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 .
DETAILED ACTION
1. This office action is in response to an amendment received on 3/19/26.
2. Claims 1-24 are pending.
3. The SUBJECT MATTER ELIGIBILITY DECLARATION (SMED) under 37 CFR 1.132 filed on 3/19/26 is insufficient to overcome the rejection of claims 1-24 based upon the claims rejected under 35 U.S.C 101 as directed to an abstract idea without significantly more as set forth in the last Office action because:
Examiner summarizes each issue and provides a response.
Issue#1
I believe that the invention of independent Claims 1 and 13 of the '611 application
provides technological improvements to cohort-based machine learning (ML)
models by reducing their cohort-specific biases, and also provides technical
improvements over non-cohort based ML models by providing for better
discriminative performance. The cohort-specific biases that are being reduced by the invention of Claim 1 of the '611 application are statistical biases, not societal and ethical biases.
The cohort-based ML modeling discussed herein and in the '611 application does
not take into account any of the gender, race and ethnicity of the retailers, and
thus, the cohort-specific biases discussed herein and in the '611 application that
are inadvertently introduced are not ethical or societal problems, but rather, the
cohort-specific biases discussed herein and in the '611 application are solely
technical problems. Indeed, it would be unlawful to take into account the gender,
race and ethnicity of retailers to adjust the credit limits of the retailers.
As recited in Claim 1 of the '611 application, each cohort, of the plurality of cohorts
referred to in Claim 1 of the '611 application, has a different respective cohort
snapshot date within a specified period of time, and each cohort, of the plurality of
cohorts, includes a group of the retailers that made at least one of the product
orders within a specified window of time following the cohort snapshot date. This
makes it clear that the retailers are grouped into the different cohorts based solely on when the retailers made product orders, not based on anything related to the gender, race and ethnicity of the retailers, which makes it clear that any biases that are introduced by the cohort-based ML models have nothing to do with ethical or societal problems.
Examiner Response
Examiner disagrees that claims 1&13 are providing an improvement to cohort-based machine learning (ML) models.
Examiner agrees that the cohort specific biases recited in claims 1&13 are statistical biases.
Examiner submits that cohort-based ML-models create statistical biases (the cohort biased ML-model is operating in its ordinary capacity).
The claimed invention is not using the ML-models to reduce statistical bias.
The claimed invention is using a computer implemented method implementing a plurality of machine learning models, recited at a high level of generality, for implementing the steps of the identified abstract idea.
Issue#2
10. As explained in paragraph [0055] of the '611 application, for which I am the sole
inventor, the performance of certain types of ML models (e.g., classification
models) can be measured as an Area Under the Curve (AUC), where the AUC
equals 1 when the ML model performs perfectly, and the worst possible score is
when the AUC equals 0, and the performance of other types of ML models (e.g.,
regression models) can be measured as a coefficient of determination (R²), where
the R² equals 1 when the ML model performs perfectly. I believe that the invention of independent Claims 1 and 13 of the '611 application also provides technological improvements to ML models by improving the discriminative power (also known as discriminative performance) of ML models, including increasing the discriminative power of ML classification models measured in terms of an Area Under the Curve (AUC), and increasing coefficient of determination (R²) scores for ML regression models.
Examiner Response
Examiner respectfully disagrees.
The determination of “lower cohort specific bias and higher coefficient of determination (R squared) score” is disclosed in the specification, however this is not an improvement to machine learning technology.
Examiner submits calculating the area under a curve is integral calculus, executed by a computer to solve a math problem.
Furthermore, calculating the coefficient of determination score (R squared) is a
a commonly understood mathematical formula, used by machine learning models, where the machine learning model is recited at a high level of generality and being used in its ordinary capacity as a tool to implement the steps of the identified abstract idea.
Therefore, there are no additional elements in the claims that are indicative of integration into a practical application.
The rejection is maintained.
Issue#3
16. To address the above described technical problems that may arise when using a single cohort-based ML model, which problems related to cohort-specific biases, I conceived of a meta ML model structure described and claimed in the '611 application that includes cohort-based ML aggregation models (labeled 604, 704, and 804 in the '611 application). Since each cohort differs from other cohorts, this approach captures more general patterns across multiple cohorts, rather than patterns specific to a single cohort. Additionally, when different cohorts contain at least some overlapping members, the ML aggregation models are better able to track their behaviors over extended periods, allowing the cohort-based ML models to better account for seasonality and cyclicality of various profiles.
