DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Amendment
This action is in response to the submission filed 10 October 2025 for application 18/471,659. Claim 1 has been canceled. Claims 2, 4, 6, 7, 9, 11, 12, 14, and 16 have been amended. Currently, claims 2-16 are pending and have been examined.
The objection to claims 6, 11, and 16 has been withdrawn in view of the amendments made.
The §112(b) rejection of claims 2-16 has been withdrawn in view of the amendments made.
Response to Arguments
Applicant’s arguments, see pages 7-9, filed 13 October 2025, in regards to rejections under 35 U.S.C. § 103, with respect to the feature “by subtracting the probability risk associated with a parent node from the probability risk associated with a child parent node” as recited in independent claim 2 (and similarly in independent claims 7 and 12) that neither combination of the cited references teach or suggest all the elements of amended claim 2.
Examiner’s response: Applicant’s arguments have been fully considered but are moot because the new ground of rejection (citing new reference Moore et al (Cached Sufficient Statistics for Efficient Machine Learning with Large Datasets, 1998) for teaching the new limitation) does not rely on any reference combination applied in the prior rejection of record for any teaching or matter specifically challenged in the argument. Hence the rejection of claim 2 (and similarly claims 7 and 12) is maintained along with the rejection of the other claims as they are dependent on one of the independent claims as shown in the detailed rejection below.
Applicant’s arguments, see page 10, filed 13 October 2025, in regards to rejections under 35 U.S.C. § 101, Applicant argues that the Office Action rejected Claims 2-16 under 35 U.S.C. § 101 as allegedly directed to an abstract idea without significantly more. While Applicant respectfully disagrees with the rejection of Claims 2-16, in order to advance prosecution, Applicant has amended independent Claims 2, 7, and 12. Applicant respectfully requests that the Office reconsider and withdraw the rejections under 35 U.S.C. § 101.
Examiner’s response: Applicant’s arguments have been fully considered but are not persuasive. Examiner disagrees that the rejection can be withdrawn because firstly even upon reconsideration the limitations that were amended are still identified as abstract ideas under Step 2A, prong 1 as shown in the detailed rejection below. Secondly, Applicant has failed to provide any explanation or reasons about which limitations of the rejection Applicant disagrees with. Hence, the rejection is maintained.
Claim Objections
Claim 4 is objected to because of the following informalities: The phrase “… for the determining a parent decision node …” on line 3 is awkwardly worded. Appropriate correction is required.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claim 2-16 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 2 recites the limitation "…the probability risk …” in line 14. There is insufficient antecedent basis for this limitation in the claim. Claims 3 - 6 depend on claim 2 and therefore inherit the same rejection.
Claim 7 recites the limitation "…the probability risk …” on Page 3 (last but one line). There is insufficient antecedent basis for this limitation in the claim. Claims 8 - 11 depend on claim 7 and therefore inherit the same rejection.
Claim 12 recites the limitation "…the probability risk …” on Page 5 (lines 10 and 11). There is insufficient antecedent basis for this limitation in the claim. Claims 13 - 16 depend on claim 12 and therefore inherit the same rejection.
Also, claim 2 is indefinite because on Page 2 (line 15), it is unclear as to what the phrase “… a child parent node…” means? What node does it refer to? Is it the child node or the parent node? Or is there any other special meaning for a “child parent node”? Hence, it is indefinite. Or perhaps it is a typo. But for the purposes of examination, Examiner is treating it as a “child node”. Claims 3 - 6 depend on claim 2 and therefore inherit the same rejection.
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 2 - 16 are rejected under 35 U.S.C. 101 because the claimed invention is directed towards abstract ideas without significantly more.
Regarding claims 2 - 6:
According to the first step (Step 1) of the 101 analysis, claims 2 - 6 are directed to a system (manufacture) and falls within one of the four statutory categories (i.e., process, machine, manufacture, or composition of matter).
Regarding claim 2:
In the next step (Step 2A, prong 1) of the analysis, the limitations of:
assign each decision node in a tree-based decision model to an adverse action code, the tree-based decision model using one or more factors to determine a fraud score for a transaction;
assign a probability of risk for each adverse action code;
determine, for the tree-based decision model, a difference in the probability of risk between child nodes and respective parent nodes beginning with a cell in the tree-based decision model associated with the probability of risk for the transaction by subtracting the probability risk associated with a parent node from the probability risk associated with a child parent node;
for each of the decision nodes having an adverse action code, associated with the one or more factors used to determine the fraud score for the transaction update the probability of risk based on the difference;
determine a parent decision node associated with a greatest positive difference in risk;
determine the adverse action code associated with the determined parent decision node;
associate the determined adverse action code with a highest contributing factor of the one or more factors affecting the fraud score for the transaction;
The above limitations, under the broadest reasonable interpretation, the above limitations are process steps that cover mental processes including an observation, evaluation, judgment or opinion that could be performed in the mind or with the aid of pencil and paper but for the recitation of a generic computer component. If a claim, under its broadest reasonable interpretation, covers a mental process but for the recitation of generic computer components, then it falls within the “Mental Process” grouping of abstract ideas.
In the next step (Step 2A, prong 2) of the analysis, the limitations:
A risk determination system, the risk determination system comprising: a non-transitory data storage configured to store computer executable instructions for a risk determination system;
a hardware processor programmed to execute the computer executable instructions in the non-transitory data storage to cause the risk determination system to:
and populate one or more templates with adverse action code information;
and a client computing device comprising one or more client computing applications and a user interface, the one or more client computing applications configured to receive the populated one or more templates.
The above limitations are considered to be additional elements and it does not integrate the abstract idea into a practical application because the additional element is recited so generically (no details whatsoever are provided other than that it is a risk determination system that comprises a non-transitory data storage configured to store computer executable instructions for a risk determination system; a hardware processor programmed to execute the computer executable instructions in the non-transitory data storage to cause the risk determination system to: populate one or more templates with adverse action code information; and a client computing device comprising one or more client computing applications and a user interface, the one or more client computing applications configured to receive the populated one or more templates) that it represents no more than mere instructions to apply the judicial exception on a computer. As discussed in MPEP 2106.05(f), mere instructions to implement an abstract idea on a computer as a tool to perform an abstract idea is not indicative of integration into a practical application.
In the same step (Step 2A, prong 2) of the analysis, the limitation of:
cause the client computing device to display the adverse action code information on the user interface.
is considered to be an additional element and as recited represents insignificant extra-solution activity that is data output, because it is a mere nominal or tangential addition to the claim and is therefore not indicative of integration into a practical application. See MPEP 2106.05(g).
