DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Status of Application
The following is a Non-Final Office Action. In response to Examiner's communication on 01/09/2026, Applicant on 02/09/2026, amended Claims 1, 12, 18 and cancelled Claims 9-11, 16, 20. Claims 1-2, 4-8, 12-15, 18-19 are now pending in this application and have been rejected below.
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 02/09/2026 has been entered.
Response to Amendment
Applicants’ amendments are insufficient to overcome the 35 USC 101 rejections set forth in the previous action. The rejections have been updated to address the amendments and maintained below.
Applicants’ amendments render moot the 35 USC 103 rejections set forth in the previous action in view of new and updated grounds for rejection necessitated by Applicants’ amendments. Therefore, these rejections are withdrawn in view of the new grounds for rejection necessitated by Applicants’ amendments, as set forth below.
Applicant’s amendments necessitate new grounds of rejection under 35 USC 112(b).
Response to Arguments – 35 USC § 101
Applicant's arguments with respect to the 35 USC 101 rejections have been fully considered but they are not persuasive.
Applicant firstly argues that technological improvements render the recited abstract ideas into a practical. Examiner respectfully disagrees.
To resolve whether the claims are directed to an improvement in technology, it is necessary to resolve the subject of improvement, as an improvement to an abstract idea does not constitute an improvement to technology per the MPEP. Examiner points to the high level of generality that Applicant’s cited limitations of iterative improvement, time series analysis, clustering classification are recited.. A time series analysis and clustering classification are both techniques that are expressly mentally performable via mental observations and judgments. What is recited is not a clear and distinct improvement to machine learning model operation, but rather a vague connection from the task at hand, namely financial form analysis, to the field of machine learning. “fine-tuning the machine learning models”, “applying a time series analysis to the machine learning models”, “applying a clustering classification to the machine learning models” – the specific mechanics of implementing these techniques are left unspecified, rendering said recitations to be generic in the art. Similar logic applies with respect to “training the machine learning models iteratively”. Progressive improvement to a statistical model is another abstract idea – if Applicant, where supported by Applicant’s specification, expounded upon the exact mechanisms of how the models are iteratively improved, that could be a relevant indicia under Step 2B, but where limitations broadly combine generic computing components and additional limitations that are specified at a high level of generality with more abstract ideas, an improvement to technology cannot be said to be present.
Accordingly, the rejections under 35 USC 101 have been maintained are updated to address the amendments below.
Response to Arguments –35 USC § 103
Applicant' s arguments with respect to the rejection of Claims 1-2, 4-16, 18-20 under 35 USC 103 have been considered but are moot in light of new grounds of rejections necessitated by applicant’s amendments.
Applicant’s arguments pertaining to the lack of support for iterative training, clustering, and reasonable pertinence are rendered moot in view of new grounds of rejection necessitated by Applicant’s amendments. Applicant respectfully points to the updated grounds of rejection below, with associated mappings and rationale to combine.
Applicant subsequently argues that the prior art does not teach or suggest time series analysis. Examiner disagrees.
The cited portions of Plehn-Dujowich are not solely relied upon to teach the totality of the limitations. In [0072] of Plehn-Dujowich, “Time series data permits trend analysis of private company stock filings”. The teaching we derive from it is specifically the trend analysis, facilitated by time series data. The broadest reasonable interpretation of “applying a time series analysis to the machine learning models” encompasses performing a time series analysis to the output of the machine learning models. In [0076] of Agrawal, “At least partly based on the comparison of the two datasets, one or more machine-learning prediction models can be trained in step 308. The machine-learning prediction models may be then employed to determine a likelihood of breach of each analyzed merchant in step 310. More than one machine-learning prediction model may be used, to improve the accuracy of data breach detection”. The plurality of models of Agrawal work by outputting scores, which can function as inputs to time series analysis, in [0072] of Agrawal, “Once trained, the machine learning prediction models 218, 220, either individually or as an ensemble scoring algorithm 222, may be used to assign a breach score to individual merchants. An output score from the prediction models 218, 220 or the ensemble scoring algorithm 222 may be normalized between 0 and 1, or scaled between 0 and 100, as a representation of likelihood of a given merchant to be breached (e.g., a confidence score, also referred herein as a “breach score””. With a broad prescription of “applying a time series analysis to the machine learning models”, our teaching of analysis, derived from Plehn-Dujowich, with inputs coming from a plurality of machine learning models, derived from Agrawal, disclose the broadest reasonable limitations of the Claims.
The logic above, as well as the rationale under Response to Arguments – 35 USC § 101 also applies to Applicant’s arguments under E. Unexpected Results. If the Applicant intends to assert that some unexpected benefit is derived or that time series/clustering are applied in a novel manner, it is missing from the broadest reasonable interpretation of the claims. What is instead claim is the conjunction of two well known techniques in a way that yields predictable results.
Claim Rejections - 35 USC § 112(b)
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-2, 4-8, 12-15, 18-19 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.
The term "unusual” in Claims 1, 12, 18 is a relative term
which renders the claim indefinite. The term "unusual" is not defined by the
claim, the specification does not provide a standard for ascertaining the
requisite degree, and one of ordinary skill in the art would not be reasonably apprised of the scope of the invention. the applicant), regards
as the invention.
The term "anomalous" in Claims 1, 12, 18 is a relative term which renders the claim indefinite. The term "anomalous" is not defined by the
claim, the specification does not provide a standard for ascertaining the
requisite degree, and one of ordinary skill in the art would not be
reasonably apprised of the scope of the invention.
Dependent Claims 2, 4-8, 13-15, 19 are rejected due to their dependency on rejected independent Claims 1, 12 and 18.
Appropriate correction is required.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-2,4-8, 12-15, 18-19 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
101 Analysis – Step 1
The claims are directed to an apparatus and method. Therefore, the claim is directed to at least one of the four statutory categories.
101 Analysis – Step 2A
Regarding Prong 1 of the Step 2A analysis in the MPEP, the claims are to be analyzed to determine whether they recite subject matter that is directed to a judicial expectation, namely a law of nature, a natural phenomenon, or one of the follow groups of abstract ideas: a) mathematical concepts, b) certain methods of organizing human activity, and/or c) mental processes.
