obtained from mapping attributes of attributes 122 may include combined values of attributes.” [par(s) 85] “In step 430, transaction processing system 100 may group attributes into sets of attributes. Transaction processing system 100 may group attributes based on relationships (e.g., correlations 224 of FIG. 2) defined in transaction processing system 100. In some embodiments, relationships between attributes may be learned from previous iterations of transaction processing by transaction processing system 100. Transaction processing system 100 may include machine learning models that may be trained to learn different types of documents, their structure, and the relationships between various attributes that may be part of documents. Machine learning models used in transaction processing system 100 may include neural networks . In some embodiments, transaction processing system 100 components may determine document types using fuzzy logic systems.” See also [par(s) 68-72] ; ) The combination of MEDALION, BURRELL, VASHISHT is combinable with BURRELL for the same rationale as set forth above with respect to claim 1. However, the combination of MEDALION, BURRELL does not appear to explicitly teach: evaluate particular PO data, of the particular PO, dynamically using a [reinforcement] learning model, VASHISHT teaches evaluate particular PO data, of the particular PO, dynamically using a reinforcement learning model, (VASHISHT [par(s) 80] “The RL agent 222 is configured to initialize Q-value function and learn the best optimal path for a particular type of payment transaction based on a reward function . The reward function depends upon approval and fraud probability scores of the particular type of transaction and the cost of applying authorizing components to the particular type of transaction. In one embodiment, the approval and fraud probability scores of the particular type of transaction are determined based on historical transaction data (i.e., a number of processed transactions that were approved or declined due to fraud). In one embodiment, the cost of applying an authorizing component is a transaction-level cost and stored at the product cost repository 228.” [par(s) 3] “Payment networks enable various types of payment transactions. A " card-not-present (CNP)" transaction is a type of payment transaction in which a consumer buys a product/service without the presence of a physical payment card (e.g., debit card, credit card, prepaid card ). In such transactions ( e.g., online/e-commerce, card-on-file), the payment card information is transmitted from a merchant, along with a flag that the payment transaction is a CNP transaction.” [par(s) 140] “The various data elements may include, but not limited to, transaction identifier, issuer identifier, issuer name, merchant name/identifier, acquirer name/identifier, cross-border transaction flag (e.g., cross border, domestic), transaction channel flag (e.g., e-commerce, POS, recurring payments), payment card type (e.g., credit, debit), card product type ( customer/commercial), card-not-present (CNP) transaction flag, response code flag (approve/decline), decline reason code (in case of declined transaction), product flag vectors (indicating applied authorizing components by the issuers), etc.” ; ) The combination of MEDALION, BURRELL, VASHISHT is combinable with VASHISHT for the same rationale as set forth above with respect to claim 1. Regarding claim 16 The claim is a system claim corresponding to the computer program product claim 13, and is directed to largely the same subject matter. Thus, it is rejected for the same reasons as given in the rejections of the computer program product claim. Regarding claim 17 The claim is a system claim corresponding to the computer program product claim 14, and is directed to largely the same subject matter. Thus, it is rejected for the same reasons as given in the rejections of the computer program product claim. Regarding claim 18 The claim is a system claim corresponding to the computer program product claim 9, and is directed to largely the same subject matter. Thus, it is rejected for the same reasons as given in the rejections of the computer program product claim. MEDALION further teaches receive PO data regarding a plurality of purchase orders (POs) and invoice data regarding a plurality of invoices, (MEDALION [fig(s) 1, 10] [par(s) 17-19] “The method may analyze the vendor offerings, identify the associated clusters, and generate a bounded list of purchases for that business. For example, a pre-defined list of purchase orders that a dry-cleaning business may select from within accounting software may include "hanger and accessories" and "pressing accessories." The method of the present disclosure may analyze the dry-cleaning business' purchases and identify associated clusters (e.g., formed by analyzing other dry cleaning services' purchases) with more specialized descriptions such as "plastic bags," "irons," and "press pads."” [par(s) 3] “In addition, it is difficult to summarize the services or products offered by a business, despite databases typically having access to the business' invoices and bank transactions. Obtaining accurate and appropriate lists of business offerings (e.g., services and/or products) requires manual entry or adherence to a