In the last step (Step 2B), the additional element does not amount to significantly more than the judicial exceptions. As explained with respect to Step 2A Prong Two, a risk determination system that comprises a non-transitory data storage configured to store computer executable instructions for a risk determination system; a hardware processor programmed to execute the computer executable instructions in the non-transitory data storage to cause the risk determination system to: populate one or more templates with adverse action code information; and a client computing device comprising one or more client computing applications and a user interface, the one or more client computing applications configured to receive the populated one or more templates is at best the equivalent of merely adding the words “apply it” to the judicial exception. See MPEP 2106.05(f). Mere instructions to apply an exception cannot provide an inventive concept and does not amount to significantly more than the judicial exception.
In the same step (Step 2B), the limitation of: cause the client computing device to display the adverse action code information on the user interface, amounts to insignificant extra solution activity because it is a mere nominal or tangential addition to the claim, amounting to mere data output (see MPEP 2106.05(g)). The courts have similarly found limitations directed to displaying a result, recited at a high level of generality, to be well-understood, routine, and conventional. See (MPEP 2106.05(d)(II), "presenting offers and gathering statistics.", “determining an estimated outcome and setting a price”). These limitations therefore remain insignificant extra-solution activity even upon reconsideration, and do not amount to significantly more. Even when considered in combination, these additional elements represent mere instructions to apply an exception and insignificant extra-solution activity, which cannot provide an inventive concept. The claim is not patent eligible.
Regarding claim 3:
In the next step (Step 2A, prong 2) of the analysis, the limitation:
wherein the tree-based decision model comprises at least one of: a random forest model or a gradient boosted model.
is considered to be an additional element and it does not integrate the abstract idea into a practical application because the additional element is recited so generically (no details whatsoever are provided other than that the risk determination system wherein the tree-based decision model comprises at least one of: a random forest model or a gradient boosted model) that it represents no more than mere instructions to apply the judicial exception on a computer. As discussed in MPEP 2106.05(f), mere instructions to implement an abstract idea on a computer as a tool to perform an abstract idea is not indicative of integration into a practical application.
In the last step (Step 2B) of the analysis, the additional element does not amount to significantly more than the judicial exceptions. As explained with respect to Step 2A Prong Two, risk determination system wherein the tree-based decision model comprises at least one of: a random forest model or a gradient boosted model is at best the equivalent of merely adding the words “apply it” to the judicial exception. See MPEP 2106.05(f). Mere instructions to apply an exception cannot provide an inventive concept and does not amount to significantly more than the judicial exception. The claim is not patent eligible.
Regarding claim 4:
In the step (Step 2A, prong 1) of the analysis, the limitation:
wherein parent decision nodes associated with differences in the probability of risk that are equal to zero or negative are not considered for the determining a parent decision node associated with the greatest difference in risk.
under the broadest reasonable interpretation, the above limitations is a step that covers mental processes including an observation, evaluation, judgment or opinion that could be performed in the mind or with the aid of pencil and paper but for the recitation of a generic computer component. If a claim, under its broadest reasonable interpretation, covers a mental process but for the recitation of generic computer components, then it falls within the “Mental Process” grouping of abstract ideas.
In the next step (Step 2A, prong 2) of the analysis, it does not integrate into a practical application because it does not add any additional elements that integrate the abstract idea into practical application.
In the last step (Step 2B) of the analysis, it does not add any additional elements that amount to significantly more than the abstract idea and thus fails to add an inventive concept. The claim is not patent eligible.
Regarding claim 5:
In the step (Step 2A, prong 1) of the analysis, the limitation:
wherein an absolute value of the difference in the probability of risk between child nodes and respective parent nodes is used to update the probability of risk.
under the broadest reasonable interpretation, the above limitations is a step that covers mental processes including an observation, evaluation, judgment or opinion that could be performed in the mind or with the aid of pencil and paper but for the recitation of a generic computer component. If a claim, under its broadest reasonable interpretation, covers a mental process but for the recitation of generic computer components, then it falls within the “Mental Process” grouping of abstract ideas.
In the next step (Step 2A, prong 2) of the analysis, it does not integrate into a practical application because it does not add any additional elements that integrate the abstract idea into practical application.
In the last step (Step 2B) of the analysis, it does not add any additional elements that amount to significantly more than the abstract idea and thus fails to add an inventive concept. The claim is not patent eligible.
Regarding claim 6:
In the step (Step 2A, prong 1) of the analysis, the limitations:
determine a next decision node associated with a next greatest difference in risk, the next greatest difference in risk being less than the greatest difference in risk and greater than other differences in risk that are associated with other decision nodes;
calculate the adverse action code associated with the next decision node;
and associate the calculated adverse action code with a next contributing factor of the one or more factors affecting the fraud score for the transaction.
under the broadest reasonable interpretation, the above limitations is a step that covers mental processes including an observation, evaluation, judgment or opinion that could be performed in the mind or with the aid of pencil and paper but for the recitation of a generic computer component. If a claim, under its broadest reasonable interpretation, covers a mental process but for the recitation of generic computer components, then it falls within the “Mental Process” grouping of abstract ideas.
In the next step (Step 2A, prong 2) of the analysis, it does not integrate into a practical application because it does not add any additional elements that integrate the abstract idea into practical application.
In the last step (Step 2B) of the analysis, it does not add any additional elements that amount to significantly more than the abstract idea and thus fails to add an inventive concept. The claim is not patent eligible.
Regarding claims 7 - 11:
According to the first step (Step 1) of the 101 analysis, claims 7 - 11 are directed to a computer-implemented method (process) and falls within one of the four statutory categories (i.e., process, machine, manufacture, or composition of matter).
Regarding claim 7:
In the next step (Step 2A, prong 1) of the analysis, the limitations of:
assigning each decision node in a tree-based decision model to an adverse action code, the tree-based decision model using one or more factors to determine a fraud score for a transaction;
assigning a probability of risk for each adverse action code;
determining, for the tree-based decision model, a difference in the probability of risk between child nodes and respective parent nodes beginning with a cell in the tree-based decision model associated with the probability of risk for the transaction by subtracting the probability risk associated with a parent node from the probability risk associated with a respective child node;
for each of the decision nodes having an adverse action code associated with the one or more factors used to determine the fraud score for the transaction, updating the probability of risk based on the difference;
determining a parent decision node associated with a greatest positive difference in risk;
determining the adverse action code associated with the determined parent decision node;
associating the determined adverse action code with a highest contributing factor of the one or more factors affecting the fraud score for the transaction;
The above limitations, under the broadest reasonable interpretation, the above limitations are process steps that cover mental processes including an observation, evaluation, judgment or opinion that could be performed in the mind or with the aid of pencil and paper but for the recitation of a generic computer component. If a claim, under its broadest reasonable interpretation, covers a mental process but for the recitation of generic computer components, then it falls within the “Mental Process” grouping of abstract ideas.