Independent Claim 1 includes limitations that recite an abstract idea and will henceforth be used as a representative claim for the 101 rejection until otherwise noted. Claim 1 recites:
A method for optimizing use of a computer processor to perform fraud detection, said method comprises automating detection of fraud by using the computer processor to eliminate manual processing of data in analysis of publicly available information about a public corporation, the method comprising: collecting automatically every 36 hours or less, using a computer processor, one or more forms which are publicly available relating to an organization from an electronic portal; cleaning and preprocessing automatically, using the computer processor, data found in the one or more forms to produce cleaned and preprocessed data; extracting automatically, using the computer processor to run one or more machine learning models, one or more sets of features from the cleaned and preprocessed data; wherein the one or more sets of features comprise a set of features related to liquid, solvency, and profitability ratio classification, a set of features related to disclosure classification, a set of features related to sentiment analysis, a set of features related to anomaly detection classification, a set of features related to ownership analysis classification, and a set of features related to ESG disclosure classification; determining automatically, using the computer processor to run machine learning models, if two or more thresholds have been exceeded indicating a risk of fraud; wherein the machine learning models comprise a liquid, solvency, and profitability ratio classification machine learning model, a disclosure classification machine learning model, a sentiment analysis machine learning model, an anomaly detection classification machine learning model, an ownership analysis classification machine learning model, and an ESG disclosure classification machine learning model; and notifying automatically an administrator, using the computer processor, when one or more thresholds have been exceeded, training the machine learning models iteratively by testing the machine learning models with new extractions of forms from the electronic portal and fine-tuning the machine learning models based on testing results; applying a time series analysis to the machine learning models to detect unusual temporal patterns; notifying the administrator when an unusual temporal pattern has been detected; applying a clustering classification to one or more machine learning models; and notifying the administrator when an anomalous cluster has been detected; wherein the computer processor automates detection of fraud and eliminates manual processing of data in the analysis of publicly available information about a public corporation.
The examiner submits that the foregoing bolded limitation(s) constitute an abstract idea because under its broadest reasonable interpretation, the claim covers a mental process. “alerting an organization about activity that may be fraudulent”, “collecting every 36 hours or less…one or more forms”, “cleaning and preprocessing…data found in the one or more forms”, “extracting…one or more sets of features, “determining…if one or more thresholds have been exceeded”, “notifying an administrator…when one or more thresholds have been exceeded”, are mental processes that could be
performed by a human with a pen and paper. Per the MPEP, merely
adapting them into the context of a technological environment with computing parts does not preclude them from being abstract
Accordingly, the claim recites at least one abstract idea.
Independent Claims 12 and 18 recite at least one abstract idea by virtue of presenting similar limitations.
Dependent Claims 2, 4-11, 13-16, 19-20 recite at least one abstract idea by virtue of their dependency from independent Claims 1, 12 and 18 respectively.
101 Analysis – Step 2A, Prong II
Regarding Prong II of the Step 2A analysis in the MPEP, the claims are to be analyzed to determine whether the claim, as a whole, integrates the abstract into practical application. As noted in the MPEP, it must be determined whether any additional elements in the claim beyond the judicial exception integrate the exception into a practical application in a manner that imposes a meaningful limit on the judicial exception. The courts have indicated that additional elements, such as merely using a computer to implement an abstract idea, adding insignificant extra solution activity, or generally linking use of a judicial exception to a particular technological environment or field of use do not integrate a judicial exception into a “practical application.
In the present case, the additional limitations beyond the above-noted abstract idea are as follows (where the underlined portions are the “additional limitations” while the bolded portions continue to represent the “abstract idea”):
A method for optimizing use of a computer processor to perform fraud detection, said method comprises automating detection of fraud by using the computer processor to eliminate manual processing of data in analysis of publicly available information about a public corporation, the method comprising: collecting automatically every 36 hours or less, using a computer processor, one or more forms which are publicly available relating to an organization from an electronic portal; cleaning and preprocessing automatically, using the computer processor, data found in the one or more forms to produce cleaned and preprocessed data; extracting automatically, using the computer processor to run one or more machine learning models, one or more sets of features from the cleaned and preprocessed data; wherein the one or more sets of features comprise a set of features related to liquid, solvency, and profitability ratio classification, a set of features related to disclosure classification, a set of features related to sentiment analysis, a set of features related to anomaly detection classification, a set of features related to ownership analysis classification, and a set of features related to ESG disclosure classification; determining automatically, using the computer processor to run machine learning models, if two or more thresholds have been exceeded indicating a risk of fraud; wherein the machine learning models comprise a liquid, solvency, and profitability ratio classification machine learning model, a disclosure classification machine learning model, a sentiment analysis machine learning model, an anomaly detection classification machine learning model, an ownership analysis classification machine learning model, and an ESG disclosure classification machine learning model; and notifying automatically an administrator, using the computer processor, when one or more thresholds have been exceeded, training the machine learning models iteratively by testing the machine learning models with new extractions of forms from the electronic portal and fine-tuning the machine learning models based on testing results;applying a time series analysis to the machine learning models to detect unusual temporal patterns;notifying the administrator when an unusual temporal pattern has been detected;applying a clustering classification to one or more machine learning models;and notifying the administrator when an anomalous cluster has been detected; wherein the computer processor automates detection of fraud and eliminates manual processing of data in the analysis of publicly available information about a public corporation.
For the following reason(s), the examiner submits that the above identified additional limitations do not integrate the above-noted abstract idea into a practical application.
As it pertains to Claim 1, the additional elements in the claims include “computer processor”, sourcing data from an “electronic portal”, the usage of “one or more machine learning models”, along with the specific implementation of models disclosed in the application. When considered in view of the claim as a whole, the additional elements do not integrate the abstract idea into a practical application because the additional elements are generic computing components that are merely used as a tool to perform the recited abstract idea and/or do no more than generally link the use of the recited abstract idea to a particular technological environment or field of use under Step 2A Prong Two.
Thus, taken alone, the additional elements do not integrate the abstract idea into a practical application. Further, looking at the additional limitation(s) as an ordered combination or as a whole, the limitation(s) add nothing that is not already present when looking at the elements taken individually. For instance, there is no indication that the additional elements, when considered as a whole, reflect an improvement in the functioning of a computer or an improvement to another technology or technical field, apply or use the above-noted judicial exception to effect a particular treatment or prophylaxis for a disease or medical condition, implement/use the above-noted judicial exception with a particular machine or manufacture that is integral to the claim, effect a transformation or reduction of a particular article to a different state or thing, or apply or use the judicial exception in some other meaningful way beyond generally linking the use of the judicial exception to a particular technological environment, such that the claim as a whole is not more than a drafting effort designed to monopolize the exception (MPEP § 2106.05). Accordingly, the additional limitation(s) does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing an abstract idea.
