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
This action is in response to the application filed 20 February 2026. Claims 1, 11, 12, 20, and 22 are amended. Claim 16 is cancelled. Claim 23 is newly added. Claims 1-15 and 17-23 are pending and have been examined.
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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 20 February 2026 has been entered.
Response to Arguments
Applicant' s arguments, see pages 15-18, filed 20 February 2026, with respect to the rejections of Claims 1-5, 7, 11-15, 17, 20, and 22 under 35 U.S.C. 102(a)(2) and Claims 6, 8-10, 16, 18, 19, and 21 under 35 U.S.C. 103 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
APPLICANT'S ARGUMENT: Applicant argues (page 15, paragraph 1) that "Belanger fails to disclose, either expressly or inherently, each and every element recited in Applicant's claims."
Applicant argues (page 16, paragraph 3) that "nothing in Belanger teaches or even suggests any processes that '... generate[s], in real-time upon receipt of the data record, output data' ... ¶ Thus, Belanger fails to teach each and every element recited similarly by Applicant's amended independent claims,
Applicant argues (page 16, paragraph 4) that "Belanger cannot anticipate these dependent claims."
EXAMINER'S RESPONSE: Examiner notes that Applicant's arguments pertain to newly claimed subject matter and that Applicant's arguments are moot, as amended Claim 1 is now rejected in view of Belanger in view of Cella. In the rejection of amended Claim 1 below, the feature of generating output data in real time is shown to be taught by Belanger.
Applicant's arguments, see pages 18-26, filed 20 February 2026, with respect to Claims 1-22 under 35 U.S.C. 101 have been fully considered but they are not persuasive.
APPLICANT'S ARGUMENT: Applicant argues (page 19, paragraph 1) that "the Office fails to provide sufficient reasoning as to why Applicant's claims recite an abstract idea or other judicial exception."
Applicant argues (page 19, paragraph 2) that "the Office fails to provide reasoning sufficient to establish that Applicant's independent claims recite a patent-ineligible mental process 'on their own or per se.'"
Applicant argues (page 20, continued paragraph) that "the Office's analysis of Applicant's claims under Prong One of Revised Step 2A of the Alice/Mayo test does not- and cannot- identify any portion of Applicant's Specification that would support its conclusion that a user could perform, via pen and paper or in the mind, any of these additional quoted elements allegedly recited by Applicant's independent claims. ¶ Further, the claimed combination of elements recited by Applicant's independent claims ... encompass artificial intelligence in a manner that cannot be performed practically in the human mind."
EXAMINER'S RESPONSE: Examiner respectfully disagrees. The step of validating a process and the various steps of generating further data based on existing data recited by amended Claim 1 appear to recite mental process steps, such as evaluation, interpretation, or opinion, when interpreted under BRI at the claimed level of generality in light of the of specification. The additional elements recited by amended Claim 1, when taken singly or in combination with the recited abstract ideas, do not appear to integrate the abstract ideas into a practical application so as to provide an improvement in the functioning of a computer or to other technology. Rather, the additional elements appear to recite steps that amount merely to implementing the abstract idea steps with a computer, or adding insignificant extra-solution activity to the mental process steps. Similarly, the additional elements of the claim do not appear to recite an inventive concept that amounts to significantly more than the mental process steps. Rather, the additional elements appear to recite steps that amount merely to implementing the abstract idea steps with a computer, or to recite well-understood, routine, and conventional activity.
APPLICANT'S ARGUMENT: Applicant argues (page 21, paragraph 2) that "the Office's analysis of Applicant's independent claims under Prong Two of Revised Step 2A of the Alice/Mayo test is inconsistent with the Office's own examination practice and with the actual language recited by Applicant's independent claims."
Applicant argues (page 22, paragraph 1) that "the Office's analysis fails to identify properly those elements recited in Applicant's independent claims that extend beyond any allegedly recited abstract idea. By way of example, even assuming that Applicant's independent claims could recite a patent-ineligible 'mental process,' which these claims do not, the Office provides no reasoning, beyond conclusory statements lacking basis within Applicant's Specification, to support its apparent assertion that the allegedly recited elements ... represent any allegedly patent-ineligible 'mental process.'"
Applicant argues (page 22, paragraph 3) that "the elements recited by Applicant's independent claims, when considered as a whole in unamended form, provide a specific, technological improvement to an existing technology or technical field, and as such, integrate any allegedly recited abstract idea into a patent-eligible, practical application."
Applicant argues (page 23, paragraph 1) that "independent claims 1, 11, and 20, as amended herein, integrate any allegedly recited abstract idea into a patent-eligible, practical application under Prong Two of Step 2A of the Alice/Mayo test."
Applicant argues (page 23, paragraph 1) that "Applicant's independent claims ... represent a specific, technological improvement to existing, computer-implemented predictive processes that ingest, operate on, and process increasing volumes of interaction data."
Applicant argues (page 23, paragraph 1) that "the claimed combination of elements recited similarly by independent claims 1, 11, and 20 extends beyond a mere application of an artificial-intelligence process to a new field of use, and instead provides a specific improvement to existing artificial intelligence processes. ... [T]he claimed combination recited by Applicant's amended independent claims, as quoted above, provides a specific improvement to these existing adaptive techniques."
Applicant argues (page 22, paragraph 2) that "the Office's analysis of Applicant's independent claims under Prong Two of Revised Step 2A of the Alice/Mayo fails to identify properly those additional elements of Applicant's independent claims that extend beyond the allegedly recited mental process and, as such, this analysis cannot evaluate properly those additional elements individually, and in combination, to determine whether they integrate the exception into a practical application."
EXAMINER'S RESPONSE: Examiner respectfully disagrees. The analysis of amended Claim 1 below has been undertaken according to the Alice/Mayo framework, and has identified the abstract element steps and additional elements of the claims below according to BRI in light of the specification. Examiner notes that, in the absence of additional elements that integrate the recited mental process steps into a practical application or provide significantly more, amended Claim 1 is directed to the mental process steps.
Examiner further notes that, as currently recited, amended Claim 1 does not appear to recite a technological improvement in processing increasing volumes of interaction data.
APPLICANT'S ARGUMENT: Applicant argues (page 25, paragraph 3) that "the above-quoted elements ..., when taken collectively as a whole, extend beyond any well-understood, routine, conventional, or human-performable activities, and represent a specific technological improvement to existing, computer-implemented predictive processes that ingest, operate on, and process increasing volumes of interaction data."
EXAMINER'S RESPONSE: Examiner respectfully disagrees. As indicated in the 35 U.S.C. 101 rejection below, the additional elements of amended Claim 1 were not found singly or in combination to provide significantly more. Therefore amended Claim 1 is directed to the recited mental process steps.
APPLICANT'S ARGUMENT: Applicant argues (page 25, paragraph 4) that "the Office appears apply the same reasoning in its analysis of Applicant's independent claims under both Prong Two of Revised Step 2A and Step 2B of the Alice/Mayo test, arguing that the additional elements '[do] not integrate the abstract idea into a practical application in Step 2A Prong Two or add significantly more in Step 2B.'"
EXAMINER'S RESPONSE: Examiner respectfully disagrees. See MPEP 2106.05(f), which states:
Another consideration when determining whether a claim integrates a judicial exception into a practical application in Step 2A Prong Two or recites significantly more than a judicial exception in Step 2B is whether the additional elements amount to more than a recitation of the words "apply it" (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer.
The Claim 1 steps of training, applying, and performing were analyzed in accordance with MPEP 2106.05(f) for Steps 2A Prong Two and Step 2B. The Claim 1 additional element step of transmitting data was analyzed according to the guidance of MPEP 2106.05(g) and MPEP 2106.05(d) regarding Step 2A Prong Two and Step 2B analysis, respectively. The steps were not found singly or in combination with the recited mental process steps to integrate the mental process steps into a practical application or provide significantly more.
APPLICANT'S ARGUMENT: Applicant argues (page 26, paragraph 3) that "Newly added claim 23 depends from amended independent claim 1 and as such, is neither anticipated, nor rendered obvious, by the references of record, and recites patent-eligible subject matter."
EXAMINER'S RESPONSE: Examiner respectfully disagrees. Claim 23 inherits the abstract ideas recited by parent Claim 1. The additional elements recited by Claim 23 do not appear to integrate the mental process steps into a practical application or recite significantly more for the reasons
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 1-15 and 17-23 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Regarding Claim 1
Step 1
Claim 1 recites an apparatus, comprising: a memory storing instructions; a communications interface; and at least one processor coupled to the memory and the communications interface, and thus the claimed machine falls within a statutory category of invention.
Step 2A Prong 1
The claim recites validate the trained artificial intelligence process based on a plurality of validation datasets associated with a second temporal interval, the second temporal interval being subsequent to the first temporal interval, which is a mental process. The claim recites generate an input dataset for the trained artificial intelligence process based on the event data, the pendency data, and the elements of first interaction data, which is a mental process. The claim recites based on the application of the trained artificial intelligence process to the input dataset, generate ... output data representative of a predicted likelihood of an occurrence of a second event within a predetermined time period subsequent to the occurrence of the first event, which is a mental process. The claim recites the computing system being configured to generate second interaction data specifying an operation associated with the occurrence of the first event based on the portion of the output data, which is a mental process.
Thus, the claim recites an abstract idea.
Step 2A Prong 2
The additional element train adaptively an artificial intelligence process based on a plurality of training datasets associated with a first temporal interval invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it"). The additional element receive, from a computing system via the communications interface, a data record comprising an identifier of a customer, event data characterizing an occurrence of a first event involving the customer during a third temporal interval, and pendency data characterizing a temporal pendency of the first event during the third temporal interval amounts to insignificant extra-solution activity (see MPEP 2106.05(g), "mere data gathering and outputting"). The additional element based on the customer identifier, obtain, from the memory, elements of first interaction data associated with the customer during the third temporal interval amounts to insignificant extra-solution activity (see MPEP 2106.05(g), "mere data gathering and outputting"). The additional element apply the trained artificial intelligence process to the input dataset invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it"). The additional element in real-time upon receipt of the data record invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it"). The additional element transmit at least a portion of the generated output data to the computing system via the communications interface amounts to insignificant extra-solution activity (see MPEP 2106.05(g), "mere data gathering and outputting"). The additional element perform the operation in accordance with the second interaction data invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it").