In the next step (Step 2A, prong 2) of the analysis, the limitations:
A computer-implemented method for determining action codes in a tree- based decision model, the computer-implemented method comprising, as implemented by one or more computing devices within a risk determination system configured with specific executable instructions:
populating one or more templates with adverse action code information;
The above limitations are considered to be additional elements and it does not integrate the abstract idea into a practical application because the additional element is recited so generically (no details whatsoever are provided other than that it is a computer-implemented method for determining action codes in a tree- based decision model, the computer-implemented method comprising, as implemented by one or more computing devices within a risk determination system configured with specific executable instructions and populating one or more templates with adverse action code information) that it represents no more than mere instructions to apply the judicial exception on a computer. As discussed in MPEP 2106.05(f), mere instructions to implement an abstract idea on a computer as a tool to perform an abstract idea is not indicative of integration into a practical application.
In the same step (Step 2A, prong 2) of the analysis, the limitation of:
receiving at a client computing device the populated one or more templates;
is considered to be an additional element and as recited represent insignificant extra-solution activity because it is mere data gathering. See MPEP 2106.05(g), discussing limitations that the Federal Circuit has considered to be insignificant extra-solution activity.
In the same step (Step 2A, prong 2) of the analysis, the limitation of:
displaying the adverse action code information on a user interface associated with the client computing device.
is considered to be an additional element and as recited represents insignificant extra-solution activity that is data output, because it is a mere nominal or tangential addition to the claim and is therefore not indicative of integration into a practical application. See MPEP 2106.05(g).
In the last step (Step 2B), the additional element does not amount to significantly more than the judicial exceptions. As explained with respect to Step 2A Prong Two, a computer-implemented method for determining action codes in a tree- based decision model, the computer-implemented method comprising, as implemented by one or more computing devices within a risk determination system configured with specific executable instructions and populating one or more templates with adverse action code information is at best the equivalent of merely adding the words “apply it” to the judicial exception. See MPEP 2106.05(f). Mere instructions to apply an exception cannot provide an inventive concept and does not amount to significantly more than the judicial exception.
In the same step (Step 2B), the limitation of: receiving at a client computing device the populated one or more templates, which is recited at a high level of generality and amounts to extra-solution activity of receiving data i.e. pre-solution activity of gathering data for use in the claimed process. The courts have found limitations directed to obtaining information electronically, recited at a high level of generality, to be well-understood, routine, and conventional (see MPEP 2106.05(d)(II), “receiving or transmitting data over a network”, "electronic record keeping," and "storing and retrieving information in memory"). These limitations therefore remain insignificant extra-solution activity even upon reconsideration, and do not amount to significantly more.
In the same step (Step 2B), the limitation of: displaying the adverse action code information on a user interface associated with the client computing device, amounts to insignificant extra solution activity because it is a mere nominal or tangential addition to the claim, amounting to mere data output (see MPEP 2106.05(g)). The courts have similarly found limitations directed to displaying a result, recited at a high level of generality, to be well-understood, routine, and conventional. See (MPEP 2106.05(d)(II), "presenting offers and gathering statistics.", “determining an estimated outcome and setting a price”). These limitations therefore remain insignificant extra-solution activity even upon reconsideration, and do not amount to significantly more. Even when considered in combination, these additional elements represent mere instructions to apply an exception and insignificant extra-solution activity, which cannot provide an inventive concept. The claim is not patent eligible.
Regarding claim 8:
In the next step (Step 2A, prong 2) of the analysis, the limitation:
wherein the tree-based decision model comprises at least one of: a random forest model or a gradient boosted model.
is considered to be an additional element and it does not integrate the abstract idea into a practical application because the additional element is recited so generically (no details whatsoever are provided other than that the computer-implemented method uses the tree-based decision model comprises at least one of: a random forest model or a gradient boosted model) that it represents no more than mere instructions to apply the judicial exception on a computer. As discussed in MPEP 2106.05(f), mere instructions to implement an abstract idea on a computer as a tool to perform an abstract idea is not indicative of integration into a practical application.
In the last step (Step 2B) of the analysis, the additional element does not amount to significantly more than the judicial exceptions. As explained with respect to Step 2A Prong Two, the computer-implemented method wherein the tree-based decision model comprises at least one of: a random forest model or a gradient boosted model is at best the equivalent of merely adding the words “apply it” to the judicial exception. See MPEP 2106.05(f). Mere instructions to apply an exception cannot provide an inventive concept and does not amount to significantly more than the judicial exception. The claim is not patent eligible.
Regarding claim 9:
In the step (Step 2A, prong 1) of the analysis, the limitation:
wherein parent decision nodes associated with differences in the probability of risk that are equal to zero or negative are not considered for determining a parent decision node associated with the greatest difference in risk.
under the broadest reasonable interpretation, the above limitations is a step that covers mental processes including an observation, evaluation, judgment or opinion that could be performed in the mind or with the aid of pencil and paper but for the recitation of a generic computer component. If a claim, under its broadest reasonable interpretation, covers a mental process but for the recitation of generic computer components, then it falls within the “Mental Process” grouping of abstract ideas.
In the next step (Step 2A, prong 2) of the analysis, it does not integrate into a practical application because it does not add any additional elements that integrate the abstract idea into practical application.
In the last step (Step 2B) of the analysis, it does not add any additional elements that amount to significantly more than the abstract idea and thus fails to add an inventive concept. The claim is not patent eligible.
Regarding claim 10:
In the step (Step 2A, prong 1) of the analysis, the limitation:
wherein an absolute value of the difference in the probability of risk between child nodes and respective parent nodes is used to update the probability of risk.
under the broadest reasonable interpretation, the above limitations is a step that covers mental processes including an observation, evaluation, judgment or opinion that could be performed in the mind or with the aid of pencil and paper but for the recitation of a generic computer component. If a claim, under its broadest reasonable interpretation, covers a mental process but for the recitation of generic computer components, then it falls within the “Mental Process” grouping of abstract ideas.
In the next step (Step 2A, prong 2) of the analysis, it does not integrate into a practical application because it does not add any additional elements that integrate the abstract idea into practical application.
In the last step (Step 2B) of the analysis, it does not add any additional elements that amount to significantly more than the abstract idea and thus fails to add an inventive concept. The claim is not patent eligible.