Independent Claims 12 and 18 do not integrate the recited abstract ideas into a practical application by virtue of presenting substantially similar limitations.
Claim 8 additionally recites “an Electronic Data Gathering, Analysis, and Retrieval (EDGAR) database”. This does not integrate the recited abstract ideas into a practical application by analogous reasoning as outlined above.
Claims 2,4-7, 9-16,18-20 do not recite additional elements beyond those found in Claims from which they depend, and therefore do not integrate recited abstract ideas into a practical application.
101 Analysis – Step 2B
Regarding Step 2B of the MPEP, representative independent claim 1 does not include additional elements (considered both individually and as an ordered combination) that are sufficient to amount to significantly more than the judicial exception for the same reasons to those discussed above with respect to determining that the claim does not integrate the abstract idea into a practical application. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements amount to generic computing components that are merely used
as a tool to perform the recited abstract idea and/or do no more than
generally link the use of the recited abstract idea to a particular
technological environment or field of use. Further, looking at the additional
elements as an ordered combination adds nothing that is not already
present when considering the additional elements individually.
Independent Claims 12 and 18 do not integrate the recited abstract ideas into a practical application or amount to significantly more by virtue of presenting substantially similar limitations.
Claim 8 additionally recites “an Electronic Data Gathering, Analysis, and Retrieval (EDGAR) database”. This does not integrate the recited abstract ideas into a practical application amount to significantly more by analogous reasoning as outlined above.
Claims 2,4-7, 9-16,18-20 do not recite additional elements beyond those found in Claims from which they depend, and therefore do not integrate recited abstract ideas into a practical application or amount to significantly more.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-2,4-8,12-15, 18-19 are rejected under 35 U.S.C. 103 as being unpatentable over Schwartz(US 11216895 B1) in view of Agrawal(US 20210279731 A1) in further view of Yan(US 20180285886 A1) in further view of Plehn-Dujowich(US 20240378674 A1).
Claim 1
As to Claim 1, Schwartz teaches:
A method for optimizing use of a computer processor to perform fraud detection, said method comprises automating detection of fraud by using the computer processor to eliminate manual processing of data in analysis of publicly available information about a public corporation, the method comprising: collecting automatically every 36 hours or less, using a computer processor, one or more forms which are publicly available relating to an organization from an electronic portal;
In Col 40 Lines 6-13, "The data fields of interest may be drawn from multiple documents and databases, including without limitation, historical data sources related to, for example, securities information. This data may be collected from a plurality of structured and unstructured data sources, as described herein, through the use of proprietary data connectors, APIs, manual upload or entry, or some other means of data transfer". In Col 13 Line 17 - Col 14 Line 6, “In some embodiments, the portfolio monitoring engine 305 determines or track losses on portfolio securities on a day, week, month or other time basis. A minimum loss threshold can be specified to configure the portfolio monitoring engine 305 to flag only those securities having asynchronous activity that meet the loss threshold. Other criteria for flagging asynchronous activity can be configured”. We consider the analytical functions to be autonomous in light of Col 8 Lines 17-28, “In some embodiments, the claims on the watch list are reviewed or screened using an automated process that correlates information relating to the asynchronous behavior of the securities on the watch list with information such as financial market updates, updates from portfolio entities, market moving or breaking news, legal activity news, and/or other information aggregated from news sites, blogs, Twitter feed, and/or other sources ((e.g., via web crawlers, data feeds, etc.) to determine if the asynchronous behavior exhibited by the securities on the watch list is likely due to securities fraud or due to other events (e.g., drop in stock price of an entity due to news about CEO's health problems).”. This can pertain to a public corporation, in Col 3 Lines 5-9, “embodiments, unstructured data may include social media data, custodian data, audio recordings, financial reports, or some other type of unstructured data. Financial reports may include Securities and Exchange Commission reports”.
cleaning and preprocessing automatically, using the computer processor, data found in the one or more forms to produce cleaned and preprocessed data;
In Col 43 Lines 22-35, " In embodiments of the aggregated use of structured and unstructured data, that is cleansed and made accurate, may allow comparison of more variables for better insights to predict possible actionable securities fraud litigation that inevitably trails disclosure of the deviations and failures, and to predict the probability of instances in which an issuer's stock price is likely inflated due to failures to disclose deviations from expected behaviors or regulatory violations before those deviations or failures are publicly disclosed. Cleansing data may be an analytic option, for example in the cognitive analytic layer of the CIOR system. An example of data cleansing may include, but is not limited to, data standardization performed prior to submitting data for analysis". As Col 8 Lines 17-28 expounds upon the automation of analysis, we consider preprocessing and cleaning data into an alternative format to implicitly be automatable, especially given the automated delivery of the analysis to human stakeholders.
Schwartz does not teach:
extracting automatically, using the computer processor to run one or more machine learning models,
However, Agrawal teaches:
extracting automatically, using the computer processor to run one or more machine learning models,
In [0068], "The modeling and detection server 112 receives a number of data sources to build model input datasets, e.g., profiles, to train machine learning prediction models. The various data sources and their use to construct model input datasets are further discussed below in relation to FIG. 2. The modeling and detection server 112 may employ one or more predictive models, e.g., machine-learning algorithms, to determine the likelihood of a given merchant having experienced a data breach event. The predictive models may be trained based on an evaluation of historic transaction data relative to known historic fraudulent activity". In [0071], “With specific reference to FIG. 2, and in non-limiting embodiments or aspects of the disclosure, provided is a method 200 for early detection of and response to a merchant data breach through machine-learning analysis. The method 200 includes a model training process 201, in which data from multiple data sources are used to create prediction model input feature datasets, e.g., model training profiles…The modeling data 212, taken from the data sources, may be used to create a number of features, e.g., feature vectors, to form model input datasets, e.g., training profiles”. Citing the example methods of a Fully Connected Neural Network in 220; we note that in modern deep learning methods, feature extraction tends to be an automatic part of the training process, and therefore automatic feature extraction is implicitly disclosed over the course of training.