pre-defined list when selecting offerings.” [par(s) 32] “Invoice preparation module 118 may be configured to extract text from invoices. In some embodiments, invoice preparation module 118 may be configured to use optical character recognition (OCR) to extract text from invoice files (e.g., PDFs, Word Documents, etc.) or images (e.g., PNG, JPEG, etc.). Invoice files or images may be stored in database 122. In some embodiments, invoice preparation module 118 may be configured to generate a representative vector of a vendor based on a plurality of vectors describing line items of the invoice.” ; ) wherein the PO data and the invoice data are received from different devices associated with different cloud systems; (MEDALION [fig(s) 1, 10] [par(s) 17-19] “The method may analyze the vendor offerings, identify the associated clusters, and generate a bounded list of purchases for that business. For example, a pre-defined list of purchase orders that a dry-cleaning business may select from within accounting software may include "hanger and accessories" and "pressing accessories." The method of the present disclosure may analyze the dry-cleaning business' purchases and identify associated clusters (e.g., formed by analyzing other dry cleaning services' purchases) with more specialized descriptions such as "plastic bags," "irons," and "press pads."” [par(s) 3] “In addition, it is difficult to summarize the services or products offered by a business, despite databases typically having access to the business' invoices and bank transactions. Obtaining accurate and appropriate lists of business offerings (e.g., services and/or products) requires manual entry or adherence to a pre-defined list when selecting offerings.” [par(s) 32] “Invoice preparation module 118 may be configured to extract text from invoices. In some embodiments, invoice preparation module 118 may be configured to use optical character recognition (OCR) to extract text from invoice files (e.g., PDFs, Word Documents, etc.) or images (e.g., PNG, JPEG, etc.). Invoice files or images may be stored in database 122. In some embodiments, invoice preparation module 118 may be configured to generate a representative vector of a vendor based on a plurality of vectors describing line items of the invoice.” ; ) Regarding claim 19 The claim is a system claim corresponding to the computer program product claim 10, and is directed to largely the same subject matter. Thus, it is rejected for the same reasons as given in the rejections of the computer program product claim. In addition, MEDALION teaches receive, via a network, PO data regarding the plurality of POs and invoice data regarding the plurality of invoices; (MEDALION [fig(s) 1, 10] [par(s) 17-19] “The method may analyze the vendor offerings, identify the associated clusters, and generate a bounded list of purchases for that business. For example, a pre-defined list of purchase orders that a dry-cleaning business may select from within accounting software may include "hanger and accessories" and "pressing accessories." The method of the present disclosure may analyze the dry-cleaning business' purchases and identify associated clusters (e.g., formed by analyzing other dry cleaning services' purchases) with more specialized descriptions such as "plastic bags," "irons," and "press pads."” [par(s) 3] “In addition, it is difficult to summarize the services or products offered by a business, despite databases typically having access to the business' invoices and bank transactions. Obtaining accurate and appropriate lists of business offerings (e.g., services and/or products) requires manual entry or adherence to a pre-defined list when selecting offerings.” [par(s) 32] “Invoice preparation module 118 may be configured to extract text from invoices. In some embodiments, invoice preparation module 118 may be configured to use optical character recognition (OCR) to extract text from invoice files (e.g., PDFs, Word Documents, etc.) or images (e.g., PNG, JPEG, etc.). Invoice files or images may be stored in database 122. In some embodiments, invoice preparation module 118 may be configured to generate a representative vector of a vendor based on a plurality of vectors describing line items of the invoice.” ; ) Regarding claim 20 The claim is a system claim corresponding to the computer program product claim 12, and is directed to largely the same subject matter. Thus, it is rejected for the same reasons as given in the rejections of the computer program product claim. Conclusion Any inquiry concerning this communication or earlier communications from the examiner should be directed to SEHWAN KIM whose telephone number is (571)270-7409. The examiner can normally be reached Mon - Fri 9:00 AM - 5:00 PM. 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, Michael J Huntley can be reached on (303) 297-4307. 