Step 2B
The additional element train an artificial intelligence process based on a plurality of training datasets associated with a first temporal interval invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it"). The additional element receive, from a computing system via the communications interface, a data record comprising an identifier of a customer, event data characterizing an occurrence of a first event involving the customer during a third temporal interval, and pendency data characterizing a temporal pendency of the first event during the third temporal interval is well-understood, routine, conventional activity (see MPEP 2106.05(d), "receiving or transmitting data over a network"). The additional element based on the customer identifier, obtain, from the memory, elements of first interaction data associated with the customer during the third temporal interval is well-understood, routine, conventional activity (see MPEP 2106.05(d), "storing and retrieving information in memory"). The additional element in real-time upon receipt of the data record invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it"). The additional element apply the trained artificial intelligence process to the input dataset invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it"). The additional element transmit at least a portion of the generated output data to a computing system via the communications interface is well-understood, routine, conventional activity (see MPEP 2106.05(d), "receiving or transmitting data over a network"). The additional element perform the operation in accordance with the second interaction data invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it").
The claim lacks additional elements that integrate it into a practical application or provide significantly more, so it is directed to an abstract idea and is ineligible.
Regarding Claim 2
Step 1
Regarding Claim 2, the rejection of Claim 1 is incorporated.
Step 2A Prong 1
Claim 2 recites those abstract ideas incorporated from the rejection of Claim 1.
Step 2A Prong 2
The additional element receive at least a portion of the elements of the first interaction data from the computing system via the communications interface amounts to insignificant extra-solution activity (see MPEP 2106.05(g), "mere data gathering"). The additional element store the received portion of the first interaction data within the memory amounts to insignificant extra-solution activity (see MPEP 2106.05(g), "mere data gathering and outputting").
Step 2B
The additional element receive at least a portion of the elements of the first interaction data from the computing system via the communications interface is well-understood, routine, conventional activity (see MPEP 2106.05(d), "receiving or transmitting data over a network"). The additional element store the received portion of the first interaction data within the memory is well-understood, routine, conventional activity (see MPEP 2106.05(d), "storing and retrieving information in memory").
The claim lacks additional elements that integrate it into a practical application or provide significantly more, so it is directed to an abstract idea and is ineligible.
Regarding Claim 3
Step 1
Regarding Claim 3, the rejection of Claim 1 is incorporated.
Step 2A Prong 1
The claim recites generate the input dataset in accordance with the data that characterizes the composition, which is a mental process.
Thus, the claim recites an abstract idea.
Step 2A Prong 2
The additional element obtain ... one or more parameters that characterize the trained artificial intelligence process amounts to insignificant extra-solution activity (see MPEP 2106.05(g), "mere data gathering"). The additional element obtain ... data that characterizes a composition of the input dataset amounts to insignificant extra-solution activity (see MPEP 2106.05(g), "mere data gathering"). The additional element apply the trained artificial intelligence process to the input dataset in accordance with the one or more parameters invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it").
Step 2B
The additional element obtain ... one or more parameters that characterize the trained artificial intelligence process is well-understood, routine, conventional activity (see MPEP 2106.05(d), "storing and retrieving information in memory"). The additional element obtain ... data that characterizes a composition of the input dataset is well-understood, routine, conventional activity (see MPEP 2106.05(d), "storing and retrieving information in memory"). The additional element apply the trained artificial intelligence process to the input dataset in accordance with the one or more parameters invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it").
The claim lacks additional elements that integrate it into a practical application or provide significantly more, so it is directed to an abstract idea and is ineligible.
Regarding Claim 4
Step 1
Regarding Claim 4, the rejection of Claim 3 is incorporated.
Step 2A Prong 1
The claim recites based on the data that characterizes the composition, perform operations, which is a mental process. The claim recites extract a first feature value from the elements of the first interaction data, which is a mental process. The claim recites compute a second feature value based on the first feature value, which is a mental process. The claim recites generate the input dataset based on at least one of the extracted first feature value or the computed second feature value, which is a mental process.
Thus, the claim recites an abstract idea.
Step 2A Prong 2, Step 2B
The claim lacks additional elements that integrate it into a practical application or provide significantly more, so it is directed to an abstract idea and is ineligible.
Regarding Claim 5
Step 1
Regarding Claim 5, the rejection of Claim 1 is incorporated.
Step 2A Prong 1
The claim recites generate ... output data representative of a predicted likelihood of an occurrence of a second event within a predetermined time period subsequent to the occurrence of the first event (as recited in Claim 1), wherein the output data comprises a numerical score indicative of the predicted likelihood of the occurrence of the second event within the predetermined time period subsequent to the occurrence of the first event, which is a mental process. The claim recites the computing system being configured to generate second interaction data specifying an operation associated with the occurrence of the first event based on the portion of the output data (as recited in Claim 1), wherein ... the computing system is further configured to generate the second interaction data that specifies the operation associated with the occurrence of the first event based on the numerical score, which is a mental process. The claim recites the computing system being configured to generate second interaction data specifying an operation associated with the occurrence of the first event based on the portion of the output data (as recited in Claim 1), wherein ... the operation is consistent with the predicted likelihood of the occurrence of the second event, which is a mental process.
Thus, the claim recites an abstract idea.
Step 2A Prong 2, Step 2B
The claim lacks additional elements that integrate it into a practical application or provide significantly more, so it is directed to an abstract idea and is ineligible.
Regarding Claim 6
Step 1
Regarding Claim 6, the rejection of Claim 1 is incorporated.
Step 2A Prong 1
Claim 6 recites the abstract ideas incorporated from the rejection of Claim 1.
Step 2A Prong 2, Step 2B
The additional element train adaptively an artificial intelligence process based on a plurality of training datasets associated with a first temporal interval (as recited in Claim 1), wherein the trained artificial intelligence process comprises a trained, gradient-boosted, decision-tree process invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it").
The claim lacks additional elements that integrate it into a practical application or provide significantly more, so it is directed to an abstract idea and is ineligible.
Regarding Claim 7
Step 1
Regarding Claim 7, the rejection of Claim 1 is incorporated.
Step 2A Prong 1
The claim recites the first interaction data is associated with a plurality of customers, each of the customers being associated with a corresponding occurrence of the first event, which is a mental process. The claim recites generate input datasets based on the first interaction data, each of the plurality of input datasets being associated with a corresponding one of the customers, which is a mental process. The claim recites based on the application of the trained artificial intelligence to each of the plurality of input datasets, generate an element of the output data representative of a predicted likelihood of a corresponding occurrence of the second event within the target temporal interval subsequent to the corresponding occurrence of the first event; and each of the generated elements of output data includes a numerical score indicative of the predicted likelihood of the corresponding occurrence of the second event for a corresponding one of the customers, which is a mental process.
Thus, the claim recites an abstract idea.
Step 2A Prong 2, Step 2B
The additional element apply the trained artificial intelligence process to each of the plurality of input datasets invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it").
The claim lacks additional elements that integrate it into a practical application or provide significantly more, so it is directed to an abstract idea and is ineligible.
Regarding Claim 8
Step 1
Regarding Claim 8, the rejection of Claim 1 is incorporated.
Step 2A Prong 1
The claim recites validate the trained artificial intelligence process based on a plurality of validation datasets associated with a second temporal interval, the second temporal interval being subsequent to the first temporal interval (as recited in Claim 1), the second temporal interval corresponds to a prior validation interval, which is a mental process. The claim recites based on the temporal identifiers, determine that a first subset of the elements of the third interaction data are associated with the prior training interval, and that a second subset of the elements of the third interaction data are associated with the prior validation interval, which is a mental process. The claim recites generate training datasets based on corresponding portions of the first subset, and perform operations that train the artificial intelligence process based on the training datasets, which is a mental process.
Thus, the claim recites an abstract idea.
Step 2A Prong 2
The additional element train adaptively an artificial intelligence process based on a plurality of training datasets associated with a first temporal interval (as recited in Claim 1), the first temporal interval corresponds to a prior training interval invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it"). The additional element obtain elements of third interaction data, each of the elements of the third interaction data comprising a temporal identifier associated with a temporal interval amounts to insignificant extra-solution activity (see MPEP 2106.05(g), "mere data gathering").
Step 2B
The additional element train adaptively an artificial intelligence process based on a plurality of training datasets associated with a first temporal interval (as recited in Claim 1), the first temporal interval corresponds to a prior training interval invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it"). The additional element obtain elements of third interaction data, each of the elements of the third interaction data comprising a temporal identifier associated with a temporal interval is well-understood, routine, conventional activity (see MPEP 2106.05(d), "receiving or transmitting data over a network").
The claim lacks additional elements that integrate it into a practical application or provide significantly more, so it is directed to an abstract idea and is ineligible.
Regarding Claim 9
Step 1
Regarding Claim 9, the rejection of Claim 8 is incorporated.
Step 2A Prong 1
The claim recites generate validation datasets based on portions of the second subset, which is a mental process. The claim recites generate additional elements of output data based on the application of the trained artificial intelligence process to the plurality of validation datasets, which is a mental process. The claim recites compute one or more validation metrics based on the additional elements of output data, which is a mental process. The claim recites based on a determined consistency between the one or more validation metrics and a threshold condition, validate the trained artificial intelligence process, which is a mental process.
Thus, the claim recites an abstract idea.
Step 2A Prong 2, Step 2B
The additional element apply the trained artificial intelligence process to the plurality of validation datasets invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it").
The claim lacks additional elements that integrate it into a practical application or provide significantly more, so it is directed to an abstract idea and is ineligible.
Regarding Claim 10
Step 1
Regarding Claim 10, the rejection of Claim 1 is incorporated.
Step 2A Prong 1
The claim recites generate ... output data representative of a predicted likelihood of an occurrence of a second event within a predetermined time period subsequent to the occurrence of the first event (as recited in Claim 1), wherein the second event comprises a default event involving the customer, which is a mental process. The claim recites the default event occurs when a pendency period of the delinquency event exceeds a threshold period, which is a mental process. The claim recites generate ... output data representative of a predicted likelihood of an occurrence of a second event within a predetermined time period subsequent to the occurrence of the first event (as recited in Claim 1), wherein the output data is representative of a predicted likelihood of an occurrence of the default event within the predetermined time period subsequent to the occurrence of the delinquency event, which is a mental process. The claim recites generate second interaction data specifying an operation (as recited in Claim 1), wherein the operation comprises a remediation process associated with the delinquency event, which is a mental process. The claim recites perform operations that implement the remediation process in accordance with the second interaction data and resolve the delinquency event based on the implementation of the remediation process, which is a mental process.