However, Yan teaches:
one or more sets of features from the cleaned and preprocessed data; wherein the one or more sets of features comprise: a set of features related to liquid,
In [0078-0085], “Thousands of information items can be used by the Computer Server 112 and the machine learning AI. For example, an exemplary listing of 100 illustrative data information attributes that can be employed by the system is shown in Table 1. The data can be pulled from public information source 316, business entity information source 318, or global partner sources 320. Data items can cover data systems including 1) firmgraphics, 2) financials, 3) trade/payment, 4) social media, 5) compliance violations, 6) public information 7) business entity scores and ratings”. See in Table 1 following [0085], “35 Current Ratio 36 Quick Ratio”. These two ratios would be well known to one of ordinary skill in the art as representing liquidity ratios.
solvency,
See in Table 1 following [0085], “34 Solvency Ratio (industry quartile)”.and profitability ratio classification,
In Table 1 following [0085], “29 Profit Margin”. It would be well known to one of ordinary skill in the art that profit margin, generally specified as a percentage/ratio, represents profitability.
Schwartz teaches:
a set of features related to disclosure classification,
In Col 43 Lines 22-31, "In embodiments of the aggregated use of structured and unstructured data, that is cleansed and made accurate, may allow comparison of more variables for better insights to predict possible actionable securities fraud litigation that inevitably trails disclosure of the deviations and failures, and to predict the probability of instances in which an issuer's stock price is likely inflated due to failures to disclose deviations from expected behaviors or regulatory violations before those deviations or failures are publicly disclosed".
a set of features related to sentiment analysis
In Col 3 Lines 5-9, "In embodiments, unstructured data may include social media data, custodian data, audio recordings, financial reports, or some other type of unstructured data. Financial reports may include Securities and Exchange Commission reports". In Col 39 Lines 42-49, " Unstructured text data may be analyzed using cognitive analytics to extract meaningful information such as, for example, entities (e.g., name of security, regulatory authority, name of the court, related parties) keywords (e.g., fraud, police, crime), tone (e.g., joy, anger sadness, fear, disgust) and sentiment (e.g., positive, negative) and key concept (e.g., insider, legal, investigation)".
, a set of features related to anomaly detection classification,
In Col 45 Lines 13-17, "In embodiments, one element of the overall DAA process may be to monitor incoming data from a filer and identify any anomalies or inconsistencies. The DAA import and posting process may include numerous processes to identify anomalies and report them.".
Schwartz does not teach:
a set of features related to ownership analysis classification,
However, Plehn-Dujowich teaches:
a set of features related to ownership analysis classification,
In [0078], "Clearly identifying corporate officers and directors of branches in a clear format may be helpful for performing meaningful investment analysis. The ability to see branch relationships, as displayed in FIG. 21, and the officers who run these entities helps to lift the “corporate veil” of ownership. This information is key to provide parent relationships, status and permit in depth background checks of officers".
Schwartz teaches:
and a set of features related to ESG disclosure classification;
In Col 43 Lines 22-31, "In embodiments of the aggregated use of structured and unstructured data, that is cleansed and made accurate, may allow comparison of more variables for better insights to predict possible actionable securities fraud litigation that inevitably trails disclosure of the deviations and failures, and to predict the probability of instances in which an issuer's stock price is likely inflated due to failures to disclose deviations from expected behaviors or regulatory violations before those deviations or failures are publicly disclosed". In Col 40 Lines 52-54, "Data used in the CIOR analytics platform may also include environment, sustainability and governance data (or “ESG data”), which may be used to refine analyses run on the platform".
Agrawal teaches:
determining automatically, using the computer processor to run machine learning models, if two or more thresholds have been exceeded indicating a risk of fraud;
In [0072], "Each model 218, 220 may output a score that, when compared to a predetermined threshold value at a given false positive tolerance level, indicates whether or not a merchant has been breached. Through ensembling, a merchant may be determined to be breached if either model 218, 220 indicates the merchant was breached. Alternatively, a merchant may be determined to be breached if both models 218, 220 indicate the merchant was breached. In non-limiting embodiments or aspects, the combined score of both models may be compared to a composite threshold, which may indicate whether or not a merchant has been breached". Note that plurality of scores generated by the models as well as the plurality of thresholds.
Agarwal teaches:
wherein the machine learning models comprise a liquid, solvency, and profitability ratio classification machine learning model, a disclosure classification machine learning model, a sentiment analysis machine learning model, an anomaly detection classification machine learning model, an ownership analysis classification machine learning model, and an ESG disclosure classification machine learning model;
In [0076], “At least partly based on the comparison of the two datasets, one or more machine-learning prediction models can be trained in step 308. The machine-learning prediction models may be then employed to determine a likelihood of breach of each analyzed merchant in step 310. More than one machine-learning prediction model may be used, to improve the accuracy of data breach detection. Different types of machine-learning algorithms may be used”. Above, we have noted that we extract data relating to a multitude of features, each of which corresponding to a machine learning model. Based on the technique of using a plurality of machine learning models as taught in Agarwal, we consider this limitation disclosed as one such arrangement of the models in Agarwal, in which each particular model focuses on a given feature.
notifying automatically an administrator, using the computer processor, when one or more thresholds have been exceeded,
Hardware is described in [0006]. In [0072], "Each model 218, 220 may
output a score that, when compared to a predetermined threshold value at
a given false positive tolerance level, indicates whether or not a merchant
has been breached. Through ensembling, a merchant may be determined
to be breached if either model 218, 220 indicates the merchant was
breached. Alternatively, a merchant may be determined to be breached if
both models 218, 220 indicate the merchant was breached. In non-limiting
embodiments or aspects, the combined score of both models may be
compared to a composite threshold, which may indicate whether or not a
merchant has been breached". In [0073], "The scores may also be used for
automated alerts 228, such as notifications on computer interfaces for
financial device holders, merchants, or security personnel".
training the machine learning models iteratively by testing the machine learning models with new extractions of forms from the electronic portal and fine-tuning the machine learning models based on testing results;
In [0104], “The system can also be configured for recursive iteration of the algorithm upon due diligence or other post facto validation of an inferred or likely TPI as a known TPI. Moreover, over time classifiers can be periodically reapplied and results reconfirmed to find the best classifiers for additional and changing data sets and windows. For example, as shown at block 503, subsequent compliance due diligence efforts, combined with the AI analytics results can generate additional confirmed TPIs or first identified as predicted or inferred TPIs…For example, there can be new, known TPIs provided from other data sources or from other due diligence, including via conventional approaches….The new confirmation of validated true non-TPIs, known TPIs and confirmed TPIs can then be used to refine and optimize the AI for more accurate predictions. Adaptive learning in next iterations make the analytics results more reliable and accurate, as well as able to reflect real-world changes and windows for TPI identification and weighting of parameters”. Note the adaptive fine tuning in response to the ingestion of new data. Data sources can encompass portals as described in [0086-0089].