Regarding claim 11:
In the step (Step 2A, prong 1) of the analysis, the limitations:
determining a next decision node associated with a next greatest difference in risk, the next greatest difference in risk being less than the greatest difference in risk and greater than other differences in risk that are associated with other decision nodes;
calculating the adverse action code associated with the next decision node;
and associating the calculated adverse action code with a next contributing factor of the one or more factors affecting the fraud score for the transaction.
under the broadest reasonable interpretation, the above limitations is a step that covers mental processes including an observation, evaluation, judgment or opinion that could be performed in the mind or with the aid of pencil and paper but for the recitation of a generic computer component. If a claim, under its broadest reasonable interpretation, covers a mental process but for the recitation of generic computer components, then it falls within the “Mental Process” grouping of abstract ideas.
In the next step (Step 2A, prong 2) of the analysis, it does not integrate into a practical application because it does not add any additional elements that integrate the abstract idea into practical application.
In the last step (Step 2B) of the analysis, it does not add any additional elements that amount to significantly more than the abstract idea and thus fails to add an inventive concept. The claim is not patent eligible.
Regarding claims 12 - 16:
According to the first step (Step 1) of the 101 analysis, claims 12 - 16 are directed to a Non-transitory computer readable medium storing computer executable instructions thereon, the computer executable instructions when executed cause a risk determination system to at least (manufacture) and falls within one of the four statutory categories (i.e., process, machine, manufacture, or composition of matter).
Regarding claim 12:
In the next step (Step 2A, prong 1) of the analysis, the limitations of:
assign each decision node in a tree-based decision model to an adverse action code, the tree-based decision model using one or more factors to determine a fraud score for a transaction;
assign a probability of risk for each adverse action code;
determine, for the tree-based decision model, a difference in the probability of risk between child nodes and respective parent nodes beginning with a cell in the tree-based decision model associated with the probability of risk for the transaction by subtracting the probability risk associated with a parent node from the probability risk associated with a respective child node;
for each of the decision nodes having an adverse action code, associated with the one or more factors used to determine the fraud score for the transaction update the probability of risk based on the difference;
determine a parent decision node associated with a greatest positive difference in risk;
determine the adverse action code associated with the determined parent decision node;
associate the determined adverse action code with a highest contributing factor of the one or more factors affecting the fraud score for the transaction;
The above limitations, under the broadest reasonable interpretation, the above limitations are process steps that cover mental processes including an observation, evaluation, judgment or opinion that could be performed in the mind or with the aid of pencil and paper but for the recitation of a generic computer component. If a claim, under its broadest reasonable interpretation, covers a mental process but for the recitation of generic computer components, then it falls within the “Mental Process” grouping of abstract ideas.
In the next step (Step 2A, prong 2) of the analysis, the limitations:
Non-transitory computer readable medium storing computer executable instructions thereon, the computer executable instructions when executed cause a risk determination system to at least:
and populate one or more templates with adverse action code information;
The above limitations are considered to be additional elements and it does not integrate the abstract idea into a practical application because the additional element is recited so generically (no details whatsoever are provided other than that it is a Non-transitory computer readable medium storing computer executable instructions thereon, the computer executable instructions when executed cause a risk determination system to at least: populate one or more templates with adverse action code information; and a client computing device comprising one or more client computing applications and a user interface, the one or more client computing applications configured to receive the populated one or more templates) that it represents no more than mere instructions to apply the judicial exception on a computer. As discussed in MPEP 2106.05(f), mere instructions to implement an abstract idea on a computer as a tool to perform an abstract idea is not indicative of integration into a practical application.
In the same step (Step 2A, prong 2) of the analysis, the limitation of:
receiving at a client computing device the populated one or more templates;
is considered to be an additional element and as recited represent insignificant extra-solution activity because it is mere data gathering. See MPEP 2106.05(g), discussing limitations that the Federal Circuit has considered to be insignificant extra-solution activity.
In the same step (Step 2A, prong 2) of the analysis, the limitation of:
display the adverse action code information on the user interface associated with the client computing device.
is considered to be an additional element and as recited represents insignificant extra-solution activity that is data output, because it is a mere nominal or tangential addition to the claim and is therefore not indicative of integration into a practical application. See MPEP 2106.05(g).
In the last step (Step 2B), the additional element does not amount to significantly more than the judicial exceptions. As explained with respect to Step 2A Prong Two Non-transitory computer readable medium storing computer executable instructions thereon, the computer executable instructions when executed cause a risk determination system to at least: populate one or more templates with adverse action code information; is at best the equivalent of merely adding the words “apply it” to the judicial exception. See MPEP 2106.05(f). Mere instructions to apply an exception cannot provide an inventive concept and does not amount to significantly more than the judicial exception.
In the same step (Step 2B), the limitation of: receiving at a client computing device the populated one or more templates, which is recited at a high level of generality and amounts to extra-solution activity of receiving data i.e. pre-solution activity of gathering data for use in the claimed process. The courts have found limitations directed to obtaining information electronically, recited at a high level of generality, to be well-understood, routine, and conventional (see MPEP 2106.05(d)(II), “receiving or transmitting data over a network”, "electronic record keeping," and "storing and retrieving information in memory"). These limitations therefore remain insignificant extra-solution activity even upon reconsideration, and do not amount to significantly more.
In the same step (Step 2B), the limitation of: display the adverse action code information on the user interface associated with the client computing device, amounts to insignificant extra solution activity because it is a mere nominal or tangential addition to the claim, amounting to mere data output (see MPEP 2106.05(g)). The courts have similarly found limitations directed to displaying a result, recited at a high level of generality, to be well-understood, routine, and conventional. See (MPEP 2106.05(d)(II), "presenting offers and gathering statistics.", “determining an estimated outcome and setting a price”). These limitations therefore remain insignificant extra-solution activity even upon reconsideration, and do not amount to significantly more. Even when considered in combination, these additional elements represent mere instructions to apply an exception and insignificant extra-solution activity, which cannot provide an inventive concept. The claim is not patent eligible.
Regarding claim 13:
In the next step (Step 2A, prong 2) of the analysis, the limitation:
wherein the tree-based decision model comprises at least one of: a random forest model or a gradient boosted model.
is considered to be an additional element and it does not integrate the abstract idea into a practical application because the additional element is recited so generically (no details whatsoever are provided other than that the risk determination system wherein the tree-based decision model comprises at least one of: a random forest model or a gradient boosted model) that it represents no more than mere instructions to apply the judicial exception on a computer. As discussed in MPEP 2106.05(f), mere instructions to implement an abstract idea on a computer as a tool to perform an abstract idea is not indicative of integration into a practical application.