18. By using the ML aggregation models recited in independent Claims 1 and 13 of
the '611 application, generalizable patterns from each cohort are reinforced while
cohort-specific patterns are filtered out, which explains why the meta ML model
structure with multiple cohort-based ML models outperforms a single cohort-based ML model when projecting outcomes for previously unseen retailers during untrained time periods.
19. As explained in paragraph [0054] of the '611 application, for which I am the sole
inventor, "[t]o reduce the probability of undesirable cohort-specific biases, while
keeping the benefits of cohort-based modeling, within each of the sets of models
501, 502, and 503 multiple cohort based ML scoring models are defined (e.g., 602-
1 through 602-6, 702-1 through 702-6, and 802-1 through 802-6) and a meta model
is defined (i.e., the cohort-based ML aggregation models 604, 704, and 804), to
generate periodic (e.g., daily) respective aggregated scores."
20. As explained in paragraph [0055] of the '611 application, for which I am the sole
inventor, "[a] benefit of using multiple cohort-based ML scoring models, within each
of the sets of models 501, 502, and 503, is that even if one of the cohort-based ML
scoring models is performing poorly, overall the models will perform well when their
scores are aggregated. More specifically, each cohort-based ML aggregation
model (604, 704, and 804) will have a better model performance than each of the
individual cohort-based ML scoring models (e.g., 602-1 through 602-6, 702-1
through 702-6, and 802-1 through 802-6) of the set of models (e.g., 501, 502, or
503) that the cohort-based ML aggregation model is within."
Examiner Response
Examiner respectfully disagrees.
It can be seen from the instant specification paras cited, that there is no technical explanation of the asserted improvement (the use of cohort- based ML models) and reflected in the claims. The additional elements( A plurality of sets of machine learning models that include a set of risk ML models, a set of risk ML models, a set of non-defaulter value ML models, a set of defaulter value ML models and wherein each of the sets of the ML models, includes at least two cohort-based ML scoring models and a cohort-based ML aggregation model) are recited at a high level of generality, operating in their ordinary capacity and as such are being used as a tool to implement the identified abstract idea.
Issue#4
22. By using the invention claimed in Claim 1 of the '611 application, for which I am
the sole inventor, the performance (measured in terms of AUC) improved by about
10% compared to the use of the single ML model that Faire Wholesale, Inc. had
previously been using to adjust credit limits of retailers.
I estimate that use of the ML aggregation models recited in independent Claims 1
and 13 of the '611 application, for which I am the sole inventor, contributed to about
2% to 3% of the improvement in the performance (measured in terms of AUC).
24. Based on the above, I believe that one of ordinary skill in the art at the time of filing
the '611 application would understand the claimed invention to provide a significant
technical improvement in the field of ML models.
Examiner Response
Examiner respectfully disagrees.
There is no technical improvement in the field of ML models.
This issue has been addressed above with respect to Examiner Response to Issues#1-3.
RESPONSE TO ARGUMENTS
Applicant argues#1
In the previous Responses filed on October 29, 2025, and June 16, 2025,
Applicant asserted that the claimed invention provides technological improvements to
technology or to computer functionality, including technical improvements to ML models,
and even more specifically to cohort-based ML models, by reducing their cohort-specific
biases, increasing their discriminative power, and increasing their coefficient of
determination (R²) scores. Applicant included similar and further arguments in the
previous Response filed on September 19, 2025, and these further arguments were
addressed to some extent in the Advisory Action issued on October 1, 2025. In the
Advisory Action issued on October 1, 2025, it was asserted that "biases inherent to cohort- based machine learning (ML) design are not technical problems" and that to "solve these biases is an ethical and societal problem (and is part of the identified abstract idea), and it is not a technical problem." Applicant respectfully disagrees with the assertion in the Advisory Action that the biases inherent to cohort-based ML design are not technical problems, but rather are ethical and societal problems. While biases in ML models could potentially be in part to ethical or social issues, biases in ML models are also deeply technical problems that can arise at multiple points in data and model pipelines. More specifically, cohort-based biases from a machine learning perspective can arise because of how data provided to ML models is collected, structured, and/or used by the ML models.