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. /SEHWAN KIM/Examiner, Art Unit 2129 Application/Control Number: 18/167,855 Page 2 Art Unit: 2129 Application/Control Number: 18/167,855 Page 3 Art Unit: 2129 Application/Control Number: 18/167,855 Page 4 Art Unit: 2129 Application/Control Number: 18/167,855 Page 5 Art Unit: 2129 Application/Control Number: 18/167,855 Page 6 Art Unit: 2129 Application/Control Number: 18/167,855 Page 7 Art Unit: 2129 Application/Control Number: 18/167,855 Page 8 Art Unit: 2129 Application/Control Number: 18/167,855 Page 9 Art Unit: 2129 Application/Control Number: 18/167,855 Page 10 Art Unit: 2129 Application/Control Number: 18/167,855 Page 11 Art Unit: 2129 Application/Control Number: 18/167,855 Page 12 Art Unit: 2129 Application/Control Number: 18/167,855 Page 13 Art Unit: 2129 Application/Control Number: 18/167,855 Page 14 Art Unit: 2129 Application/Control Number: 18/167,855 Page 15 Art Unit: 2129 Application/Control Number: 18/167,855 Page 16 Art Unit: 2129 Application/Control Number: 18/167,855 Page 17 Art Unit: 2129 Application/Control Number: 18/167,855 Page 18 Art Unit: 2129 Application/Control Number: 18/167,855 Page 19 Art Unit: 2129 Application/Control Number: 18/167,855 Page 20 Art Unit: 2129 Application/Control Number: 18/167,855 Page 21 Art Unit: 2129 Application/Control Number: 18/167,855 Page 22 Art Unit: 2129 Application/Control Number: 18/167,855 Page 23 Art Unit: 2129 Application/Control Number: 18/167,855 Page 24 Art Unit: 2129 Application/Control Number: 18/167,855 Page 25 Art Unit: 2129 Application/Control Number: 18/167,855 Page 26 Art Unit: 2129 Application/Control Number: 18/167,855 Page 27 Art Unit: 2129 Application/Control Number: 18/167,855 Page 28 Art Unit: 2129 Application/Control Number: 18/167,855 Page 29 Art Unit: 2129 Application/Control Number: 18/167,855 Page 30 Art Unit: 2129 Application/Control Number: 18/167,855 Page 31 Art Unit: 2129 Application/Control Number: 18/167,855 Page 32 Art Unit: 2129 Application/Control Number: 18/167,855 Page 33 Art Unit: 2129 Application/Control Number: 18/167,855 Page 34 Art Unit: 2129 Application/Control Number: 18/167,855 Page 35 Art Unit: 2129 Application/Control Number: 18/167,855 Page 36 Art Unit: 2129 Application/Control Number: 18/167,855 Page 37 Art Unit: 2129 Application/Control Number: 18/167,855 Page 38 Art Unit: 2129 Application/Control Number: 18/167,855 Page 39 Art Unit: 2129 Application/Control Number: 18/167,855 Page 40 Art Unit: 2129 Application/Control Number: 18/167,855 Page 41 Art Unit: 2129 Application/Control Number: 18/167,855 Page 42 Art Unit: 2129 Application/Control Number: 18/167,855 Page 43 Art Unit: 2129 Application/Control Number: 18/167,855 Page 44 Art Unit: 2129 Application/Control Number: 18/167,855 Page 45 Art Unit: 2129 Application/Control Number: 18/167,855 Page 46 Art Unit: 2129 Application/Control Number: 18/167,855 Page 47 Art Unit: 2129 Application/Control Number: 18/167,855 Page 48 Art Unit: 2129 Application/Control Number: 18/167,855 Page 49 Art Unit: 2129 Application/Control Number: 18/167,855 Page 50 Art Unit: 2129 Application/Control Number: 18/167,855 Page 51 Art Unit: 2129 Application/Control Number: 18/167,855 Page 52 Art Unit: 2129 Application/Control Number: 18/167,855 Page 53 Art Unit: 2129 Application/Control Number: 18/167,855 Page 54 Art Unit: 2129 Application/Control Number: 18/167,855 Page 55 Art Unit: 2129 Application/Control Number: 18/167,855 Page 56 Art Unit: 2129 Application/Control Number: 18/167,855 Page 57 Art Unit: 2129 Application/Control Number: 18/167,855 Page 58 Art Unit: 2129 Application/Control Number: 18/167,855 Page 59 Art Unit: 2129 Application/Control Number: 18/167,855 Page 60 Art Unit: 2129 Application/Control Number: 18/167,855 Page 61 Art Unit: 2129 Application/Control Number: 18/167,855 Page 62 Art Unit: 2129 Application/Control Number: 18/167,855 Page 63 Art Unit: 2129 Application/Control Number: 18/167,855 Page 64 Art Unit: 2129 Application/Control Number: 18/167,855 Page 65 Art Unit: 2129 Application/Control Number: 18/167,855 Page 66 Art Unit: 2129 Application/Control Number: 18/167,855 Page 67 Art Unit: 2129 Application/Control Number: 18/167,855 Page 68 Art Unit: 2129 Application/Control Number: 18/167,855 Page 69 Art Unit: 2129 Application/Control Number: 18/167,855 Page 70 Art Unit: 2129 Application/Control Number: 18/167,855 Page 71 Art Unit: 2129 Application/Control Number: 18/167,855 Page 72 Art Unit: 2129 Application/Control Number: 18/167,855 Page 73 Art Unit: 2129 Application/Control Number: 18/167,855 Page 74 Art Unit: 2129 Application/Control Number: 18/167,855 Page 75 Art Unit: 2129 Application/Control Number: 18/167,855 Page 76 Art Unit: 2129 Application/Control Number: 18/167,855 Page 77 Art Unit: 2129 Application/Control Number: 18/167,855 Page 78 Art Unit: 2129 Application/Control Number: 18/167,855 Page 79 Art Unit: 2129 Application/Control Number: 18/167,855 Page 80 Art Unit: 2129 Application/Control Number: 18/167,855 Page 81 Art Unit: 2129 Application/Control Number: 18/167,855 Page 82 Art Unit: 2129 Application/Control Number: 18/167,855 Page 83 Art Unit: 2129 Application/Control Number: 18/167,855 Page 84 Art Unit: 2129 Application/Control Number: 18/167,855 Page 85 Art Unit: 2129 Application/Control Number: 18/167,855 Page 86 Art Unit: 2129 Application/Control Number: 18/167,855 Page 87 Art Unit: 2129 Application/Control Number: 18/167,855 Page 88 Art Unit: 2129 Application/Control Number: 18/167,855 Page 89 Art Unit: 2129 Application/Control Number: 18/167,855 Page 90 Art Unit: 2129 Application/Control Number: 18/167,855 Page 91 Art Unit: 2129 Application/Control Number: 18/167,855 Page 92 Art Unit: 2129 Application/Control Number: 18/167,855 Page 93 Art Unit: 2129
Prosecution Insights