Thus, the claim recites an abstract idea.
Step 2A Prong Two
The additional element receive ... event data characterizing an occurrence of a first event involving the customer during a third temporal interval (as recited in Claim 1), wherein the first event comprises a delinquency event involving a customer amounts to insignificant extra-solution activity (see MPEP 2106.05(g), "mere data gathering and outputting").
Step 2B
The additional element receive ... event data characterizing an occurrence of a first event involving the customer during a third temporal interval (as recited in Claim 1), wherein the first event comprises a delinquency event involving a customer is well-understood, routine, conventional activity (see MPEP 2106.05(d), "receiving or transmitting data over a network").
The claim lacks additional elements that integrate it into a practical application or provide significantly more, so it is directed to an abstract idea and is ineligible.
Regarding Claim 11
Step 1
Claim 11 recites a computer-implemented method, and thus the claimed process falls within a statutory category of invention.
Step 2A Prong 1
The claim recites validating the trained artificial intelligence process based on a plurality of validation datasets associated with a second temporal interval, the second temporal interval being subsequent to the first temporal interval, which is a mental process. The claim recites generating an input dataset for the trained artificial intelligence process based on the event data, the pendency data, and the elements of first interaction data, which is a mental process. The claim recites based on the application of the trained artificial intelligence process to the input dataset, generating ... output data representative of a predicted likelihood of an occurrence of a second event within a predetermined time period subsequent to the occurrence of the first event, which is a mental process. The claim recites the computing system being configured to generate second interaction data specifying an operation associated with the occurrence of the first event based on the portion of the output data, which is a mental process.
Thus, the claim recites an abstract idea.
Step 2A Prong 2
The additional element using at least one processor, training adaptively an artificial intelligence process based on a plurality of training datasets associated with a first temporal interval invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it"). The additional element receiving, from a computing system using the at least one processor, a data record comprising an identifier of a customer, event data characterizing an occurrence of a first event involving the customer during a third temporal interval, and pendency data characterizing a temporal pendency of the first event during the third temporal interval amounts to insignificant extra-solution activity (see MPEP 2106.05(g), "mere data gathering and outputting"). The additional element based on the customer identifier, obtaining, from a data repository using the at least one processor, elements of first interaction data associated with the customer during the third temporal interval amounts to insignificant extra-solution activity (see MPEP 2106.05(g), "mere data gathering and outputting"). The additional element using the at least one processor invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it"). The additional element using the at least one processor, applying the trained artificial intelligence process to the input dataset invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it"). The additional element in real-time upon receipt of the data record invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it"). The additional element transmitting, using the at least one processor, at least a portion of the generated output data to the computing system amounts to insignificant extra-solution activity (see MPEP 2106.05(g), "mere data gathering and outputting"). The additional element perform the operation in accordance with the second interaction data invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it").
Step 2B
The additional element using at least one processor, training adaptively an artificial intelligence process based on a plurality of training datasets associated with a first temporal interval invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it"). The additional element receiving, from a computing system using the at least one processor, a data record comprising an identifier of a customer, event data characterizing an occurrence of a first event involving the customer during a third temporal interval, and pendency data characterizing a temporal pendency of the first event during the third temporal interval is well-understood, routine, conventional activity (see MPEP 2106.05(d), "receiving or transmitting data over a network"). The additional element based on the customer identifier, obtaining, from a data repository using the at least one processor, elements of first interaction data associated with the customer during the third temporal interval is well-understood, routine, conventional activity (see MPEP 2106.05(d), "receiving or transmitting data over a network"). The additional element using the at least one processor invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it"). The additional element using the at least one processor, applying the trained artificial intelligence process to the input dataset invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it"). The additional element in real-time upon receipt of the data record invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it"). The additional element transmitting, using the at least one processor, at least a portion of the generated output data to the computing system is well-understood, routine, conventional activity (see MPEP 2106.05(d), "receiving or transmitting data over a network"). The additional element perform the operation in accordance with the second interaction data invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it").
The claim lacks additional elements that integrate it into a practical application or provide significantly more, so it is directed to an abstract idea and is ineligible.
Claims 12-15 and 17-19, dependent on Claim 11, incorporate the rejection of Claim 11. Claims 12-15 and 17-19 incorporate substantively limitations of Claims 2-5 and 7-9, respectively, and are rejected under the same rationales.
Regarding Claim 20
Step 1
Claim 20 recites an apparatus, comprising: a memory storing instructions; a communications interface; and at least one processor coupled to the memory and the communications interface, the at least one processor being configured to execute the instructions, and thus the claimed machine falls within a statutory category of invention.
Step 2A Prong 1
The claim recites validate the trained artificial intelligence process based on a plurality of validation datasets associated with a third temporal interval, the third temporal interval being subsequent to the second temporal interval, which is a mental process. The claim recites generate an input dataset for the trained artificial intelligence process based on the event data, the pendency data, and the elements of first interaction data, which is a mental process. The claim recites generate ... the elements of output data based on an application of the trained artificial intelligence process to the input dataset, which is a mental process. The claim recites based on the elements of output data, generate elements of second interaction data that specify one or more operations associated with the occurrence of the first event, which is a mental process.
Thus, the claim recites an abstract idea.
Step 2A Prong 2
The additional element transmit, to a computing system via the communications interface, a data record comprising an identifier of a customer, event data characterizing an occurrence of a first event involving the customer during a first temporal interval, and pendency data characterizing a temporal pendency of the first event during the first temporal interval amounts to insignificant extra-solution activity (see MPEP 2106.05(g), "mere data gathering and outputting"). The additional element receive elements of output data from the computing system via the communications interface, the elements of output data being representative of a predicted likelihood of an occurrence of a second event within a predetermined time period subsequent to the occurrence of the first event amounts to insignificant extra-solution activity (see MPEP 2106.05(g), "mere data gathering and outputting"). The additional element the computing system being configured to: train adaptively an artificial intelligence process based on a plurality of training datasets associated with a second temporal interval invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it"). The additional element based on the customer identifier, obtain elements of first interaction data associated with the customer during the first temporal interval amounts to insignificant extra-solution activity (see MPEP 2106.05(g), "mere data gathering and outputting"). The additional element in real-time invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it"). The additional element perform operations that implement the one or more specified operations in accordance with the elements of second interaction data invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it").
Step 2B
The additional element transmit, to a computing system via the communications interface, a data record comprising an identifier of a customer, event data characterizing an occurrence of a first event involving the customer during a first temporal interval, and pendency data characterizing a temporal pendency of the first event during the first temporal interval is well-understood, routine, conventional activity (see MPEP 2106.05(d), "receiving or transmitting data over a network"). The additional element receive elements of output data from the computing system via the communications interface, the elements of output data being representative of a predicted likelihood of an occurrence of a second event within a predetermined time period subsequent to the occurrence of the first event is well-understood, routine, conventional activity (see MPEP 2106.05(d), "receiving or transmitting data over a network"). The additional element the computing system being configured to: train adaptively an artificial intelligence process based on a plurality of training datasets associated with a second temporal interval invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it"). The additional element based on the customer identifier, obtain elements of first interaction data associated with the customer during the first temporal interval is well-understood, routine, conventional activity (see MPEP 2106.05(d), "storing and retrieving information in memory"). The additional element in real-time invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it"). The additional element perform operations that implement the one or more specified operations in accordance with the elements of second interaction data invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it").
The claim lacks additional elements that integrate it into a practical application or provide significantly more, so it is directed to an abstract idea and is ineligible.
Regarding Claim 21
Step 1
Regarding Claim 21, the rejection of Claim 1 is incorporated.
Step 2A Prong One
The claim recites generate the input dataset in accordance with elements of composition data that characterize the composition, which is a mental process. The claim recites based on at least the output data, compute a value of one or more metrics characterizing the application of the trained artificial intelligence process to the input dataset, which is a mental process. The claim recites determine an inconsistency between the one or more metric values and at least one threshold condition, which is a mental process. The claim recites perform operations that modify at least one of the composition data or the one or more parameter values in accordance with the determined inconsistency, which is a mental process.
Thus, the claim recites an abstract idea.
Step 2A Prong Two
The additional element obtain ... a value of one or more parameters that characterize the trained artificial intelligence process amounts to insignificant extra-solution activity (see MPEP 2106.05(g), "mere data gathering"). The additional element obtain ... composition data that characterizes a composition of the input dataset amounts to insignificant extra-solution activity (see MPEP 2106.05(g), "mere data gathering"). The additional element apply the trained artificial intelligence process to the input dataset in accordance with the one or more parameter values invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it").
Step 2B
The additional element obtain ... a value of one or more parameters that characterize the trained artificial intelligence process is well-understood, routine, conventional activity (see MPEP 2106.05(d), "receiving or transmitting data over a network"). The additional element obtain ... composition data that characterizes a composition of the input dataset is well-understood, routine, conventional activity (see MPEP 2106.05(d), "receiving or transmitting data over a network"). The additional element apply the trained artificial intelligence process to the input dataset in accordance with the one or more parameter values invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it").
The claim lacks additional elements that integrate it into a practical application or provide significantly more, so it is directed to an abstract idea and is ineligible.
Regarding Claim 22
Step 1
Regarding Claim 22, the rejection of Claim 1 is incorporated.
Step 2A Prong One
The claim recites generate an input dataset for the trained artificial intelligence process based on the event data, the pendency data, and the elements of first interaction data (as recited by Claim 1), wherein: each of the elements of first interaction data comprises a temporal identifier that indicates a temporal position within the third temporal interval, which is a mental process. The claim recites generate an input dataset for the trained artificial intelligence process based on the event data, the pendency data, and the elements of first interaction data (as recited by Claim 1), wherein ... at least one of the elements of first interaction data characterizes the occurrence of the first event during the third temporal interval, which is a mental process. The claim recites generate an input dataset for the trained artificial intelligence process based on the event data, the pendency data, and the elements of first interaction data (as recited by Claim 1), wherein ... a subset of the elements of interaction are being associated with temporal positions disposed subsequent to the occurrence of the first event during the third temporal interval, which is a mental process.
Thus, the claim recites an abstract idea.
Step 2A Prong Two, Step 2B
The claim lacks additional elements that integrate it into a practical application or provide significantly more, so it is directed to an abstract idea and is ineligible.