In the last step (Step 2B) of the analysis, the additional element does not amount to significantly more than the judicial exceptions. As explained with respect to Step 2A Prong Two, risk determination system wherein the tree-based decision model comprises at least one of: a random forest model or a gradient boosted model is at best the equivalent of merely adding the words “apply it” to the judicial exception. See MPEP 2106.05(f). Mere instructions to apply an exception cannot provide an inventive concept and does not amount to significantly more than the judicial exception. The claim is not patent eligible.
Regarding claim 14:
In the step (Step 2A, prong 1) of the analysis, the limitation:
wherein parent decision nodes associated with differences in the probability of risk that are equal to zero or negative are not considered for determining a parent decision node associated with the greatest difference in risk.
under the broadest reasonable interpretation, the above limitations is a step that covers mental processes including an observation, evaluation, judgment or opinion that could be performed in the mind or with the aid of pencil and paper but for the recitation of a generic computer component. If a claim, under its broadest reasonable interpretation, covers a mental process but for the recitation of generic computer components, then it falls within the “Mental Process” grouping of abstract ideas.
In the next step (Step 2A, prong 2) of the analysis, it does not integrate into a practical application because it does not add any additional elements that integrate the abstract idea into practical application.
In the last step (Step 2B) of the analysis, it does not add any additional elements that amount to significantly more than the abstract idea and thus fails to add an inventive concept. The claim is not patent eligible.
Regarding claim 15:
In the step (Step 2A, prong 1) of the analysis, the limitation:
wherein an absolute value of the difference in the probability of risk between child nodes and respective parent nodes is used to update the probability of risk.
under the broadest reasonable interpretation, the above limitations is a step that covers mental processes including an observation, evaluation, judgment or opinion that could be performed in the mind or with the aid of pencil and paper but for the recitation of a generic computer component. If a claim, under its broadest reasonable interpretation, covers a mental process but for the recitation of generic computer components, then it falls within the “Mental Process” grouping of abstract ideas.
In the next step (Step 2A, prong 2) of the analysis, it does not integrate into a practical application because it does not add any additional elements that integrate the abstract idea into practical application.
In the last step (Step 2B) of the analysis, it does not add any additional elements that amount to significantly more than the abstract idea and thus fails to add an inventive concept. The claim is not patent eligible.
Regarding claim 16:
In the step (Step 2A, prong 1) of the analysis, the limitations:
determine a next decision node associated with a next greatest difference in risk, the next greatest difference in risk being less than the greatest difference in risk and greater than other differences in risk that are associated with other decision nodes;
calculate the adverse action code associated with the next decision node;
and associate the calculated adverse action code with a next contributing factor of the one or more factors affecting the fraud score for the transaction.
under the broadest reasonable interpretation, the above limitations is a step that covers mental processes including an observation, evaluation, judgment or opinion that could be performed in the mind or with the aid of pencil and paper but for the recitation of a generic computer component. If a claim, under its broadest reasonable interpretation, covers a mental process but for the recitation of generic computer components, then it falls within the “Mental Process” grouping of abstract ideas.
In the next step (Step 2A, prong 2) of the analysis, it does not integrate into a practical application because it does not add any additional elements that integrate the abstract idea into practical application.
In the last step (Step 2B) of the analysis, it does not add any additional elements that amount to significantly more than the abstract idea and thus fails to add an inventive concept. The claim is not patent eligible.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
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.
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.
Claims 2, 3, 5-8, 10-13, 15, and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Robida et al (US 7711636 B2) in view of Xiong et al (US 20160127319 A1) and further in view of Moore et al (Cached Sufficient Statistics for Efficient Machine Learning with Large Datasets, 1998).
Regarding claim 2:
Robida teaches: A risk determination system, the risk determination system comprising: a non-transitory data storage configured to store computer executable instructions for a risk determination system ([Column 4, Lines 38-47] The computing system 100 includes, for example, a personal computer that is IBM, Macintosh, or Linux/Unix compatible. In one embodiment, the exemplary computing system 100 includes a central processing unit ("CPU") 105, which may include a conventional microprocessor. The computing system 100 further includes a memory 130, such as random access memory ("RAM") for temporary storage of information and a read only memory ("ROM") for permanent storage of information, and a mass storage device 120, such as a hard drive, diskette, or optical media storage device);
a hardware processor programmed to execute the computer executable instructions in the non-transitory data storage to cause the risk determination system to ([Column 4, Lines 38-42] The computing system 100 includes, for example, a personal computer that is IBM, Macintosh, or Linux/Unix compatible. In one embodiment, the exemplary computing system 100 includes a central processing unit ("CPU") 105, which may include a conventional microprocessor):
assign each decision node in a tree-based decision model to an adverse action code ([Column 20, Lines 16-27] Although the embodiment of FIG. 16 begins the process of allocating adverse action codes at the final segment to which the individual is assigned and moves upward through the segmentation hierarchy, it is understood that the process of allocating adverse action codes to segments may be performed in the opposite direction, or in any other order. In one embodiment, adverse action code allotment begins at the first segmentation level, with the entire population segment 310 (FIG. 7), for example, and then moves to the children nodes, such as to the previous bankruptcy segment 410, then to the higher risk segment 510, and then to the higher bankruptcy risk segment 610),
assign a probability of risk for each adverse action code ([Column 18, Lines 62-67] For example, an adverse action code for an individual assigned to the higher bankruptcy risk segment 610 (FIG. 7) may indicate that the individual was assigned to the higher bankruptcy risk segment. Additionally, the individual assigned to the higher bankruptcy risk segment 610 may also receive an adverse action code indicating that [column 19, Lines 1-2] the individual was assigned to a higher risk segment, for example, the higher risk segment 510);
determine, for the tree-based decision model, a difference in the probability of risk between child nodes and respective parent nodes beginning with a cell in the tree-based decision model associated with the probability of risk for the transaction ([Column 6, Lines 7-15] A segmentation structure may include multiple segments arranged in a tree configuration, wherein certain segments are parents, or children, of other segments. A segment hierarchy includes the segment to which an individual is assigned and each of the parent segments to the assigned segment. FIG. 7, described in detail below, illustrates a segmentation structure having multiple levels of segments to which individuals may be assigned. [Column 18, Lines 13-19] In certain embodiments, adverse action code may indicate that a final risk score is less than the maximum partly because of the segment, or segment hierarchy, to which the individual was assigned. However, for different individuals, the actual affect of being assigned in a particular segment or in a segment hierarchy on the final risk score may be significantly different. [Column 20, Lines 6-11] Thus, the process of determining a percentage drop of the final risk score due to a penalty for assignment to a particular segment and allotment of adverse action codes based on the determined percentage may be performed for each segment in the segmentation hierarchy for the individual. [Column 20, Lines 56-67] Continuing to a block 1720, an adverse action code related to being assigned to the previous bankruptcy segment is allotted if the ratio of the penalty for assignment to the previous bankruptcy segment to the difference between the highest available final risk score and the actual final risk score is larger than a predetermined ratio. In the example of FIG. 17, the penalty for assignment to the previous bankruptcy segment is 20 and the difference between the highest final risk score and the actual final risk score is 50 (for example, 100-50=50). Thus, the determined ratio is 40%. In this example, one adverse action code is allotted to indicate segmentation to the previous bankruptcy segment if the ratio is greater than [Column 21, Lines 1-3] 12.5%. Because the determined ratio of 40% is greater than 12.5%, an adverse action code is assigned to indicate segmentation to the previous bankruptcy segment);