Submitted herewith is a Subject Matter Eligibility Declaration (SMED) under 37
C.F.R. 1.132 from the sole inventor of the present application. The SMED further clarifies why the cohort specific biases that are reduced using the claimed invention is a technical problem, not a societal or ethical problem. Further, the SMED further clarifies that the claimed invention improves ML model performance, and thereby, provides a technical improvement, including by providing for better discriminative performance.
Examiner Response
Examiner respectfully disagrees.
This argument has been addressed above, see Examiner Response to Issues#1-2.
The rejection is maintained.
Applicant argues#2
Applicant asserts that there is a nexus between the invention as claimed in
independent claims 1 and 13 and the evidence provided in the SMED. Further, Applicant believes that the SMED further clarifies how and why claimed invention is an eligible improvement in machine learning (ML) technology, and asserts that the specification provides sufficient details such that one of ordinary skill in the art would recognize the claimed invention as providing a technological improvement and the claims reflect the disclosed improvement.
The SMED provides facts that describe the state of the art at the time of filing the
present application, provide objective evidence as to how the invention improved upon
the state of the art, and provides a factual basis for determining that one of ordinary skill
in the art would have concluded that the invention improved the underlying technology.
Applicant respectfully asserts, as noted above, that the SMED further clarifies how
and why claimed invention is an eligible improvement in ML technology. For example, the SMED clarifies how and why the invention claimed in independent Claims 1 and 13 of the '611 provide improvements over prior uses of a single non-cohort based ML model, as well as over use of a single cohort-based ML model, including by providing for better discriminative performance.
Further, the SMED clarifies that the cohort-specific biases that are being reduced
by the invention of independent Claims 1 and 13 of the '611 application are statistical
biases, not societal and ethical biases.
Examiner Response
This issue has been addressed above with respect to Examiner Response to Issues#1-3.
The rejection is maintained.
Applicant argues#3
In the previous Response filed on October 29, 2025, Applicant explained that a
search of the USPTO's database revealed that there are numerous issued U.S. patents
that attempt to address technical problems related to ML biases, but in different manners than the claimed embodiments in the present application, and Applicant included a list of a half-dozen such issued U.S. patents. In the Office Action issued on December 19, 2025, it was asserted that "each allowed patent stands on its own merits, and the reasons for allowance are specific to the facts of the case," and that "[a]s was explained previously, solving cohort specific bias is a societal and ethical problem, and as such is part of the identified abstract idea, and it is not a technical problem." While Applicant understands that the reasons for allowance in each patent are specific to the facts of the case, Applicant still believes the USPTO having previously issued numerous patents that attempt to address reducing statistical biases of ML models is probative evidence that such biases are a technical problem, not a societal and ethical problem, and that overcoming such a problem is a technical solution. Further, just as prior caselaw having similar facts to a case being decided by a court can serve as persuasive authority, the USPTO having issued numerous patents attempting to address similar problems to those being addressed by the presently claimed invention can also serve as persuasive authority.
Examiner Response
This issue has been addressed above with respect to Examiner Response to Issues#1 above.
Examiner reiterates that each allowed patent stands on its own merits, and the reasons for allowance are specific to the facts of the case.
Examiner has stated above, that the cohort specific biases are statistical biases, see the Response to Issue #1.
However the claims of the instant invention are not using the ML models to reduce statistical biases.
The claimed invention using a computer implemented method implementing a plurality of machine learning models, recited at a high level of generality, for implementing the steps of the identified abstract idea.
The rejection is maintained.
Applicant argues#4
As explained in the previously filed Response, there are technical problems with
using cohort-based modeling in machine learning, some of which relate to the biases
inherent to cohort-based ML design. As explained in paragraph [0054] of the present
application "[a] potential disadvantage of using cohort-based modeling to model the behaviors of retailers is that because different cohorts of retailers may behave inconsistently, cohort-based modeling may inadvertently introduce cohort-specific biases, which is undesirable." Other technical problems with using cohort-based ML modeling is that individual cohort-based ML models can perform poorly, which can cause individual cohort-based ML models to generate inaccurate predictions. Additional technical problems with using cohort-based modeling in machine learning is that a single cohort-based ML model may overfit to characteristics specific to a single cohort while failing to capture more generally applicable patterns, which can also cause the model to generate inaccurate predictions.