Regarding Claim 23
Step 1
Regarding Claim 23, the rejection of Claim 1 is incorporated.
Step 2A Prong One
Claim 23 recites the abstract ideas recited by parent Claim 1.
Step 2A Prong 2
The additional element receive ... pendency data characterizing a temporal pendency of the first event during the third temporal interval (as recited by Claim 1) wherein the pendency data comprises a value of the temporal pendency of the first event amounts to insignificant extra-solution activity (see MPEP 2106.05(g), "mere data gathering and outputting"). The additional element the at least one processor is further configured to execute the instructions to: obtain the value of the temporal pendency from the data record amounts to insignificant extra-solution activity (see MPEP 2106.05(g), "mere data gathering and outputting"). The additional element perform operations that package the value of the temporal pendency into the input dataset at a corresponding sequential position invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it").
Step 2B
The additional element receive ... pendency data characterizing a temporal pendency of the first event during the third temporal interval (as recited by Claim 1) wherein the pendency data comprises a value of the temporal pendency of the first event is well-understood, routine, conventional activity (see MPEP 2106.05(d), "receiving or transmitting data over a network"). The additional element the at least one processor is further configured to execute the instructions to: obtain the value of the temporal pendency from the data record is well-understood, routine, conventional activity (see MPEP 2106.05(d), "receiving or transmitting data over a network"). The additional element perform operations that package the value of the temporal pendency into the input dataset at a corresponding sequential position invokes a computer or other machinery merely as a tool to perform an existing process (see MPEP 2106.05(f), "apply it").
The claim lacks additional elements that integrate it into a practical application or provide significantly more, so it is directed to an abstract idea and is ineligible.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1-5, 7, 10-15, 17, 20, 22, and 23 is rejected under 35 U.S.C. 103 as being unpatentable over Belanger, et al. (U.S. 2019/0340684 A1, hereinafter "Belanger") in view of Cella (U.S. 2020/0294128 A1, hereinafter "Cella").
Regarding Claim 1, Belanger teaches:
an apparatus, comprising: a memory storing instructions; a communications interface; and at least one processor coupled to the memory and the communications interface, the at least one processor being configured to execute the instructions to: (Belanger, Claim 1: "A tangible, non-transitory, machine-readable medium storing instructions that when executed by one or more processors effectuate operations," and [0075]: "Computing system 1000 may include one or more processors (e.g., processors 1010a-1010n) coupled to system memory 1020, an input/output I/O device interface 1030, and a network interface 1040")
train adaptively an artificial intelligence process (Belanger, [0002]: "The present disclosure relates to artificial intelligence (AI) and more particularly to machine learning used for monitoring or controlling continuous stochastic processes based on discrete samples and other events in time series data" and [0038]: "In some cases, a predictive model (e.g., a vector of weights) may be calculated as a batch process run periodically") based on a plurality of training datasets (Belanger, [0028]: "The availability of actions and events on many time series ... are used to train machine learning models to estimate a risk index" and [0029]: "Training these models with diverse risk data, possibly from a variety of sources, is expected to enrich their ability to address as many different types of risk, system components, and workflows or sections as are contained in the training data," where Belanger's data from a variety of sources corresponds to the instant's plural datasets) associated with a first temporal interval (Belanger, [0047]: "In some embodiments, the exogenous events are stochastic, and some embodiments may associate with exogenous events an estimated or known probability distribution, like likelihoods of occurring within threshold durations of time," where Belanger's threshold durations of time correspond to the instant temporal interval), and validate the trained artificial intelligence process based on a plurality of validation datasets associated with a second temporal interval (Belanger, [0039]: "In some cases, a subset of the training set may be withheld in each of several iterations of training the model to cross validate the model. The model may be trained periodically, e.g., monthly, in advance of use of the model"), the second temporal interval being subsequent to the first temporal interval (Belanger, [0038]: "The resulting, trained model, e.g., a vector of weights or biases, may be stored in memory and later retrieved for application to new calculations on newly calculated risk scores. In some cases, cyclic loops in the network may be unrolled during training");
receive ... a data record (Belanger, [0042]: "the controller 12 is configured to ... use the trained model to characterize risk as a continuous stochastic variable that is updated as current events are received via the event streams 17," where Belanger's events streams comprise events that correspond to the instant data records, as in [0021]: "Some embodiments create a customer risk journey (e.g., a temporally (or at least sequence) indexed record of events that may affect risk-related metrics and indicators).... Journeys may be encoded in memory as a set of time-stamped or sequenced entries in an interaction-event record, each entry, in some cases, including an event and information about that event") comprising an identifier of a customer (Belanger, [0048]: "each interaction- event record may correspond to a different subject entity, such as a different ... person.... In some embodiments, each interaction-event record may have associated therewith a unique identifier of the subject, in some cases a pseudonymous identifier" and [0072]: "the interaction-event records 402 may include previous purchase data 404(1), finance data (e.g., associated with consumer financing) 404(2), demographic data (e.g., customer's age, income, zip code, and the like)"), event data characterizing an occurrence of a first event involving the customer during a third temporal interval (Belanger, [0033]: "an event timeline or other interaction-event record that includes one or more interactions between a customer and a supplier may be determined or otherwise obtained ( e.g., from historical logs of a CRM system, complain logs, invoicing systems, and the like). A starting risk value may be assigned to individual events in the event timeline. A sub-sequence comprising a portion of the event timeline that includes at least one reference event may be selected. A classifier may be used to determine a previous relative risk value for a previous event that occurred before the reference event and to determine a next relative risk value for a next event that occurred after the reference event until all events in the event timeline have been processed," where Belanger's event timeline corresponds to the instant temporal interval) ... from a computing system via the communications interface (Belanger, Fig. 1, 17, "event stream," depicting computing device 12 receiving an event stream, where [0040]: "the above techniques may be implemented in a computing environment 10 shown in FIG. 1, for example, with the illustrated continuous stochastic process controller 12" and [0042]: "the controller 12 is configured to train a risk scoring machine learning model based upon historical interaction-event records 14 and then use the trained model to characterize risk as a continuous stochastic variable that is updated as current events are received via the event streams 17" and [0041]: "In some embodiments, some or all of the components of the computing environment 10 may be hosted by different entities, for instance in remote datacenters that communicate via the Internet or other networks.... In some embodiments, some or all of the illustrated components may be implemented as distinct services executing on different network hosts that communicate with one another via messages exchanged via network stacks of the respective hosts");
... obtain ... elements of first interaction data associated with the customer during the third temporal interval, and generate an input dataset for the trained artificial intelligence process based on the event data ... and the elements of first interaction data (Belanger, [0033]: "an event timeline or other interaction-event record that includes one or more interactions between a customer and a supplier may be determined or otherwise obtained ( e.g., from historical logs of a CRM system, complain logs, invoicing systems, and the like). ... A sub-sequence comprising a portion of the event timeline that includes at least one reference event may be selected. A classifier may be used to determine a previous relative risk value for a previous event that occurred before the reference event and to determine a next relative risk value for a next event that occurred after the reference event until all events in the event timeline have been processed") ... based on the customer identifier ... from the memory (Belanger, [0048]: "records describing histories of events experienced or potentially experienced by subjects are stored in the interaction-event record repository 14 in interaction-event record. ... In some embodiments, each interaction-event record may have associated therewith a unique identifier of the subject, in some cases a pseudonymous identifier");
apply the trained artificial intelligence process to the input dataset (Belanger, [0021]: "Some embodiments create a customer risk journey (e.g., a temporally (or at least sequence). ... Machine learning may be used to extract the appropriate patterns. The models built and trained with the risk journey time series may be used to score a step's (in the journey) risk posture," where Belanger's used to score a step corresponds to the instant apply), and based on the application of the trained artificial intelligence process to the input dataset, generate... output data representative of a predicted likelihood of an occurrence of a second event within a predetermined time period subsequent to the occurrence of the first event (Belanger, [0021]: "The models built and trained with the risk journey time series may be used to score a step's (in the journey) risk posture in the form of a risk index. ... The ability to assess the risk index is not limited to past and present events, in some embodiments, but it may also be used to predict the risk index for future events. As a result, the models may be used to predict the likelihood a risk incident may happen" and [0047]: "the events may further include exogenous events ... to which the subject entity is exposed or potentially exposed. ... Other examples include phenomena like recessions, changes in interest rates, and other macro-economic phenomena.... In some embodiments, the exogenous events are stochastic ... like likelihoods of occurring within threshold durations of time") ... in real-time upon receipt of the data record (Belanger, [0023]: "Some embodiments manage (e.g., infer and effectuate decisions based on) continuous risk as a time series of events and actions taken (or not) within a system's context ... and implement a methodology to continuously assess a continuous risk posture" and [0028]: "These models, in some embodiments, can then be used to predict (e.g., may execute the act of predicting) the likelihood of future incidents, thus providing a continuous assessment of continuous risk," where Belanger's continuous risk assessment corresponds to the instant real-time) ... ; and
transmit at least a portion of the generated output data to the computing system via the communications interface (Belanger, [0055]: "the controller 12 may exercise control (which may be outcome determinative control or merely control over downstream inputs that merely influence a downstream process), via one or more of the action-channel servers 18") the computing system being configured to generate second interaction data specifying an operation associated with the occurrence of the first event based on the portion of the output data (Belanger, [0055]: "the controller 12 may exercise control ... via one or more of the action channel servers 18. ... Examples of resulting actions may also include outputting a score.... Examples of resulting actions may also include configuring a parameter of a product, like an interest rate" where the instant second interaction data corresponds to the data of Belanger's resulting actions), and perform the operation in accordance with the second interaction data (Belanger, [0055]: "different servers 18 may communicate with various downstream systems, like loan or insurance underwriting computer systems, ERP systems, or CRM systems, to update those systems on predicted current state of risk, e.g., ... by pushing updates," where the instant second interaction data corresponds to the data that Belanger's action-channel server uses to communicate downstream).
Belanger teaches receiving a data record comprising event data characterizing an occurrence of a first event involving the customer during a third temporal interval and generating an input dataset for the trained artificial intelligence process based on the event data and the elements of first interaction data.
Belanger does not explicitly teach a data record comprising ... event data characterizing an occurrence of a first event ..., and pendency data characterizing a temporal pendency of the first event and generate an input dataset for the ... artificial intelligence process based ... the pendency data.