Plehn-Dujowich teaches:
applying a time series analysis…to detect one or more unusual temporal patterns;
In [0075], "FIG. 18 illustrates a stock class series daily graph 1858 configured to permit early warning and alerts of private companies' activities, according to an aspect. Daily time series of stock filing data, normally not available, permits early warning and alerts of a private company's activities (FIG. 18)". In [0072], “Time series data permits trend analysis of private company stock filings”.
Agarwal teaches:
to one or more machine learning models;
In [0076], “At least partly based on the comparison of the two datasets, one or more machine-learning prediction models can be trained in step 308. The machine-learning prediction models may be then employed to determine a likelihood of breach of each analyzed merchant in step 310. More than one machine-learning prediction model may be used, to improve the accuracy of data breach detection”.
notifying the administrator
In [0070], "With further reference to FIG. 1, and in non-limiting
embodiments or aspects, the communication server 116 may generate and
communicate an alert to a financial device holder's communication device
122, such as a mobile device, to notify the financial device holder 102 that
a merchant 106 with whom they have transacted has likely experienced a
data breach"
Schwartz teaches:
when an unusual temporal pattern has been detected;
As an example of the monitoring criteria of Schwartz in Col 17 Lines 46-59,
"Foreign-Jill Smith Group Date(s): Pension Fund X held 4,745,426 shares
Loss (one-day) $1,784,491.04 USD JSG LN Equity (London) had a
negative 12.07% spread VS. its GICS-defined sub-industry cohort (hotels,
resorts and cruise lines). No securities class action lawsuits have been filed
against this company in the past year. The company has not issued any
stock in the last 36 months, but issued rights in June 2012". These unusual
events can be derived from the time series predictions - in [0081] of Plehn-
Dujowich, "In an embodiment, it may also be useful to explore the potential
of forecasting of time series trends. This forecast may include: future stock
authorizations of private companies; future earnings of public companies;
market movements; industry trends; likelihood of expansion or contraction
of a company or peer group; forecasting or predicting housing prices and
forecasting or predicting the commercial lease behavior of tenants, i.e.
whether a tenant renews the lease, expand the lease, etc". In this case, we
analogize the temporal patterns to be the predicted metrics derived from
the time series analysis of Plehn-Dujowich.
Yan teaches:
applying a clustering classification…cluster
In [0069], “For example, in an embodiment, the system can be configured to employ a clustering algorithm on a training database to group companies identified as TPIs and unidentified companies to generate classes and classifiers for TPI identification. The classifier results can then be compared against databases of known TPI companies to confirm accuracy and refine the classifiers. In at least one of the various embodiments Training databases can be compiled by mapping business entity data from business entity databases and other databases as described herein to pre-identified TPI companies to create weights for factors in TPI identification, for example, firmographics, business entity analytical scoring, past financial statements (e.g.: 5 year window), legal indicators and Business Linkage information”.
Agarwal teaches:
To one or more machine learning models;
In [0068], "The modeling and detection server 112 receives a number of data sources to build model input datasets, e.g., profiles, to train machine learning prediction models. The various data sources and their use to construct model input datasets are further discussed below in relation to FIG. 2. The modeling and detection server 112 may employ one or more predictive models, e.g., machine-learning algorithms, to determine the likelihood of a given merchant having experienced a data breach event".
and notifying the administrator when an anomalous … has been detected;
In [0070], "With further reference to FIG. 1, and in non-limiting
embodiments or aspects, the communication server 116 may generate and
communicate an alert to a financial device holder's communication device
122, such as a mobile device, to notify the financial device holder 102 that
a merchant 106 with whom they have transacted has likely experienced a
data breach".
wherein the computer processor automates detection of fraud
Hardware is described in [0006]. In [0072], “Each model 218, 220 may output a score that, when compared to a predetermined threshold value at a given false positive tolerance level, indicates whether or not a merchant has been breached. Through ensembling, a merchant may be determined to be breached if either model 218, 220 indicates the merchant was breached. Alternatively, a merchant may be determined to be breached if both models 218, 220 indicate the merchant was breached. In non-limiting embodiments or aspects, the combined score of both models may be compared to a composite threshold, which may indicate whether or not a merchant has been breached”. In [0073], “The scores may also be used for automated alerts 228, such as notifications on computer interfaces for financial device holders, merchants, or security personnel”.
Schwartz teaches:
and eliminates manual processing of data in the analysis of publicly available information about a public corporation.
We consider the analytical functions to be autonomous in light of Col 8 Lines 17-28, “In some embodiments, the claims on the watch list are reviewed or screened using an automated process that correlates information relating to the asynchronous behavior of the securities on the watch list with information such as financial market updates, updates from portfolio entities, market moving or breaking news, legal activity news, and/or other information aggregated from news sites, blogs, Twitter feed, and/or other sources ((e.g., via web crawlers, data feeds, etc.) to determine if the asynchronous behavior exhibited by the securities on the watch list is likely due to securities fraud or due to other events (e.g., drop in stock price of an entity due to news about CEO's health problems).”. This can pertain to a public corporation, in Col 3 Lines 5-9, “embodiments, unstructured data may include social media data, custodian data, audio recordings, financial reports, or some other type of unstructured data. Financial reports may include Securities and Exchange Commission reports”.
The primary reference, Schwartz discloses a system for identifying and predicting instances of securities claims, relating to fraud. Agrawal discloses means of using a plurality of machine learning models to detect fraud. The teachings of Agrawal are applicable to Schwartz as they both pertain to the problem of fraud detection, and Schwartz already supports the usage of machine learning models for fraud detection; see Col 41 Lines 5-12, Agrawal merely discloses the implementation of a specific means for coordinating the usage of said plurality.