As explained in the previously filed Response, to address the above described
technical problems that may arise when using cohort-based ML models, the inventor of
the present application conceived of a meta ML model structure that includes the cohort- based ML aggregation models 604, 704, and 804 in the present application. Since each cohort differs from others, this approach captures more general patterns across multiple cohorts, rather than patterns specific to a single cohort. Additionally, when different cohorts contain at least some overlapping members (e.g., retailers), the ML aggregation models are better able to track their behaviors over extended periods, allowing the ML models to better account for seasonality and cyclicality of various profiles. To illustrate this concept, think of the meta ML model structure as a committee of individual cohort- based ML models that can each make predictions about a same retailer from a different perspective while following consistent guiding principles. Each ML model can evaluate numerous retailers and learn to assess performance based on behaviors from various time periods. When making a decision or prediction, the cohort-based ML models can be thought of as voting to reach a joint conclusion. Although individual cohort-based ML models may have statistical biases that stem from accessing information from only certain respective cohorts and training on specific time periods, these limited-data statistical biases beneficially tend to cancel each other out when aggregated using the claimed invention, and their common conclusions become stronger when using the claimed invention. By using the ML aggregation models of the claimed invention, generalizable patterns from each cohort are reinforced while cohort-specific patterns are filtered out. This explains why the meta ML model structure with multiple cohort-based ML models outperforms a single ML model when projecting outcomes for previously unseen retailers during untrained time periods. These are some of the reasons that Applicant asserts that the claimed invention is clearly improving upon technical problems that may arise when using ML models in generally, and that especially may arise when using cohort-based ML models.
Consistent with MPEP 2106.05(a), the disclosure of the present application,
including the disclosure in paragraphs [0054]-[0055] of the present application discussed above, and the disclosure in paragraph [0039]-[0041] of the present application (which was quoted and discussed in the previous Response filed on June 16, 2025), "provides sufficient details such that one of ordinary skill in the art would recognize the claimed invention as providing an improvement." Additionally, consistent with MPEP 2106.05(a), "[a]n indication that the claimed invention provides an improvement" is clearly included in the present application and the present application also includes "a discussion in the specification that identifies a technical problem and explains the details of an unconventional technical solution expressed in the claim, or identifies technical improvements realized by the claim over the prior art" (e.g., see paragraphs [0039]-[0041] and [0055]-[0056] of the present application).
Examiner Response
Examiner respectfully disagrees.
Paras 38-41, and 55-56 are reproduced below:
[0038] Referring again to FIG. 3, step 307 involves selecting a retailer for whom their respective credit limit is to be analyzed, using ML models, for the purpose of determining whether their credit limit should be adjusted. For example, step 307 can involve selecting a retailer that has utilized the B2B marketplace platform (e.g., 212) to make product orders during the specified period of time (e.g., within the past six months, or within the past year).
[0039] Step 308 involves using the at least two (e.g., six) cohort-based ML scoring models 602 of the set of risk ML models 501 to produce at least two (e.g., six) risk scores. More specifically, at step 308, each of the cohort-based ML scoring models 602 of the set of risk ML models 501 produces a respective risk score indicative of a predictive risk that the retailer will default on at least one product order made within a further specified window of time (e.g., within the next month, or the next two months). Step 310 involves using the cohort-based ML aggregation model 604 of the set of risk ML models 501 to aggregate the at least two (e.g., six) risk scores, generated by the cohort-based ML scoring models 602, into an aggregated risk score indicative of a predictive risk that the retailer will default on at least one product order made within a further specified window of time (e.g., within the next month, or the next two months). Beneficially, on the test cohort, the aggregated risk score generated by the cohort-based ML aggregation model 604 has a lower cohort-specific bias and a better discriminative power than any individual one of the individual risk scores produced by the at least two (e.g., six) cohort-based ML scoring models 602 of the set of risk ML models 501.