However, Cella teaches:
a data record (Cella, [1154]: "The system may include a data collection circuit 5112 structured to interpret entity information 5102, collateral data 5104, and the like.... The system may include a loan management circuit 5132 structured to interpret loan related actions 5134 and/or events 5138 in response to the entity information ... and the loan terms and conditions where the loan related events are associated with the loan") comprising ... event data characterizing an occurrence of a first event (Cella, [0137]: "The term loan related event(s) ... may include any event related to terms of the loan or events triggered by the agreement associated with the loan. Loan-related events may include default on loan, breach of contract, fulfillment, repayment, payment, change in interest, late fee assessment, refund assessment, distribution, and the like," where Cella's default corresponds to the instant event) ..., and pendency data characterizing a temporal pendency of the first event (Cella, [0180]: "The term loan-related action, events, and activities, as noted herein, may also more specifically be utilized to describe a context for calling of a loan. ... For example, a loan-related action for calling of the loan may occur when a borrower misses three payments in a row, such that there is a severe delinquency in the loan payment schedule, and the loan goes into default. ... In some circumstances a ... robotic process automation system may initiate, administrate or process loan-related actions for calling of the loan," where Cella's period of delinquency corresponds to the instant temporal pendency)
generate an input dataset (Cella, [1154]: "The system may include a data collection circuit 5112 structured to interpret entity information 5102, collateral data 5104, and the like, such as corresponding to entities related to a lending transaction corresponding to the loan, collateral conditions, and the like. ... The received data 5102 5104 and the collateral condition 5110 may be provided to AI circuits 5142 which may include an automated agent circuit 5114 (e.g., processing events 5118, 5120)") for the ... artificial intelligence process based ... the pendency data (Cella, [1153]: "The AI systems 5062 may include an automated agent circuit 5070 that takes action based on collateral events, loan-events and the like. Actions may include loan-related actions such as ... calling the loan ... , providing notices required to be provided to a borrower ... , and the like").
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Belanger regarding Belanger teaches receiving a data record comprising event data characterizing an occurrence of a first event involving the customer during a third temporal interval and generating an input dataset for the trained artificial intelligence process based on the event data and the elements of first interaction data. with those of Cella regarding a data record comprising event data characterizing an occurrence of a first event, and pendency data characterizing a temporal pendency of the first event and generate an input dataset for the artificial intelligence process based the pendency data.
The motivation to do so would be to take advantage of improved loan-related outcomes based on iterative training (Cella, [0178]: "The term robotic process automation system as utilized herein may be understood broadly to describe a system capable of performing tasks or providing needs for a system of the present disclosure. For example, a robotic process automation system, without limitation, can be configured for: ... loan collection, ... configuring a data collection and monitoring action based on a set of attributes of a loan ... and iteratively training and improving based on a set of outcomes. A robotic process automation system may include: a set of data collection and monitoring services, an artificial intelligence system, and another robotic process automation system which is a component of the higher level robotic process automation system").
Claim 11 incorporates substantively all the limitations of Claim 1 in a computer-implemented method (Belanger, [0009]: "FIG. 1 is a logical architecture block diagram of an example continuous stochastic process controller and its computing environment in accordance with some embodiments of the present techniques") and is rejected under the same rationale.
Regarding Claim 2, the rejection of Claim 1 is incorporated. The Belanger/Cella combination teaches:
wherein the at least one processor is further configured to: receive at least a portion of the elements of the first interaction data from the computing system via the communications interface (Belanger, [0043]: "In some embodiments, interaction-event records reflecting previous customer risk journeys may be obtained from the interaction-event record repositories 14. Examples of such repositories include ... customer relationship management databases," where the instant computing system here corresponds to the system comprising Belanger's interaction-event record repositories, and use of a communication interface is inherent in communication with a database) and store the received portion of the first interaction data within the memory (Belanger, [0067]: "Some embodiments may then store the sets of event-risk scores in memory, as indicated by block 60").
Claim 12 incorporates substantively all the limitations of Claim 2 in a computer-implemented method and is rejected under the same rationale.
Regarding Claim 3, the rejection of Claim 1 is incorporated. The Belanger/Cella combination teaches:
wherein the at least one processor is further configured to: obtain (i) one or more parameters that characterize the trained artificial intelligence process (Belanger, [0061]: "In some embodiments, the explainability module 30 may access the risk scores and trained model parameters from the model 24 to advise users, like consumers, those offering products, regulators or auditors, on the causes of the risk scores that are calculated.") obtain ... (ii) data that characterizes a composition of the input dataset (Belanger, [0049]: "In some embodiments, each interaction-event record may further include attributes of the subject");
generate the input dataset in accordance with the data that characterizes the composition (Belanger, [0051]: "In some embodiments, the event attributes may not express whether such attributes are good or bad, merely provide a metric that may be interpreted with, for example, a reward function of the controller 12"); and
apply the trained artificial intelligence process to the input dataset in accordance with the one or more parameters (Belanger, [0061]: "In some embodiments, the explainability module 30 may access the risk scores and trained model parameters from the model 24 to advise users, like consumers, those offering products, regulators or auditors, on the causes of the risk scores that are calculated.").
Claim 13 incorporates substantively all the limitations of Claim 3 in a computer-implemented method and is rejected under the same rationale.
Regarding Claim 4, the rejection of Claim 3 is incorporated. Belanger teaches:
wherein the at least one processor is further configured to: based on the data that characterizes the composition, perform operations that at least one of (i) extract a first feature value from the elements of the first interaction data (Belanger, [0049]: "each interaction-event record may further include attributes of the subject," where Belanger's included attributes correspond to the instant first feature value) or (ii) compute a second feature value based on the first feature value (Belanger, [0069]: "Some embodiments may further determine a measure of contribution of events or types of events to the sets of event-risk scores, as indicated by block 64, for example with the above-described explainability module 30. In some embodiments may cause the measure of contribution to be presented to a user to the instruct the user on how to modulate risk, as indicated by block 60," where Belanger's computed measure of contribution corresponds to the instant second feature value); and
generate the input dataset based on at least one of the extracted first feature value or the computed second feature value (Belanger, [0089]: "for a journey of interaction-events available in a fraud detection platform, extract or receive information from the journey of interaction-events regarding attributes related to an entity's progress in the journey of interaction-events...; for each entity of the plurality of entities, apply a classifying function to classify the entity progress in the journey of interaction-events as normal or as outlying based upon the attributes provided by the journey of interaction-events").
Claim 14 incorporates substantively all the limitations of Claim 4 in a computer-implemented method and is rejected under the same rationale.
Regarding Claim 5, the rejection of Claim 1 is incorporated. The Belanger/Cella combination teaches:
the output data comprises a numerical score indicative of the predicted likelihood of the occurrence of the second event within the predetermined time period subsequent to the occurrence of the first event (Belanger, [0016]: "in some embodiments, risk is no longer a static factor ... Risk (e.g., a likelihood of an undesirable outcome, or expected utility thereof) may be a continuously managed signal (e.g., a parameter) that can be used to ... trigger activities"); and
the computing system is further configured to generate the second interaction data that specifies the operation associated with the occurrence of the first event based on the numerical score (Belanger, [0055] "In some cases, the controller 12 may exercise control (which may be outcome determinative control or merely control over downstream inputs that merely influence a downstream process), via one or more of the action-channel servers 18. ... Examples of resulting actions may also include outputting a score indicative of whether a consumer should be sold or marketed a particular product like those described above"); and
the operation is consistent with the predicted likelihood of the occurrence of the second event (Belanger, [0051]: "In some embodiments, events corresponding to actions by the subject-entity may have attributes indicating whether the respective event is advancing a goal, such as ... maximizing likelihood of good outcomes").
Claim 15 incorporates substantively all the limitations of Claim 5 in a computer-implemented method and is rejected under the same rationale.
Regarding Claim 7, the rejection of Claim 1 is incorporated. The Belanger/Cella combination teaches:
wherein: the first interaction data is associated with a plurality of customers, each of the customers being associated with a corresponding occurrence of the first event (Belanger, [0044]: "In some embodiments, each record may be timeseries of events for one of a relatively large number of independent entities for which actions are selected to influence behavior or responsive to predicted behavior, such as of different people in a population"); and
the at least one processor is further configured to execute the instructions to: generate input datasets based on the first interaction data, each of the plurality of input datasets being associated with a corresponding one of the customers (Belanger, [0021]: "Some embodiments create a customer risk journey (e.g., a temporally (or at least sequence) indexed record of events that may affect risk-related metrics and indicators) in the form of an event timeline integrating the different events that impact or otherwise reflect the risk behavior of a customer");
apply the trained artificial intelligence process to each of the plurality of input datasets (Belanger, [0021]: "Machine learning may be used to extract the appropriate patterns. The models built and trained with the risk journey time series may be used to score a step's (in the journey) risk posture in the form of a risk index.") and based on the application of the trained artificial intelligence to each of the plurality of input datasets, generate an element of the output data representative of a predicted likelihood of a corresponding occurrence of the second event (Belanger, [0021]: "As a result, the models may be used to predict the likelihood a risk incident may happen, as well as plan actions (future steps) to decrease the risk index and thus improve continuous risk posture") within the target temporal interval subsequent to the corresponding occurrence of the first event (Belanger, [0047]: "In some embodiments, the exogenous events are stochastic, and some embodiments may associate with exogenous events an estimated or known probability distribution, like likelihoods of occurring within threshold durations of time"); and
each of the generated elements of output data includes a numerical score indicative of the predicted likelihood of the corresponding occurrence of the second event for a corresponding one of the customers (Belanger, [0021]: "Some embodiments create a customer risk journey (e.g., a temporally (or at least sequence) indexed record of events that may affect risk-related metrics and indicators) in the form of an event timeline integrating the different events that impact or otherwise reflect the risk behavior of a customer.... The ability to assess the risk index is not limited to past and present events, in some embodiments, but it may also be used to predict the risk index for future events. As a result, the models may be used to predict the likelihood a risk incident may happen").
Claim 17 incorporates substantively all the limitations of Claim 7 in a computer-implemented method and is rejected under the same rationale.
Regarding Claim 10, the rejection of Claim 1 is incorporated.