Plehn-Dujowich discloses a system for monitoring stock filings to perform analysis on companies. Yan discloses a system meant to audit companies on the basis of financial data. The teachings of Plehn-Dujowich and Yan are applicable to Schwartz as they both involve analyzing the financial information of companies – in order to have a more robust analysis regarding the possibility of fraud, it would naturally serve to be able to capture a more complete image of a company’s financials.
It would have been obvious to one having ordinary skill in the art at the effective filling date of the invention to apply the plurality of machine learning models as taught in Agrawal and apply that to the means for fraud detection as taught in Schwartz. Motivation to do so comes from the fact that the claim is plainly directed to the predictable result of combining known items in the prior art, with the expected benefit that adopting the means for coordinating a plurality of specialized models would enable users to capture more granular relationships in the data. With respect to the systems for performing analysis for financial analysis disclosed in Yan and Plehn-Dujowich, benefit would be found in enriching the analysis by paying attention to various aspects of the data, as well as the incorporation of new analytical techniques.
Claims 12 and 18 are rejected as reciting substantially similar limitations as Claim 1.
With respect to additional limitations of Claim 12,
Schwartz teaches:
"collecting automatically every 45 days or less from an electronic portal of a Security and Exchange Commission (SEC), using a computer processor, one or more forms submitted to the SEC relating to a first organization; wherein: the one or more forms submitted to the SEC comprise SEC Form 10-K, SEC Form 8-K, SEC Form 10-Q, SEC Form 4, and SEC Form SD"
In Col 13 Line 17 - Col 14 Line 6, “In some embodiments, the portfolio monitoring engine 305 determines or track losses on portfolio securities on a day, week, month or other time basis. A minimum loss threshold can be specified to configure the portfolio monitoring engine 305 to flag only those securities having asynchronous activity that meet the loss threshold. Other criteria for flagging asynchronous activity can be configured”. These forms are disclosed in the EDGAR database, whose access is taught by Schwartz in Col 42 Lines 37-40, "the CIOR system may also be used to create variables from unstructured data by ingesting, for example, textural and voice data from various proprietary and publicly available data sources, including but not limited to EDGAR". EDGAR can be accessed through an official API that would be well known to those of ordinary skill in the art. Citing the attached non-patent literature Volume II: EDGAR Filing, Chapter 3: Index to Forms, we can find the disclosure of Form 10-K on page 3, 8-K on page 11, 10-Q on page 67, Form 4 on page 78, and Form SD on page 31.
Agarwal teaches:
"determining automatically, using the computer processor to run one or more machine learning models, if two or more thresholds have been exceeded indicating a risk of fraud”
In [0072], "Each model 218, 220 may output a score that, when compared to a predetermined threshold value at a given false positive tolerance level, indicates whether or not a merchant has been breached. Through ensembling, a merchant may be determined to be breached if either model 218, 220 indicates the merchant was breached. Alternatively, a merchant may be determined to be breached if both models 218, 220 indicate the merchant was breached. In non-limiting embodiments or aspects, the combined score of both models may be compared to a composite threshold, which may indicate whether or not a merchant has been breached". Note that plurality of scores generated by the models as well as the plurality of thresholds.
It would have been obvious to one having ordinary skill in the art at the effective filling date of the invention to apply the plurality of machine learning models as taught in Agrawal and apply that to the means for fraud detection as taught in Schwartz. Motivation to do so comes from the same rationale as outlined above with respect to Claim 1.
Schwartz teaches:
"notifying automatically an administrator in a second organization, using the computer processor, when one or more thresholds have been exceeded; and informing automatically the administrator, using the computer processor, with an identity of the one or more thresholds which have been exceeded"
"In Col 17 Lines 34-44, ""In some embodiments, notifications of potential claims (e.g., claims corresponding to asynchronous activities) can be generated and sent to clients via email and can also be displayed on the client dashboard. A notification of a potential claim can include information identifying whether the potential claim is related to a domestic or foreign exchange traded security, security identifier (e.g., CUSIP), relevant dates, amount of securities held, loss amounts and/or other relevant information. Examples of notifications of potential claim are provided below." In Col 6 Lines 42-45, "Examples of the users of the CIOR system 110 include clients or customers (e.g., customer asset manager), admin users and analyst users supporting the CIOR system 110".
With respect to additional limitations of Claim 18,
Schwartz teaches:
"a computer processor; a fraud detection engine comprising: a storage module; a cleaning module; a preprocessing module; a features extraction module; a machine learning module; wherein the computer processor is configured to run the fraud detection engine by performing steps comprising: collect automatically every 36 hours or less one or more forms submitted to a Security and Exchange Commission (SEC) relating to a first organization from an electronic portal of the SEC; wherein the one or more forms submitted to the SEC comprise SEC Form 10-K, SEC Form 8-K, SEC Form 10-Q, SEC Form 4, and SEC Form SD"
In Col 6 Lines 13-20 of Schwartz, "Embodiments of the CIOR system 110 can be implemented in distributed computing environments, where tasks or modules are performed by remote processing devices, which are linked through a communications network, such as a Local Area Network (“LAN”), Wide Area Network (“WAN”), or the Internet. In a distributed computing environment, program modules or subroutines may be located in both local and remote memory storage devices.” We consider the disclosure of the capabilities of this module to thus teach the individual modules. In Col 13 Line 17 - Col 14 Line 6, “In some embodiments, the portfolio monitoring engine 305 determines or track losses on portfolio securities on a day, week, month or other time basis. A minimum loss threshold can be specified to configure the portfolio monitoring engine 305 to flag only those securities having asynchronous activity that meet the loss threshold. Other criteria for flagging asynchronous activity can be configured”. These forms are disclosed in the EDGAR database, whose access is taught by Schwartz in Col 42 Lines 37-40, "the CIOR system may also be used to create variables from unstructured data by ingesting, for example, textural and voice data from various proprietary and publicly available data sources, including but not limited to EDGAR". EDGAR can be accessed through an official API that would be well known to those of ordinary skill in the art. Citing the attached non-patent literature Volume II: EDGAR Filing, Chapter 3: Index to Forms, we can find the disclosure of Form 10-K on page 3, 8-K on page 11, 10-Q on page 67, Form 4 on page 78, and Form SD on page 31. For machine learning, see Col 41 Lines 5-12 of Schwartz.
store automatically the one or more forms in the storage module
Looking at Figure 3 of Schwartz, we note the various databases that store collected information relevant to the analytical functions of the system.