[0040] Step 312 involves using the at least two (e.g., six) cohort-based ML scoring models 702 of the set of non-defaulter value ML models 502 to produce at least two non- defaulter value scores. More specifically, at step 312, each of the cohort-based ML scoring models 602 of the set of non-defaulter value ML models 502 produces a respective non- defaulter risk score indicative of a predicted value of one or more orders the retailer will make if the retailer does not default on any product orders within the further specified window of time. Step 314 involves using the cohort-based ML aggregation model 704 of the set of non-defaulter value ML models 502 to aggregate the at least two non-defaulter value scores into an aggregated non-defaulter value score indicative of a predicted value of one or more orders the retailer will make if the retailer does not default on any product orders within the further specified window of time. Beneficially, on the test cohort, the aggregated non-defaulter value score generated by the cohort-based ML aggregation model 704 has a lower cohort-specific bias and a higher coefficient of determination (R²) score than any individual one of the individual non-defaulter value scores produced by the at least two (e.g., six) cohort-based ML scoring models 702 of the set of non-defaulter value models 502.
[0041] Step 316 involves using the at least two (e.g., six) cohort-based ML scoring models 802 of the set of defaulter value ML models 503 to produce at least two defaulter value scores. More specifically, at step 316, each of the cohort-based ML scoring models 602 of the set of defaulter value ML models 503 produces a respective defaulter value score indicative of a predicted value of one or more orders the retailer will make if the retailer does default on at least one product order within the further specified window of time. Step 318 involves using the cohort-based ML aggregation model 804 of the set of defaulter value ML models 503 to aggregate the at least two defaulter value scores into an aggregated defaulter value score indicative of a predicted value of one or more orders the retailer will make if the retailer does default on at least one product order made within the further specified window of time. Beneficially, on the test cohort, the aggregated defaulter value score generated by the cohort-based ML aggregation model 804 has a lower cohort-specific bias and a higher coefficient of determination (R²) score than any individual one of the individual defaulter value scores produced by the at least two (e.g., six) cohort-based ML scoring models 802 of the set of defaulter value models 503.
[0055] A benefit of using multiple cohort-based ML scoring models, within each of the sets of models 501, 502, and 503, is that even if one of the cohort-based ML scoring models is performing poorly, overall the models will perform well when their scores are aggregated. More specifically, each cohort-based ML aggregation model (604, 704, and 804) will have a better model performance than each of the individual cohort-based ML scoring models (e.g., 602-1 through 602-6, 702-1 through 702-6, and 802-1 through 802- of the set of models (e.g., 501, 502, or 503) that the cohort-based ML aggregation model is within. For example, the performance of certain types of ML models (e.g., classification models) can be measured as an Area Under the Curve (AUC), where the AUC equals 1 when the ML model performs perfectly, and the worst possible score is when the AUC equals 0. The performance of other types of ML models (e.g., regression models) can be measured as a coefficient of determination (R²), where the R² equals 1 when the ML model performs perfectly. Continuing with this example, the cohort-based ML aggregation model 604, of the set of risk models 501, will have a greater testing cohort AUC than any individual one of the cohort-based ML scoring models 602-1 through 602- 6; the cohort-based ML aggregation model 704, of the set of non-defaulter value models 502, will have a greater testing cohort coefficient of determination (R²) score than any individual one of the cohort-based ML scoring models 702-1 through 702-6; and the cohort-based ML aggregation model 804, of the set of defaulter value models 503, will have a greater testing cohort coefficient of determination (R²) score than any individual one of the cohort-based ML scoring models 802-1 through 802-6.
[0056] In accordance with certain embodiments, when each of the cohort-based ML aggregation models 604, 704, and 804 is being trained, one or more cohort-based ML scoring models whose scores the aggregation model aggregates (e.g., one or more of the cohort-based ML scoring models 601-1 through 602-6, one or more of the cohort- based ML scoring models 701-1 through 702-6, and/or one or more of the cohort-based ML scoring models 801-1 through 802-6) may be dropped during a aggregation stage, if the cohort-based ML aggregation model doesn't benefit from those dropped models, so long as at least one cohort-based training ML models remains within each of the sets of models 501, 502, 503. In accordance with certain embodiments of the present technology, each of the ML models described herein are serialized objects stored in the cloud.
It can be seen from the instant specification that there is no technical explanation of the asserted improvement (the use of cohort- based ML models) and reflected in the claims. The additional elements( A plurality of sets of machine learning models that include a set of risk ML models, a set of risk ML models, a set of non-defaulter value ML models, a set of defaulter value ML models and wherein each of the sets of the ML models, includes at least two cohort-based ML scoring models and a cohort-based ML aggregation model) are recited at a high level of generality, operating in their ordinary capacity and as such are being used as a tool to implement the identified abstract idea (steps for adjusting a respective credit limit for one or more retailers).