for each of the decision nodes having an adverse action code, update the probability of risk based on the difference ([Column 20, Lines 34-52] FIG. 17 is one embodiment of a flowchart illustrating an exemplary process of allocating adverse action codes to various segments in a segment hierarchy. FIG. 17 also includes an example of application of the general formulas described in the flowchart using exemplary data related to an exemplary individual. In the example illustrated in FIG. 17, it is assumed that the highest final risk score possible for an individual is 100, the penalty for being assigned to the previous bankruptcy segment 410 (FIG. 7) is 20, and the penalty for assignment to the higher bankruptcy risk segment 610 is 15. Thus, in the example discussed with reference to FIG. 17, for an individual assigned to the higher bankruptcy risk segment 610, the total possible final risk score is 65. For purposes of example, an individual assigned to the higher bankruptcy risk segment 610 and having a final score of 50, for example, having 15 points deducted for reasons other than being assigned to the higher bankruptcy risk segment 610, is discussed with reference to the adverse action code allotment method);
determine a parent decision node associated with a greatest positive difference in risk ([Column 18, Lines 16-23] However, for different individuals, the actual affect of being assigned in a particular segment or in a segment hierarchy on the final risk score may be significantly different. For example, for a first individual, assignment to lower bankruptcy risk segment 620 (FIG. 7) may have had a larger percentage impact on the individuals final risk score than for a second individual that was also assigned to the lower bankruptcy risk segment 620);
determine the adverse action code associated with the determined parent decision node ([Column 18, Lines 24-25] Thus, providing an adverse action code related to segmentation of the first individual may be appropriate. [Column 18, Lines 29-33] Accordingly, described herein with respect to FIGS. 15-17 are exemplary methods of allotting adverse action codes related to segmentation of an individual based on the relevance of the segmentation decision on the final risk score assigned to the individual);
associate the determined adverse action code with a highest contributing factor of the one or more factors affecting the fraud score for the transaction ([Column 6, Lines 20-25] For example, many customers request information regarding the factors that had the most impact on an individual's risk score. Thus, in one embodiment the computing system 100 selects one or more adverse action codes that are indicative of reasons that a particular score was assigned to an individual. [Column 18, Lines 53-57] The allotment of adverse action codes for various levels of a segmentation hierarchy may be determined based on several factors, such as the relative impact of assignments to each level of the segment hierarchy had on the final risk score for the individual);
and populate one or more templates with adverse action code information ([Column 19, Lines 59-63] Accordingly, after allotment of adverse action codes to the higher bankruptcy risk segment 610, the computing device 100 determines at block 1650 that additional parent groups in the segment hierarchy are present and additional adverse action code allotment should be considered);
and a client computing device comprising one or more client computing applications and a user interface, the one or more client computing applications configured to receive the populated one or more templates and cause the client computing device to display the adverse action code information on the user interface ([Column 4, Lines 66 & 67] The exemplary computing system 100 includes one or more commonly available input/output (I/O) devices and [Column 5, Lines 1-4] interfaces 110, such as a keyboard, mouse, touchpad, and printer. In one embodiment, the I/O devices and interfaces 110 include one or more display device, such as a monitor, that allows the visual presentation of data to a user. [Column 6, Lines 18-32] After assigning a score to an individual, the computing system 100 may also select and provide reasons related to why the individual was assigned a particular score. For example, many customers request information regarding the factors that had the most impact on an individual's risk score. Thus, in one embodiment the computing system 100 selects one or more adverse action codes that are indicative of reasons that a particular score was assigned to an individual. In certain embodiments, the assignment of an individual to a particular segment may be a factor that was relevant in arriving at the risk score for the individual. Thus, in one embodiment, one or more adverse action codes provided to a customer may be related to the assignment of the individual to a particular segment, or to particular segments in the segment hierarchy. [Column 20, Lines 11-15] After each of the segments in the segmentation hierarchy are considered for allotment analysis, the method continues from block 1650 to a block 1670, where the adverse action codes allotted to various segments are generated and provided to the customer).
However, Robida does not explicitly disclose: the tree-based decision model using one or more factors to determine a fraud score for a transaction; by subtracting the probability risk associated with a parent node from the probability risk associated with a child parent node; associated with the one or more factors used to determine the fraud score for the transaction.
Xiong teaches, in an analogous system: the tree-based decision model using one or more factors to determine a fraud score for a transaction ([0045] Use of the GBRT allows for more stable threes that have better performance than conventionally used random forest (RF) techniques. Regression trees may be used instead of decision trees. The regression trees may output different probabilities for fraud for different rules (e.g., instead of binary “yes” or “no” decisions), as represented by the weighting factor associated with each rule. In some embodiments, the probability of fraud from applying one or more applicable rules may be defined with respect to a “fraud score” that is determined as a function of the weighting factors associated with each applicable rule);
associated with the one or more factors used to determine the fraud score for the transaction ([0045] Use of the GBRT allows for more stable threes that have better performance than conventionally used random forest (RF) techniques. Regression trees may be used instead of decision trees. The regression trees may output different probabilities for fraud for different rules (e.g., instead of binary “yes” or “no” decisions), as represented by the weighting factor associated with each rule. In some embodiments, the probability of fraud from applying one or more applicable rules may be defined with respect to a “fraud score” that is determined as a function of the weighting factors associated with each applicable rule).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the risk determination system of Robida to incorporate the teachings of Xiong to use the tree-based decision model using one or more factors to determine a fraud score for a transaction; associated with the one or more factors used to determine the fraud score for the transaction. One would have been motivated to do this modification because doing so would give the benefit of outputting different probabilities for fraud for different rules as taught by Xiong [0045].