Agarwal teaches:
determine automatically if a threshold indicating a risk of fraud is exceeded by using two or more machine learning models to analyze features from the features extraction module
In [0072], "Each model 218, 220 may output a score that, when compared to a predetermined threshold value at a given false positive tolerance level, indicates whether or not a merchant has been breached. Through ensembling, a merchant may be determined to be breached if either model 218, 220 indicates the merchant was breached. Alternatively, a merchant may be determined to be breached if both models 218, 220 indicate the merchant was breached. In non-limiting embodiments or aspects, the combined score of both models may be compared to a composite threshold, which may indicate whether or not a merchant has been breached". Note that plurality of scores generated by the models as well as the plurality of thresholds.
It would have been obvious to one having ordinary skill in the art at the effective filling date of the invention to apply the plurality of machine learning models as taught in Agrawal and apply that to the means for fraud detection as taught in Schwartz. Motivation to do so comes from the same rationale as outlined above with respect to Claim 1.
Claim 2
As to Claim 2, Schwartz combined with Yan, Plehn-Dujowich, and Agrawal teaches all the limitations of Claim 1 as discussed above.
Agrawal teaches:
The method of claim 1, wherein: exceeding a threshold when running a liquid, solvency, and profitability ratio classification machine learning model
Above, we have noted that we extract data relating to a multitude of features, each of which corresponds to a machine learning model. Based on the technique of using a plurality of machine learning models according to configurable logic for combining their individual thresholds as taught in Agarwal in [0072], we consider this limitation disclosed as one such arrangement of the models in Agarwal, in which each particular model focuses on a given feature.
indicates where the liquid, solvency, and profitability ratio classification misguides the public to make incorrect conclusions about an ability of an organization to operate
In [0074] of Agarwal, it is described that we can convey more granular information regarding the details of various threats and suspicious metrics, and in light of the breadth of “indicates where”, we understand disclosing the particular suspicious metrics as disclosing this limitation, “With further reference to FIG. 2, and in non-limiting embodiments or aspects, the services 225 may be integrated with additional features for user interaction 233. For example, merchant scores and other metrics of data breaches may be sorted/filtered in display interfaces for users (e.g., security personnel), to facilitate prioritization 234 of cases of merchant breach. When assigned to a security personnel, the scores may be integrated with case disposition 236 systems, such as an issue ticketing and resolution system. The scores may also be incorporated into any number of graphs and charts 238, to represent various metrics including, but not limited to, fraud risk of a merchant over time, number of financial device declines over time, number of financial device declines that are indicative of fraud over time, number of financial devices with security testing transaction activity, days since breach occurred and/or was detected, estimated value of fraudulent transaction activity, ratio/percentage of PANs exhibiting fraudulent behavior (see FIG. 4 for a more detailed description), and/or the like”. In [0045] of Yan, "In step S27, the determining and sorting module 15 is configured to determine any ones of the stocks, whose analysis data matches predetermined selection criteria, as the target stocks…. In other embodiments, the predetermined selection criteria are further associated with fundamental indices data as indicated in Table 2, and technical indices based on trading price or trading volume as indicated in Table 3". Citing Table 2 of Yan, "Acid-test ratio(quick ratio)...Cost of capital for operations current (liquid) assets to total liability current (liquid) assets to total assets current (liquid) assets turnover rate current liabilities turnover current liability to total liability current liability to equity current liability to inventory Current ratio..."Gross Margin Growth gross margin of sales Gross profit margin(gross profit ratio/margin) Gross profit ratio/margin Gross profit margin Gross profit/gross loss...Debt ratio debt to capital ratio debt to equity ratio Debt to total assets" are disclosed.
exceeding a threshold when running a disclosure classification machine learning model
Above, we have noted that we extract data relating to a multitude of features, each of which corresponds to a machine learning model. Based on the technique of using a plurality of machine learning models according to configurable logic for combining their individual thresholds as taught in Agarwal in [0072], we consider this limitation disclosed as one such arrangement of the models in Agarwal, in which each particular model focuses on a given feature.
indicates a detection of one or more ambiguous disclosures;
In Col 3 Lines 5-9 of Schwartz, "In embodiments, unstructured data may include social media data, custodian data, audio recordings, financial reports, or some other type of unstructured data. Financial reports may include Securities and Exchange Commission reports". In Col 43 Lines 22-31, "In embodiments of the aggregated use of structured and unstructured data, that is cleansed and made accurate, may allow comparison of more variables for better insights to predict possible actionable securities fraud litigation that inevitably trails disclosure of the deviations and failures, and to predict the probability of instances in which an issuer's stock price is likely inflated due to failures to disclose deviations from expected behaviors or regulatory violations before those deviations or failures are publicly disclosed".
exceeding a threshold when running a sentiment analysis machine learning model
Above, we have noted that we extract data relating to a multitude of features, each of which corresponds to a machine learning model. Based on the technique of using a plurality of machine learning models according to configurable logic for combining their individual thresholds as taught in Agarwal in [0072], we consider this limitation disclosed as one such arrangement of the models in Agarwal, in which each particular model focuses on a given feature.
indicates a detection of one or more erroneous statements about the organization;
In Col 39 Lines 42-49, " Unstructured text data may be analyzed using cognitive analytics to extract meaningful information such as, for example, entities (e.g., name of security, regulatory authority, name of the court, related parties) keywords (e.g., fraud, police, crime), tone (e.g., joy, anger sadness, fear, disgust) and sentiment (e.g., positive, negative) and key concept (e.g., insider, legal, investigation)".
exceeding a threshold when running an anomaly detection classification machine learning model
Above, we have noted that we extract data relating to a multitude of features, each of which corresponds to a machine learning model. Based on the technique of using a plurality of machine learning models according to configurable logic for combining their individual thresholds as taught in Agarwal in [0072], we consider this limitation disclosed as one such arrangement of the models in Agarwal, in which each particular model focuses on a given feature.
indicates a detection of one or more anomalies;
In Col 45 Lines 13-17, "In embodiments, one element of the overall DAA process may be to monitor incoming data from a filer and identify any anomalies or inconsistencies. The DAA import and posting process may include numerous processes to identify anomalies and report them.".