Cella further teaches:
the first event (Cella, [0137]: "The term loan related event(s) ... may include any event related to terms of the loan or events triggered by the agreement associated with the loan") comprises a delinquency event (Cella, [0180]: "A calling of a loan is an action wherein the lender can demand the loan be repaid, usually triggered by some other condition or term, such as delinquent payment") involving a customer (Cella, [0180]: "a loan-related action for calling of the loan may occur when a borrower misses three payments in a row"), and the second event comprises a default event involving the customer (Cella, [0180]: "a loan-related action for calling of the loan may occur when ... there is a severe delinquency in the loan payment schedule, and the loan goes into default");
the default event occurs when a pendency period of the delinquency event exceeds a threshold period Cella, [0180]: "A calling of a loan is an action wherein the lender can demand the loan be repaid, usually triggered by some other condition or term, such as delinquent payment(s). For example, a loan-related action for calling of the loan may occur when a borrower misses three payments in a row, such that there is a severe delinquency in the loan payment schedule, and the loan goes into default");
the output data is representative of a predicted likelihood of an occurrence of the default event (Cella, [0216]: "Condition classification may include grouping or labeling entities, or clustering the entities, as similarly positioned with regard to some aspect of the classified condition (e.g., a risk, ..., likelihood to default, or some other aspect of the related debt)") within the predetermined time period subsequent to the occurrence of the delinquency event (Cella, [0128]: "financial condition may further include an assessment of the ability of the entity to survive future risk scenarios" and [0137]: "considerations ... in determining whether a contemplated data is a loan related event ... include ... a time period and/or response time associated with the event (e.g., ... time that is allotted from the time the event is triggered to when processing or detection of the event is desired to occur," where delinquency is a condition, as in [0180]: "A calling of a loan is an action wherein the lender can demand the loan be repaid, usually triggered by some other condition or term, such as delinquent payment(s)" and where risk conditions are predicted, as in [0143]: "considerations ... in determining whether a contemplated data is a loan condition ... include ... risk associated with the loan (conditions may depend on the probability that the loan may not be repaid)" and [0477]: "risk factors of the borrower ... including predicted risk based on one or more predictive models using artificial intelligence 156)" and [1072]: "For classification problems, one output is produced (with a separate set of weights and summation units) for each target category. The value output for a category is the probability that the case being evaluated has that category");
the operation comprises a remediation process associated with the delinquency event (Cella, [0180]: "a loan-related action for calling of the loan may occur when a borrower misses three payments in a row, such that there is a severe delinquency in the loan payment schedule, and the loan goes into default. ... In such a scenario, perhaps the borrower pays a sum to cure the delinquency and penalties, which may also be considered as a loan-related action for calling of the loan"); and
the computing system is further configured to perform operations that implement the remediation process ... based on the implementation of the remediation process (Cella, [0180]: "In some circumstances a smart contract or robotic process automation system may initiate, administrate or process loan-related actions for calling of the loan") in accordance with the second interaction data and resolve the delinquency event (Cella, [0180]: "loan-related actions for calling of the loan ... may including providing notice, researching and collecting payment history, or other tasks performed as a part of the calling of the loan").
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of the Belanger/Cella combination regarding predicted likelihoods of first and second events made by a trained model, with the further teachings of Cella regarding delinquency and default event remediation based on predicted occurrences of events.
The motivation to do so would be to take advantage of improved predictions made according to loan-related events (Cella, [0137]: "determining whether a contemplated data is a loan related event and/or whether aspects of the present disclosure can benefit or enhance the contemplated transaction system include, without limitation: the impact of the related event on the loan (events that cause default or termination of the loan may have higher impact), the cost (capital and/or operating) associated with the event, the cost ( capital and/or operating) associated with monitoring for an occurrence of the event").
Regarding Claim 20, Belanger teaches:
an apparatus, comprising: a memory storing instructions; a communications interface; and at least one processor coupled to the memory and the communications interface, the at least one processor being configured to execute the instructions (Belanger, [0075]: "Computing system 1000 may include one or more processors (e.g., processors 1010a-1010n) coupled to system memory 1020, an input/output I/O device interface 1030, and a network interface 1040 via an input/output (I/O) interface 1050. ... A processor may be any suitable processor capable of executing or otherwise performing instructions. ... A processor may receive instructions and data from a memory (e.g., system memory 1020).") to:
transmit ... to a computing system via the communications interface ... (Belanger, [0054]: "state to which controller 12 is responsive ( e.g., in online use cases after training) may be ingested from various event streams 17 ( some may be continuously feeds, and some may be batch feeds), which may take the form of a series of events like those described above. ... In some embodiments, the stream may be a batch process," where Belanger's batch process reasonably suggests aggregation and transmission of event records) a data record (Belanger, [0042]: "the controller 12 is configured to ... use the trained model to characterize risk as a continuous stochastic variable that is updated as current events are received via the event streams 17," where Belanger's events streams comprise events that correspond to the instant data records, as in [0021]: "Some embodiments create a customer risk journey (e.g., a temporally (or at least sequence) indexed record of events that may affect risk-related metrics and indicators).... Journeys may be encoded in memory as a set of time-stamped or sequenced entries in an interaction-event record, each entry, in some cases, including an event and information about that event") comprising an identifier of a customer (Belanger, [0048]: "each interaction- event record may correspond to a different subject entity, such as a different ... person.... In some embodiments, each interaction-event record may have associated therewith a unique identifier of the subject, in some cases a pseudonymous identifier" and [0072]: "the interaction-event records 402 may include previous purchase data 404(1), finance data (e.g., associated with consumer financing) 404(2), demographic data (e.g., customer's age, income, zip code, and the like)"), event data characterizing an occurrence of a first event involving the customer during a first temporal interval (Belanger, [0033]: "an event timeline or other interaction-event record that includes one or more interactions between a customer and a supplier may be determined or otherwise obtained ( e.g., from historical logs of a CRM system, complain logs, invoicing systems, and the like). A starting risk value may be assigned to individual events in the event timeline. A sub-sequence comprising a portion of the event timeline that includes at least one reference event may be selected. A classifier may be used to determine a previous relative risk value for a previous event that occurred before the reference event and to determine a next relative risk value for a next event that occurred after the reference event until all events in the event timeline have been processed," where Belanger's event timeline corresponds to the instant temporal interval);
receive elements of output data from the computing system via the communications interface (Belanger, [0043]: "In some embodiments, the controller 12 may train and apply various machine learning models to inputs from the components 14, 16, and 17 to effectuate various actions implemented via the action-channel servers 18." with [0041]: "A variety of different computing architectures are contemplated. In some embodiments, some or all of the components of the computing environment 10 may be hosted by different entities, for instance in remote datacenters that communicate via the Internet or other networks"), the elements of output data being representative of a predicted likelihood of an occurrence of a second event within a predetermined time period subsequent to the occurrence of the first event (Belanger, [0059]: "In some embodiments, after training, the risk scoring model 24 may update risk-scores in the risk-score repository 26 responsive to new events in event streams 17 ... the risk scores may account for both a likelihood of an event and an expected cost or other measure of undesirability ... of that event, e.g., with the product of the two values" and [0021]: "The models built and trained with the risk journey time series may be used to score a step's (in the journey) risk posture in the form of a risk index ... As a result, the models may be used to predict the likelihood a risk incident may happen, as well as plan actions (future steps) to decrease the risk index" with [0047]: "some embodiments may associate with exogenous events an estimated or known probability distribution, like likelihoods of occurring within threshold durations of time"), the computing system being configured to: train adaptively (Belanger, [0026]: "In some embodiments, machine learning models may be trained using adaptive network-based fuzzy inference systems to convert the current static risk management models into a nonlinear mapping system") an artificial intelligence process (Belanger, [0002]: "The present disclosure relates to artificial intelligence (AI) and more particularly to machine learning used for monitoring or controlling continuous stochastic processes based on discrete samples and other events in time series data" and [0038]: "In some cases, a predictive model (e.g., a vector of weights) may be calculated as a batch process run periodically") based on a plurality of training datasets (Belanger, [0028]: "The availability of actions and events on many time series ... are used to train machine learning models to estimate a risk index" and [0029]: "Training these models with diverse risk data, possibly from a variety of sources, is expected to enrich their ability to address as many different types of risk, system components, and workflows or sections as are contained in the training data," where Belanger's data from a variety of sources corresponds to the instant's plural datasets) associated with second temporal interval (Belanger, [0047]: "In some embodiments, the exogenous events are stochastic, and some embodiments may associate with exogenous events an estimated or known probability distribution, like likelihoods of occurring within threshold durations of time," where Belanger's threshold durations of time correspond to the instant temporal interval), and validate the trained artificial intelligence process based on a plurality of validation datasets associated with a third temporal interval (Belanger, [0039]: "In some cases, a subset of the training set may be withheld in each of several iterations of training the model to cross validate the model. The model may be trained periodically, e.g., monthly, in advance of use of the model"), the third temporal interval being subsequent to the second temporal interval (Belanger, [0038]: "The resulting, trained model, e.g., a vector of weights or biases, may be stored in memory and later retrieved for application to new calculations on newly calculated risk scores. In some cases, cyclic loops in the network may be unrolled during training");
... obtain elements of first interaction data associated with the customer during the first temporal interval; generate an input dataset for the trained artificial intelligence process based on the event data ... and the elements of first interaction data (Belanger, [0033]: "an event timeline or other interaction-event record that includes one or more interactions between a customer and a supplier may be determined or otherwise obtained ( e.g., from historical logs of a CRM system, complain logs, invoicing systems, and the like). ... A sub-sequence comprising a portion of the event timeline that includes at least one reference event may be selected. A classifier may be used to determine a previous relative risk value for a previous event that occurred before the reference event and to determine a next relative risk value for a next event that occurred after the reference event until all events in the event timeline have been processed") ... based on the customer identifier (Belanger, [0048]: "records describing histories of events experienced or potentially experienced by subjects are stored in the interaction-event record repository 14 in interaction-event record. ... In some embodiments, each interaction-event record may have associated therewith a unique identifier of the subject, in some cases a pseudonymous identifier"); and
generate ... the elements of output data (Belanger, [0021]: "The models built and trained with the risk journey time series may be used to score a step's (in the journey) risk posture in the form of a risk index. ... The ability to assess the risk index is not limited to past and present events, in some embodiments, but it may also be used to predict the risk index for future events. As a result, the models may be used to predict the likelihood a risk incident may happen" and [0047]: "the events may further include exogenous events ... to which the subject entity is exposed or potentially exposed. ... Other examples include phenomena like recessions, changes in interest rates, and other macro-economic phenomena.... In some embodiments, the exogenous events are stochastic ... like likelihoods of occurring within threshold durations of time") based on an application of the trained artificial intelligence process to the input dataset (Belanger, [0021]: "Some embodiments create a customer risk journey (e.g., a temporally (or at least sequence). ... Machine learning may be used to extract the appropriate patterns. The models built and trained with the risk journey time series may be used to score a step's (in the journey) risk posture," where Belanger's used to score a step corresponds to the instant apply) ... in real time (Belanger, [0023]: "Some embodiments manage (e.g., infer and effectuate decisions based on) continuous risk as a time series of events and actions taken (or not) within a system's context ... and implement a methodology to continuously assess a continuous risk posture" and [0028]: "These models, in some embodiments, can then be used to predict (e.g., may execute the act of predicting) the likelihood of future incidents, thus providing a continuous assessment of continuous risk," where Belanger's continuous risk assessment corresponds to the instant real-time)
based on the elements of output data, generate elements of second interaction data that specify one or more operations associated with the occurrence of the first event (Belanger, [0055]: "the controller 12 may exercise control ... via one or more of the action channel servers 18. ... Examples of resulting actions may also include outputting a score.... Examples of resulting actions may also include configuring a parameter of a product, like an interest rate" where the instant second interaction data corresponds to the data of Belanger's resulting actions), and perform operations that implement the one or more specified operations in accordance with the elements of second interaction data (Belanger, [0055]: "different servers 18 may communicate with various downstream systems, like loan or insurance underwriting computer systems, ERP systems, or CRM systems, to update those systems on predicted current state of risk, e.g., ... by pushing updates," where the instant second interaction data corresponds to the data that Belanger's action-channel server uses to communicate downstream).