Moore teaches, in an analogous system: by subtracting the probability risk associated with a parent node from the probability risk associated with a child parent node ([Page 87, Last Paragraph] Imagine that all values of all attributes in the dataset are independent random binary variables, taking value 2 with probability p and taking value 1 with probability 1p. Then the further p is from 0:5, the smaller we can expect the ADtree to be. This is because, on average, the less common value of a Vary node will match fraction min(p; 1 p) of its parent's records. And, on average, the number of records matched at the kth level of the tree will be R(min(p; 1 p))k. Thus, the maximum level in the tree at which we may and a node matching one or more records is approximately b(log2 R)=(log2 q)c, where q = min(p; 1 p). And so the total number of nodes in the tree is approximately).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combined teachings of Robida and Xiong to incorporate the teachings of Moore to use subtracting the probability risk associated with a parent node from the probability risk associated with a child parent node. One would have been motivated to do this modification because doing so would give the benefit of bringing enormous savings in memory as taught by Moore [Page 88, Paragraph 1].
Regarding claim 3:
The system of Robida, Xiong, and Moore teaches: The risk determination system of Claim 2 (as shown above).
However, Robida does not explicitly disclose: wherein the tree-based decision model comprises at least one of: a random forest model or a gradient boosted model.
Xiong teaches, in an analogous system: wherein the tree-based decision model comprises at least one of: a random forest model or a gradient boosted model ([0012] FIG. 4 shows an example of a gradient boosting regression tree (GBRT) according to an embodiment of the present invention. [0049] Gradient boosting may be applied to generate multiple regression trees, such as in a sequential manner to maximize accuracy).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the system of Robida to incorporate the teachings of Xiong wherein the tree-based decision model comprises at least one of: a random forest model or a gradient boosted model. One would have been motivated to do this modification because doing so would give the benefit of generating multiple regression trees, such as in a sequential manner to maximize accuracy as taught by Xiong [0049].
Regarding claim 5:
The system of Robida, Xiong, and Moore teaches: The risk determination system of Claim 2 (as shown above).
Robida further teaches: wherein an absolute value of the difference in the probability of risk between child nodes and respective parent nodes is used to update the probability of risk ([Column 20, Lines 34-52] FIG. 17 is one embodiment of a flowchart illustrating an exemplary process of allocating adverse action codes to various segments in a segment hierarchy. FIG. 17 also includes an example of application of the general formulas described in the flowchart using exemplary data related to an exemplary individual. In the example illustrated in FIG. 17, it is assumed that the highest final risk score possible for an individual is 100, the penalty for being assigned to the previous bankruptcy segment 410 (FIG. 7) is 20, and the penalty for assignment to the higher bankruptcy risk segment 610 is 15. Thus, in the example discussed with reference to FIG. 17, for an individual assigned to the higher bankruptcy risk segment 610, the total possible final risk score is 65. For purposes of example, an individual assigned to the higher bankruptcy risk segment 610 and having a final score of 50, for example, having 15 points deducted for reasons other than being assigned to the higher bankruptcy risk segment 610, is discussed with reference to the adverse action code allotment method. Note: All the numbers shown are absolute numbers as they are all positive).
Regarding claim 6:
The system of Robida, Xiong, and Moore teaches: The risk determination system of Claim 2 (as shown above).
Robida further teaches: wherein the risk determination system is further caused to: determine a next decision node associated with a next greatest difference in risk, the next greatest difference in risk being less than the greatest difference in risk and greater than other differences in risk that are associated with other decision nodes ([Column 21, Lines 8-18] Moving to a block 1730, an adverse action code related to being assigned to a subgroup, or segment configured as a child of the previous bankruptcy segment, is allotted if the ratio of the penalty for assignment to the particular subgroup to the difference in the highest available final risk score and the actual final risk score is larger than a predetermined ratio. In the example of FIG. 17, the penalty for assignment to the higher bankruptcy risk segment 610 is 15 and a difference between the highest final risk score and the actual final risk score is 50 (for example, 100-50=50). Accordingly, the determined ratio is 30%);
calculate the adverse action code associated with the next decision node ([Column 21, Lines 18-25] In this example, if the ratio is between 12.5% and 37.5%, one adverse action code is allotted to indicate segmentation to the subgroup; and if the ratio is greater than 37.5%, two adverse action codes are allotted to indicate segmentation to the subgroup. Using the exemplary figures provided herein, the ratio is 30% and, thus, one adverse action code is allotted for indicating segmentation to the higher bankruptcy risk segment 610).
However, Robida does not explicitly disclose: and associate the calculated adverse action code with a next contributing factor of the one or more factors affecting the fraud score for the transaction.
Xiong further teaches, in an analogous system: and associate the calculated adverse action code with a next contributing factor of the one or more factors affecting the fraud score for the transaction ([0045] Use of the GBRT allows for more stable threes that have better performance than conventionally used random forest (RF) techniques. Regression trees may be used instead of decision trees. The regression trees may output different probabilities for fraud for different rules (e.g., instead of binary “yes” or “no” decisions), as represented by the weighting factor associated with each rule. In some embodiments, the probability of fraud from applying one or more applicable rules may be defined with respect to a “fraud score” that is determined as a function of the weighting factors associated with each applicable rule. [0046] For example, in some embodiments, the fraud score for a set of variable values of a transaction when applying rule 0, rule 3 and rule 10 (e.g., as determined by the variable values applied to a regression tree) may be given by: Fraud score=1/1+e.sup.(−(w0*r0)−(w3*r3)−(w10*r10)+bias) [0047] Here, (w0*r0) is the weight of the rule 0, (w3*r3) is the weight of the rule 3, (w10*r10) is the weight of a rule 10, and bias is a (e.g., optional) constant value output from the machine learning algorithm. Similarly, the fraud score equation may be applied to different sets of rules and their associated values, and may be defined by the equation: Fraud Score=1/1+e.sup.(−(Σwi)+bias)).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the system of Robida to incorporate the teachings of Xiong and associate the calculated adverse action code with a next contributing factor of the one or more factors affecting the fraud score for the transaction. One would have been motivated to do this modification because doing so would give the benefit of the fraud score equation being applied to different sets of rules and their associated values as taught by Xiong [0047].