The determination of “lower cohort specific bias and higher coefficient of determination (R squared) score” is disclosed in the specification, however this is not an improvement to machine learning technology.
Examiner submits calculating the coefficient of determination score (R squared) is a
a commonly understood mathematical formula, used by machine learning models, where the machine learning model is recited at a high level of generality and being used in its ordinary capacity as a tool to implement the steps of the identified abstract idea.
Therefore, there are no additional elements in the claims that are indicative of integration into a practical application.
The rejection is maintained.
Applicant argues#5
In view of the SMED, the arguments presented above, and the arguments
previously presented in the Responses filed on October 29, 2025, and June 16, 2025,
Applicant respectfully requests that the rejection of the claims under 35 U.S.C. 101 be
reconsidered and withdrawn.
Examiner Response
Examiner respectfully disagrees.
See the Response to Isssue#1-4 above, and the Response to Applicant argues#1-4 above.
The rejection is maintained.
Claim Rejections- 35 U.S.C § 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.
1. Claims 1-24 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Claims 1, 13 are directed to a system, method and system which are statutory categories of invention. (Step 1: YES).
Representative claim 1 recites the limitations of:
A computer implemented method for using machine learning to adjust a respective credit limit of one or more retailers that utilize a business-to-business (B2B) marketplace platform to make product orders, the method comprising:
collecting retailer data for retailers that have utilized the B2B marketplace platform to make product orders during a specified period of time;
identifying a plurality of cohorts based on the retailer data;
wherein each cohort, of the plurality of cohorts, has a different respective cohort snapshot date within the specified period of time, and
wherein each cohort, of the plurality of cohorts, includes a group of the retailers that made at least one of the product orders within a specified window of time following the cohort snapshot date;
producing, based on the retailer data, a plurality of sets of machine learning (ML) models that include a set of risk ML models, a set of non-defaulter value ML models, and a set of defaulter value ML models, wherein each of the sets of the ML models includes a plurality of cohort-based ML models, and wherein each of the plurality of cohort-based ML models, of each of the sets of the ML models, includes at least two cohort-based ML scoring models, and a cohort-based ML aggregation model;
wherein each of the plurality of cohort-based ML models, of each of the sets of the ML models, corresponds to a different cohort of the plurality of cohorts;
for each retailer, of at least some of the retailers that have utilized the B2B marketplace platform to make product orders during the specified period of time, using the at least two cohort-based ML scoring models of the set of risk ML models to produce at least two risk scores, each having a respective cohort specific bias and
using the cohort-based ML aggregation model of the set of risk ML models to aggregate the at least two risk scores into an aggregated risk score having a cohort-specific bias that is lower than any individual one of the at two risk scores;
using the at least two cohort-based ML scoring models of the set of non-defaulter value ML models to produce at least two non-defaulter value scores, each having a respective cohort-specific bias and using the cohort-based ML aggregation model of the set of non-defaulter value ML models to aggregate the at least two non-defaulter value scores into an aggregated non-defaulter value score having a cohort- specific bias that is lower than any individual one of the at least two non-defaulter value scores;
using the at least two cohort-based ML scoring models of the set of defaulter value ML models to produce at least two defaulter value scores, each having a respective cohort-specific bias, and using the cohort-based ML aggregation model of the set of defaulter value ML models to aggregate the at least two defaulter value scores into an aggregated defaulter value score having a cohort-specific bias that is lower than any individual one of the at least two defaulter value scores; and
determining whether to adjust the credit limit of the retailer based on the aggregated risk score, the aggregated non-defaulter value score, and the aggregated defaulter value score for the retailer, that were produced using the plurality of sets of ML models.
These limitations, under their broadest reasonable interpretation, cover performance of the limitation as certain methods of organizing human activity.