Belanger teaches receiving a data record comprising event data characterizing an occurrence of a first event involving the customer during a third temporal interval and generating an input dataset for the trained artificial intelligence process based on the event data and the elements of first interaction data.
Belanger does not explicitly teach a data record comprising ... event data characterizing an occurrence of a first event ..., and pendency data characterizing a temporal pendency of the first event and generate an input dataset for the ... artificial intelligence process based ... the pendency data.
However, Cella teaches:
a data record (Cella, [1154]: "The system may include a data collection circuit 5112 structured to interpret entity information 5102, collateral data 5104, and the like.... The system may include a loan management circuit 5132 structured to interpret loan related actions 5134 and/or events 5138 in response to the entity information ... and the loan terms and conditions where the loan related events are associated with the loan") comprising ... event data characterizing an occurrence of a first event (Cella, [0137]: "The term loan related event(s) ... may include any event related to terms of the loan or events triggered by the agreement associated with the loan. Loan-related events may include default on loan, breach of contract, fulfillment, repayment, payment, change in interest, late fee assessment, refund assessment, distribution, and the like," where Cella's default corresponds to the instant event) ..., and pendency data characterizing a temporal pendency of the first event (Cella, [0180]: "The term loan-related action, events, and activities, as noted herein, may also more specifically be utilized to describe a context for calling of a loan. ... For example, a loan-related action for calling of the loan may occur when a borrower misses three payments in a row, such that there is a severe delinquency in the loan payment schedule, and the loan goes into default. ... In some circumstances a ... robotic process automation system may initiate, administrate or process loan-related actions for calling of the loan," where Cella's period of delinquency corresponds to the instant temporal pendency)
generate an input dataset (Cella, [1154]: "The system may include a data collection circuit 5112 structured to interpret entity information 5102, collateral data 5104, and the like, such as corresponding to entities related to a lending transaction corresponding to the loan, collateral conditions, and the like. ... The received data 5102 5104 and the collateral condition 5110 may be provided to AI circuits 5142 which may include an automated agent circuit 5114 (e.g., processing events 5118, 5120)") for the ... artificial intelligence process based ... the pendency data (Cella, [1153]: "The AI systems 5062 may include an automated agent circuit 5070 that takes action based on collateral events, loan-events and the like. Actions may include loan-related actions such as ... calling the loan ... , providing notices required to be provided to a borrower ... , and the like").
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Belanger regarding Belanger teaches receiving a data record comprising event data characterizing an occurrence of a first event involving the customer during a third temporal interval and generating an input dataset for the trained artificial intelligence process based on the event data and the elements of first interaction data. with those of Cella regarding a data record comprising event data characterizing an occurrence of a first event, and pendency data characterizing a temporal pendency of the first event and generate an input dataset for the artificial intelligence process based the pendency data.
The motivation to do so would be to take advantage of improved loan-related outcomes based on iterative training (Cella, [0178]: "The term robotic process automation system as utilized herein may be understood broadly to describe a system capable of performing tasks or providing needs for a system of the present disclosure. For example, a robotic process automation system, without limitation, can be configured for: ... loan collection, ... configuring a data collection and monitoring action based on a set of attributes of a loan ... and iteratively training and improving based on a set of outcomes. A robotic process automation system may include: a set of data collection and monitoring services, an artificial intelligence system, and another robotic process automation system which is a component of the higher level robotic process automation system").
Regarding Claim 22, the rejection of Claim 1 is incorporated. Belanger teaches:
wherein: each of the elements of first interaction data comprises a temporal identifier that indicates a temporal position within the third temporal interval (Belanger, [0021]: "Some embodiments create a customer risk journey (e.g., a temporally (or at least sequence) indexed record of events that may affect risk-related metrics and indicators) in the form of an event timeline integrating the different events that impact or otherwise reflect the risk behavior of a customer. ... Journeys may be encoded in memory as a set of time-stamped or sequenced entries in an interaction-event record," where Belanger's temporally indexed corresponds to the instant comprising a temporal identifier indicating temporal position);
at least one of the elements of first interaction data characterizes the occurrence of the first event during the third temporal interval (Belanger, [0033]: "an event timeline or other interaction-event record that includes one or more interactions between a customer and a supplier may be determined or otherwise obtained .... A sub-sequence comprising a portion of the event timeline that includes at least one reference event may be selected. A classifier may be used to determine a previous relative risk value for a previous event that occurred before the reference event and to determine a next relative risk value for a next event that occurred after the reference event until all events in the event timeline have been processed," where Belanger's reference event corresponds to the instant first event); and
a subset of the elements of interaction are being associated with temporal positions disposed subsequent to the occurrence of the first event during the third temporal interval (Belanger, [0033]: "A sub-sequence comprising a portion of the event timeline that includes at least one reference event may be selected. A classifier may be used to determine a previous relative risk value for a previous event that occurred before the reference event and to determine a next relative risk value for a next event that occurred after the reference event until all events in the event timeline have been processed," where Belanger's next and remaining events corresponds to the instant subsequent subset).
Regarding Claim 23, the rejection of Claim 1 is incorporated.
Cella further teaches:
the pendency data comprises a value of the temporal pendency of the first event (Cella, [0180]: "The term loan-related action, events, and activities, as noted herein, may also more specifically be utilized to describe a context for calling of a loan. ... For example, a loan-related action for calling of the loan may occur when a borrower misses three payments in a row, such that there is a severe delinquency in the loan payment schedule, and the loan goes into default. ... In some circumstances a ... robotic process automation system may initiate, administrate or process loan-related actions for calling of the loan"); the at least one processor is further configured to execute the instructions (Cella, [2314]: "The methods and/or processes described above, and steps associated therewith, may be realized in hardware, software or any combination of hardware and software suitable for a particular application. The hardware may include a general-purpose computer and/or dedicated computing device or specific computing device or particular aspect or component of a specific computing device") to:
obtain the value of the temporal pendency from the data record (Cella, [1154]: "The system may include a data collection circuit 5112 structured to interpret entity information 5102, collateral data 5104, and the like, such as corresponding to entities related to a lending transaction corresponding to the loan, collateral conditions, and the like"); and
perform operations that package the value of the temporal pendency into the input dataset (Cella, [1154]: "The received data 5102 5104 and the collateral condition 5110 may be provided to AI circuits 5142 which may include an automated agent circuit 5114 (e.g., processing events 5118, 5120)") at a corresponding sequential position (Cella, [1075]: "methods and systems described herein that involve an expert system or self-organization capability may use a recurrent neural network, which may allow for a bi-directional flow of data, such as where connected units (e.g., neurons or nodes) form a directed cycle. ... In embodiments, the recurrent neural network may use internal memory to process a sequence of inputs ... In embodiments, the recurrent neural network may also be used for pattern recognition, such as for recognizing a machine, component, agent, or other item based on a behavioral signature, a profile, a set of feature vectors (such as in an audio file or image), or the like," where Cella's sequential inputs for an RNN reasonably suggests the instant corresponding sequential position).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of the Belanger/Cella combination regarding event data characterizing an occurrence of a first event and pendency data characterizing a temporal pendency of the first event with the further teachings of Cella regarding the pendency data comprising a value of the temporal pendency of the first event; the at least one processor is further configured to execute the instructions to: obtain the value of the temporal pendency from the data record; and perform operations that package the value of the temporal pendency into the input dataset at a corresponding sequential position.
The motivation to do so would be to facilitate use of models of market or transactional resources that have dynamic behavior (Cella, [1119]: "methods and systems described herein that involve an expert system or self-organization capability may use a recurrent neural network.... Such a network may be used to model or exhibit dynamic temporal behavior ... where dynamic system behavior involves complex interactions that a user may desire to understand, predict, control and/or optimize. For example, the recurrent neural network may be used to anticipate the state of a market, such as one involving a dynamic process or action, such as a change in state of a resource that is traded in or that enables a marketplace of transactional environment. In embodiments, the recurrent neural network may use internal memory to process a sequence of inputs, such as from ... data inputs from or about the transactional environment").
Claim 6 is rejected under 35 U.S.C. 103 as being unpatentable over Belanger, et al. (U.S. 2019/0340684 A1, hereinafter "Belanger") in view of Cella (U.S. 2020/0294128 A1, hereinafter "Cella") in view of Harris, et al. (U.S. 2020/0097817 A1, hereinafter "Harris").
Regarding Claim 6, the rejection of Claim 1 is incorporated.
The Belanger/Cella combination teaches a trained artificial intelligence process.
The Belanger/Cella combination does not explicitly teach wherein the trained artificial intelligence process comprises a trained, gradient-boosted, decision-tree process.