exceeding a threshold when running an ownership analysis classification machine learning model
Above, we have noted that we extract data relating to a multitude of features, each of which corresponds to a machine learning model. Based on the technique of using a plurality of machine learning models according to configurable logic for combining their individual thresholds as taught in Agarwal in [0072], we consider this limitation disclosed as one such arrangement of the models in Agarwal, in which each particular model focuses on a given feature.
indicates a detection of one or more suspicious owners;
In [0078] of Plehn-Dujowich, "Clearly identifying corporate officers and directors of branches in a clear format may be helpful for performing meaningful investment analysis. The ability to see branch relationships, as displayed in FIG. 21, and the officers who run these entities helps to lift the “corporate veil” of ownership. This information is key to provide parent relationships, status and permit in depth background checks of officers".
and exceeding a threshold when running an ESG disclosure classification machine learning model
Above, we have noted that we extract data relating to a multitude of features, each of which corresponds to a machine learning model. Based on the technique of using a plurality of machine learning models according to configurable logic for combining their individual thresholds as taught in Agarwal in [0072], we consider this limitation disclosed as one such arrangement of the models in Agarwal, in which each particular model focuses on a given feature.
indicates a detection of one or more fraudulent ESG disclosures are detected.
In Col 43 Lines 22-31, "In embodiments of the aggregated use of structured and unstructured data, that is cleansed and made accurate, may allow comparison of more variables for better insights to predict possible actionable securities fraud litigation that inevitably trails disclosure of the deviations and failures, and to predict the probability of instances in which an issuer's stock price is likely inflated due to failures to disclose deviations from expected behaviors or regulatory violations before those deviations or failures are publicly disclosed". In Col 40 Lines 52-54 of Schwartz, "Data used in the CIOR analytics platform may also include environment, sustainability and governance data (or “ESG data”), which may be used to refine analyses run on the platform.
It would have been obvious to one having ordinary skill in the art at the effective filling date of the invention to apply the plurality of machine learning models as taught in Agrawal and the mechanisms for performing financial analysis of Yan and Plehn-Dujowich to the means for fraud detection as taught in Schwartz. Motivation to do so can be found in the rationale outlined with respect to Claim 1.
Claims 13, 19 is rejected as disclosing substantially similar limitations as Claim 2.
Claim 4
As to Claim 4, Schwartz teaches all the limitations of Claim 1 as discussed above.
Schwartz teaches:
The method of claim 1, wherein the administrator is part of the organization.
In Col 6 Lines 42-44, "Examples of the users of the CIOR system 110 include clients or customers (e.g., customer asset manager)". Given the existence of ETFs, or exchange tradeable funds who would have public disclosures, we consider this limitation to be disclosed as the customer asset manager would support the manager of such a fund.
Claim 5
As to Claim 5, Schwartz combined with Agrawal, Yan, and Plehn-Dujowich teaches all the limitations of Claim 1 as discussed above.
Schwartz teaches:
The method of claim 1, wherein: the organization is a first organization; and the administrator is part of a second organization.
In Col 6 Lines 42-45, "Examples of the users of the CIOR system 110 include clients or customers (e.g., customer asset manager), admin users and analyst users supporting the CIOR system 110".
Claim 6
As to Claim 6, Schwartz combined with Agrawal, Yan, and Plehn-Dujowich teaches all the limitations of Claim 1 as discussed above.
Schwartz teaches:
The method of claim 1, wherein the electronic portal is a portal of a Securities and Exchange Commission (SEC).
In Col 3 Lines 5-10, "In embodiments, unstructured data may include social media data, custodian data, audio recordings, financial reports, or some other type of unstructured data. Financial reports may include Securities and Exchange Commission reports".
Claim 7
As to Claim 7, Schwartz combined with Agrawal, Yan, and Plehn-Dujowich teaches all the limitations of Claim 6 as discussed above.
Schwartz teaches:
The method of claim 6, wherein the one or more forms comprise SEC Form 10-K, SEC Form 8-K, SEC Form 10-Q, SEC Form 4, and SEC Form SD.
These forms are disclosed in the EDGAR database, whose access is taught by Schwartz in Col 42 Lines 37-40, "the CIOR system may also be used to create variables from unstructured data by ingesting, for example, textural and voice data from various proprietary and publicly available data sources, including but not limited to EDGAR". EDGAR can be accessed through an official API that would be well known to those of ordinary skill in the art. Citing the attached non-patent literature Volume II: EDGAR Filing, Chapter 3: Index to Forms, we can find the disclosure of Form 10-K on page 3, 8-K on page 11, 10-Q on page 67, Form 4 on page 78, and Form SD on page 31.
Claim 8
As to Claim 8, Schwartz combined with Agrawal, Yan, and Plehn-Dujowich teaches all the limitations of Claim 1 as discussed above.
Schwartz teaches:
The method of claim 1, wherein the electronic portal is an Electronic Data Gathering, Analysis, and Retrieval (EDGAR) database.
In Col 42 Lines 37-40, "the CIOR system may also be used to create variables from unstructured data by ingesting, for example, textural and voice data from various proprietary and publicly available data sources, including but not limited to EDGAR".
Claim 14
As to Claim 14, Schwartz combined with Agrawal, Yan, and Plehn-Dujowich teaches all the limitations of Claim 12 as discussed above.
Schwartz teaches:
The method of claim 12, wherein collecting automatically the one or more forms from the electronic portal occurs every 36 hours or less.
In Col 13 Line 17 - Col 14 Line 6, “In some embodiments, the portfolio monitoring engine 305 determines or track losses on portfolio securities on a day, week, month or other time basis.
Claim 15
As to Claim 15, Schwartz combined with Agrawal, Yan, and Plehn-Dujowich teaches all the limitations of Claim 12 as discussed above.
Schwartz teaches:
The method of claim 12, wherein the first organization and the second organization are different organizations.
In Col 6 Lines 42-45, "Examples of the users of the CIOR system 110 include clients or customers (e.g., customer asset manager), admin users and analyst users supporting the CIOR system 110".
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to THEODORE L XIE whose telephone number is (571)272-7102. The examiner can normally be reached M-F 9-5.
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, Rutao Wu can be reached at 571-272-6045. 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.
/THEODORE XIE/Examiner, Art Unit 3623
/CHARLES GUILIANO/Primary Examiner, Art Unit 3623
Prosecution Insights