The claim recites elements that are in bold above, which covers performance of the limitation as a commercial interaction, steps for adjusting a respective credit limit for one or more retailers (e.g., to adjust a respective credit limit of one or more retailers that utilize a business-to-business (B2B) marketplace platform to make product orders, collecting retailer data for retailers that have utilized the B2B marketplace platform to make product orders during a specified period of time;
identifying a plurality of cohorts based on the retailer data; wherein each cohort, of the plurality of cohorts, has a different respective cohort snapshot date within the specified period of time, and wherein each cohort, of the plurality of cohorts, includes a group of the retailers that made at least one of the product orders within a specified window of time following the cohort snapshot date;
for each retailer, of at least some of the retailers that have utilized the B2B marketplace platform to make product orders during the specified period of time, to produce at least two risk scores, each having a respective cohort specific bias, to aggregate the at least two risk scores into an aggregated risk score having a cohort-specific bias that is lower than any individual one of the at two risk scores; to produce at least two non-defaulter value scores, each having a respective cohort-specific bias and to aggregate the at least two non-defaulter value scores into an aggregated non-defaulter value score having a cohort- specific bias that is lower than any individual one of the at least two non-defaulter value scores; to produce at least two defaulter value scores, each having a respective cohort-specific bias, and to aggregate the at least two defaulter value scores into an aggregated defaulter value score having a cohort-specific bias that is lower than any individual one of the at least two defaulter value scores; and
determining whether to adjust the credit limit of the retailer based on the aggregated risk score, the aggregated non-defaulter value score, and the aggregated defaulter value score for the retailer, that were produced)
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation as a Commercial Interaction, then it falls within the “Certain Methods of Organizing Human Activity” grouping of abstract ideas.
Claims 13 is abstract for similar reasons.
(Step 2A-Prong 1: YES. The claims are abstract).
This judicial exception is not integrated into a practical application. Limitations that are not indicative of integration into a practical application include: (1) Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea (MPEP 2106.05.f), (2) Adding insignificant extra solution activity to the judicial exception (MPEP 2106.05.g), (3) Generally linking the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05.h).
Claims 1, 13 includes the following additional elements:
-A plurality of sets of machine learning models that include a set of risk ML models, a set of risk ML models, a set of non-defaulter value ML models, a set of defaulter value ML models and wherein each of the sets of the ML models, includes at least two cohort-based ML scoring models and a cohort-based ML aggregation model
-A data store
-One or more processors
The plurality of sets of machine learning models, the data store, the one or more processors and each of the plurality of cohort-based ML models, of each of the sets of the ML models and the cohort based ML aggregation model are recited at a high level of generality and are being used in their ordinary capacity and are being used as a tool for implementing the steps of the identified abstract idea, see MPEP 2106.05(f), where applying a computer or using a computer as a tool to perform the abstract idea is not indicative of a practical application.
Accordingly, these additional elements, when considered separately and as an ordered combination, do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea
Therefore claims 1, 13 are directed to an abstract idea without a practical application. (Step 2A-Prong 2: NO. The additional claimed elements are not integrated into a practical application)
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because, when considered separately and as an ordered combination, they do not add significantly more (also known as an “inventive concept”) to the exception. As discussed above with respect to integration of the abstract idea into a practical application, there are no additional elements recited in the claim beyond the judicial exception.
Mere instructions to implement an abstract idea, on or with the use of generic computer components, or even without any computer components, cannot provide an inventive concept - rendering the claim patent ineligible. Thus claims 1,13 are not patent eligible. (Step 2B: NO. The claims do not provide significantly more)
Dependent claims 2-12, 14-24 further define the abstract idea that is present in their respective independent claims 1, 13 and thus correspond to Certain Methods of Organizing Human Activity and hence are abstract for the reasons presented above.
Claims 3, 15 further defines the identified abstract idea as recited in claims 1,13. The additional element of (the set of risk ML models comprise classification models) are recited a high level of generality, operating in their ordinary capacity, and are being used as a tool to implement the steps of the identified abstract idea, see MPEP 2106.05(f)
Therefore, the dependent claims do not include any additional elements that integrate the abstract idea into a practical application or are sufficient to amount to significantly more than the judicial exception when considered both individually and as an ordered combination. Therefore, the dependent claims (2-12, 14-24) are directed to an abstract idea. Thus, the claims 1-24 are not patent-eligible.
CONCLUSION
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MOHAMMAD Z SHAIKH whose telephone number is (571)270-3444. The examiner can normally be reached M-T, 9-600; Fri, 8-11, 3-5.
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, BENNETT SIGMOND can be reached at 303-297-4411. 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.
/MOHAMMAD Z SHAIKH/Primary Examiner, Art Unit 3694 6/3/2026
Prosecution Insights