However, Harris teaches:
wherein the trained artificial intelligence process comprises a trained, gradient-boosted, decision-tree process (Harris, [0110]: "At step 508, after storing the test sample, the analysis computer can train one or more models with the interaction data. ... The one or more models can include any suitable models, for example, in some embodiments, the one or more models can include a decision tree. The analysis computer can create the decision tree in part by gradient tree boosting using the training sample and a predefined target feature").
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of the Belanger/Cella combination regarding a trained artificial intelligence process with those of Harris regarding use of a trained, gradient-boosted decision tree.
The motivation to do so would be to provide a means for accounting for model error with respect to each input interaction data (Harris, [0084]: "The analysis computer can then create a model, for example, a gradient boosted tree, which may predict whether or not interaction data is fraudulent interaction data. The analysis computer can then evaluate the performance of the resulting model using a sample data set. The analysis computer can determine, based on the analysis of the performance, residual error values for each input interaction data. The analysis computer can also determine a total error").
Claims 8, 9, 18, 19, and 21 are rejected under 35 U.S.C. 103 as being unpatentable over Belanger, et al. (U.S. 2019/0340684 A1, hereinafter "Belanger") in view of Cella (U.S. 2020/0294128 A1, hereinafter "Cella") in view of Nori, et al. (U.S. 2021/0224602 A1, hereinafter "Nori").
Regarding Claim 8, the rejection of Claim 1 is incorporated. The Belanger/Cella combination teaches:
wherein: the first temporal interval corresponds to a prior training interval (Belanger, [0047]: "the exogenous events are stochastic.... Records of such events may be obtained from repository 16 in some embodiments" and [0065]: "the process 50 includes obtaining historical interaction-event records, as indicated by block 52, for example, with the above-described classifier 20 from the interaction-event record repositories 14. In some embodiments, this may further include obtaining exogenous event records from the repository 16 described above" and [0066]: "Next, some embodiments may train a machine learning model on the historical interaction-event records, as indicated by block 54. In some embodiments, this may be performed by the above-described model trainer 22. Some embodiments may then obtain current interaction-event records, as indicated by block 56, again for example, with the classifier 20 from the repositories 14 or streams 17 described above");
the second temporal interval corresponds to a prior validation interval (Belanger, [0039]: "a subset of the training set may be withheld in each of several iterations of training the model to cross validate the model").
The Belanger/Cella combination may not explicitly teach the at least one processor is further configured to execute the instructions to: obtain elements of third interaction data, each of the elements of the third interaction data comprising a temporal identifier associated with a temporal interval; based on the temporal identifiers, determine that a first subset of the elements of the third interaction data are associated with the prior training interval, and that a second subset of the elements of the third interaction data are associated with the prior validation interval; and generate training datasets based on corresponding portions of the first subset, and perform operations that train the artificial intelligence process based on the training datasets.
However, Nori teaches:
the at least one processor is further configured to execute the instructions to: obtain elements of third interaction data, each of the elements of the third interaction data comprising a temporal identifier associated with a temporal interval (Nori, [0033]: "Embodiments of the present disclosure provide for data labelling according to a defined ruleset using a dual prediction model system, improving accuracy of the generated data labels with respect to long-term predictions (e.g., predictions based on historical data from X years ago, where X is a number or a range of numbers) and enabling labelling of data associated with candidates that were previously discarded");
based on the temporal identifiers, determine that a first subset of the elements of the third interaction data are associated with the prior training interval (Nori, [0090]: "In some embodiments, the apparatus 200 is configured to identify the candidate data subsets 312A-312C based on corresponding index dates and a particular timestamp interval. For example, for each candidate identifier, the subset of candidate data records may include data records associated with a record timestamp within a short term record threshold of time before the index date"), and that a second subset of the elements of the third interaction data are associated with the prior validation interval (Nori, [0095]: "It should be appreciated that, in some such embodiments, training data may be matched (e.g., candidate identifiers and/or corresponding candidate data records of a positive candidate data set 312A and negative candidate data set 312C), but validation and/or test data remains umnodified to ensure validity of the validation or test"); and
generate training datasets based on corresponding portions of the first subset, and perform operations that train the artificial intelligence process based on the training datasets (Nori, [0033]-[0034]: "In the context of labelling candidates as a case or control for purposes of a RCT [randomized clinical trial], the generated candidate positive-label probability represents a probability that the data associated with the candidate ( e.g., electronic health records and/or claim data having particular features) indicates the candidate should be labeled a case ... The candidate label probabilistic model learns label probability generating to ensure that candidates with the same features represented within their corresponding data records have the same probability of receiving a particular label (e.g., same probability of being labeled a case). In the context of labeling candidates for an RCT into a case cohort or a control cohort, for example, this means that candidates with the medical features are scored to have the same probability of having a case label assigned").
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of the Belanger/Cella combination regarding training of machine learning models with those of Nori regarding devising training sets to train models using historical data, where the models provide result labels with probability scores when the data was previously discarded from training or labeling.
The motivation to do so would be to condition the inclusion of data in training sets according to criteria depending on the use case of the trained model (Nori [0027]: "Errors in data labelling can lead a candidate to wrongly being included in and/or excluded from an RCT, which in many instances is a costly and/or harmful, if not deadly, mistake. To avoid such mistakes, RCTs often target only the specific, limited population that may be definitively labelled as a case, for example satisfying every rule of a particular ruleset associated with the disease diagnosis, or a control, for example failing to satisfy any rule of the particular ruleset associated with the disease diagnosis. However, relying on such limited candidate pools lead to significant delays and/or clinical trial failures").
Claim 18 incorporates substantively all the limitations of Claim 8 in a computer-implemented method and is rejected under the same rationale.
Regarding Claim 9, the rejection of Claim 8 is incorporated. Nori further teaches:
wherein the at least one processor is further configured to execute the instructions to: generate validation datasets based on portions of the second subset (Nori , [0093]: "In this regard, remaining data records not utilized for training may be utilized for model training validation and/or testing" and [0095]: "It should be appreciated that, in some such embodiments, training data may be matched (e.g., candidate identifiers and/or corresponding candidate data records of a positive candidate data set 312A and negative candidate data set 312C), but validation and/or test data remains unmodified to ensure validity of the validation or test");
apply the trained artificial intelligence process to the plurality of validation datasets, and generate additional elements of output data based on the application of the trained artificial intelligence process to the plurality of validation datasets (Nori, [0122]: "At optional block 814, the apparatus 200 includes means, such as the label prediction module 210, communications module 208, processor 202, and/or the like, or a combination thereof, configured for utilizing the historical prediction model to generate a long-term candidate positive-label probability associated with at least one candidate identifier. ... Alternatively or additionally, in some embodiments, the apparatus 200 is configured to utilize the historical prediction model to generate a long-term candidate positive-label probability set for candidate identifiers in a test set," where the instant additional elements of output data correspond to Nori's labels);
compute one or more validation metrics based on the additional elements of output data (Nori, previous at [0122], where the instant validation metrics correspond to Nori's long-term candidate positive-label probability set) and
based on a determined consistency between the one or more validation metrics and a threshold condition, validate the trained artificial intelligence process (Nori, [0093]: " In such embodiments, the apparatus 200 is configured to utilize known data set separation and/or partitioning for training, validation, and/or testing to ensure generation of a properly trained candidate label probabilistic model 316").
Claim 19 incorporates substantively all the limitations of Claim 9 in a computer-implemented method and is rejected under the same rationale.
Regarding Claim 21, the rejection of Claim 1 is incorporated. The Belanger/Cella combination teaches:
wherein the at least one processor is further configured to execute the instructions to: obtain (i) a value of one or more parameters that characterize the trained artificial intelligence process (Belanger, [0061]: "In some embodiments, the explainability module 30 may access the risk scores and trained model parameters from the model 24 to advise users, like consumers, those offering products, regulators or auditors, on the causes of the risk scores that are calculated.") and (ii) composition data that characterizes a composition of the input dataset (Belanger, [0049]: "In some embodiments, each interaction-event record may further include attributes of the subject");
generate the input dataset in accordance with elements of composition data that characterize the composition (Belanger, [0051]: "In some embodiments, the event attributes may not express whether such attributes are good or bad, merely provide a metric that may be interpreted with, for example, a reward function of the controller 12"), and apply the trained artificial intelligence process to the input dataset in accordance with the one or more parameter values (Belanger, [0061]: "In some embodiments, the explainability module 30 may access the risk scores and trained model parameters from the model 24 to advise users, like consumers, those offering products, regulators or auditors, on the causes of the risk scores that are calculated.");
based on at least the output data, compute a value of one or more metrics characterizing the application of the trained artificial intelligence process to the input dataset (Belanger, [0069]: "Some embodiments may further determine a measure of contribution of events or types of events to the sets of event-risk scores, as indicated by block 64, for example with the above-described explainability module," where Belanger's measure of contribution corresponds to the instant metrics).
The Belanger/Cella combination may not explicitly teach determine an inconsistency between the one or more metric values and at least one threshold condition, and perform operations that modify at least one of the composition data or the one or more parameter values in accordance with the determined inconsistency.
However, Nori teaches:
determine an inconsistency between the one or more metric values and at least one threshold condition (Nori, Fig. 9, "NO" branch following 906: "Total score neighborhood count satisfies a neighborhood count threshold?," where Nori's total score corresponds to the instant metric value), and perform operations that modify at least one of the composition data or the one or more parameter values in accordance with the determined inconsistency and perform operations that modify the at least one process parameter value in accordance with the determined inconsistency (Nori, Fig. 9, 908: "Adjusting the score adjustment range by multiplying the score adjustment range with a score adjustment factor," where Nori's adjustment corresponds to the instant operation, and where Nori's range corresponds to the instant composition data).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of the Belanger/Cella combination regarding applying the trained artificial intelligence process to the input dataset and computing a metric characterizing the application with those of Nori regarding determining an inconsistency between a metric value and a threshold condition and modify the composition data or parameter values accordingly.
The motivation to do so would be to ensure that the model remains calibrated to expected data (Nori, [0123]: "FIG. 9 illustrates additional example process for label predicting using a dual-model system, specifically for generating a well-calibrated adjusted candidate positive label probability").
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ROBERT N DAY whose telephone number is (703)756-1519. 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, Kakali Chaki can be reached at (571) 272-3719. 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.
/R.N.D./Examiner, Art Unit 2122
/KAKALI CHAKI/Supervisory Patent Examiner, Art Unit 2122
Prosecution Insights