Regarding claim 7:
Robida teaches: A computer-implemented method for determining action codes in a tree- based decision model, the computer-implemented method comprising, as implemented by one or more computing devices within a risk determination system configured with specific executable instructions ([Column 4, Lines 38-47] The computing system 100 includes, for example, a personal computer that is IBM, Macintosh, or Linux/Unix compatible. In one embodiment, the exemplary computing system 100 includes a central processing unit ("CPU") 105, which may include a conventional microprocessor. The computing system 100 further includes a memory 130, such as random access memory ("RAM") for temporary storage of information and a read only memory ("ROM") for permanent storage of information, and a mass storage device 120, such as a hard drive, diskette, or optical media storage device);
assigning each decision node in a tree-based decision model to an adverse action code ([Column 20, Lines 16-27] Although the embodiment of FIG. 16 begins the process of allocating adverse action codes at the final segment to which the individual is assigned and moves upward through the segmentation hierarchy, it is understood that the process of allocating adverse action codes to segments may be performed in the opposite direction, or in any other order. In one embodiment, adverse action code allotment begins at the first segmentation level, with the entire population segment 310 (FIG. 7), for example, and then moves to the children nodes, such as to the previous bankruptcy segment 410, then to the higher risk segment 510, and then to the higher bankruptcy risk segment 610),
assigning a probability of risk for each adverse action code ([Column 18, Lines 62-67] For example, an adverse action code for an individual assigned to the higher bankruptcy risk segment 610 (FIG. 7) may indicate that the individual was assigned to the higher bankruptcy risk segment. Additionally, the individual assigned to the higher bankruptcy risk segment 610 may also receive an adverse action code indicating that [column 19, Lines 1-2] the individual was assigned to a higher risk segment, for example, the higher risk segment 510);
determining, for the tree-based decision model, a difference in the probability of risk between child nodes and respective parent nodes beginning with a cell in the tree-based decision model associated with the probability of risk for the transaction ([Column 6, Lines 7-15] A segmentation structure may include multiple segments arranged in a tree configuration, wherein certain segments are parents, or children, of other segments. A segment hierarchy includes the segment to which an individual is assigned and each of the parent segments to the assigned segment. FIG. 7, described in detail below, illustrates a segmentation structure having multiple levels of segments to which individuals may be assigned. [Column 18, Lines 13-19] In certain embodiments, adverse action code may indicate that a final risk score is less than the maximum partly because of the segment, or segment hierarchy, to which the individual was assigned. However, for different individuals, the actual affect of being assigned in a particular segment or in a segment hierarchy on the final risk score may be significantly different. [Column 20, Lines 6-11] Thus, the process of determining a percentage drop of the final risk score due to a penalty for assignment to a particular segment and allotment of adverse action codes based on the determined percentage may be performed for each segment in the segmentation hierarchy for the individual. [Column 20, Lines 56-67] Continuing to a block 1720, an adverse action code related to being assigned to the previous bankruptcy segment is allotted if the ratio of the penalty for assignment to the previous bankruptcy segment to the difference between the highest available final risk score and the actual final risk score is larger than a predetermined ratio. In the example of FIG. 17, the penalty for assignment to the previous bankruptcy segment is 20 and the difference between the highest final risk score and the actual final risk score is 50 (for example, 100-50=50). Thus, the determined ratio is 40%. In this example, one adverse action code is allotted to indicate segmentation to the previous bankruptcy segment if the ratio is greater than [Column 21, Lines 1-3] 12.5%. Because the determined ratio of 40% is greater than 12.5%, an adverse action code is assigned to indicate segmentation to the previous bankruptcy segment);
for each of the decision nodes having an adverse action code, update the probability of risk based on the difference ([Column 20, Lines 34-52] FIG. 17 is one embodiment of a flowchart illustrating an exemplary process of allocating adverse action codes to various segments in a segment hierarchy. FIG. 17 also includes an example of application of the general formulas described in the flowchart using exemplary data related to an exemplary individual. In the example illustrated in FIG. 17, it is assumed that the highest final risk score possible for an individual is 100, the penalty for being assigned to the previous bankruptcy segment 410 (FIG. 7) is 20, and the penalty for assignment to the higher bankruptcy risk segment 610 is 15. Thus, in the example discussed with reference to FIG. 17, for an individual assigned to the higher bankruptcy risk segment 610, the total possible final risk score is 65. For purposes of example, an individual assigned to the higher bankruptcy risk segment 610 and having a final score of 50, for example, having 15 points deducted for reasons other than being assigned to the higher bankruptcy risk segment 610, is discussed with reference to the adverse action code allotment method);
determining a parent decision node associated with a greatest positive difference in risk ([Column 18, Lines 16-23] However, for different individuals, the actual affect of being assigned in a particular segment or in a segment hierarchy on the final risk score may be significantly different. For example, for a first individual, assignment to lower bankruptcy risk segment 620 (FIG. 7) may have had a larger percentage impact on the individuals final risk score than for a second individual that was also assigned to the lower bankruptcy risk segment 620);
determining the adverse action code associated with the determined parent decision node ([Column 18, Lines 24-25] Thus, providing an adverse action code related to segmentation of the first individual may be appropriate. [Column 18, Lines 29-33] Accordingly, described herein with respect to FIGS. 15-17 are exemplary methods of allotting adverse action codes related to segmentation of an individual based on the relevance of the segmentation decision on the final risk score assigned to the individual);
associating the determined adverse action code with a highest contributing factor of the one or more factors affecting the fraud score for the transaction ([Column 6, Lines 20-25] For example, many customers request information regarding the factors that had the most impact on an individual's risk score. Thus, in one embodiment the computing system 100 selects one or more adverse action codes that are indicative of reasons that a particular score was assigned to an individual. [Column 18, Lines 53-57] The allotment of adverse action codes for various levels of a segmentation hierarchy may be determined based on several factors, such as the relative impact of assignments to each level of the segment hierarchy had on the final risk score for the individual);
populating one or more templates with adverse action code information ([Column 19, Lines 59-63] Accordingly, after allotment of adverse action codes to the higher bankruptcy risk segment 610, the computing device 100 determines at block 1650 that additional parent groups in the segment hierarchy are present and additional adverse action code allotment should be considered);
receiving at a client computing device the populated one or more templates; and displaying the adverse action code information on a user interface associated with the client computing device ([Column 4, Lines 66 & 67] The exemplary computing system 100 includes one or more commonly available input/output (I/O) devices and [Column 5, Lines 1-4] interfaces 110, such as a keyboard, mouse, touchpad, and printer. In one embodiment, the I/O devices and interfaces 110 include one or more display device, such as a monitor, that allows the visual presentation of data to a user. [Column 6, Lines 18-32] After assigning a score to an individual, the computing system 100 may also select and provide reasons related to why the individual was assigned a particular score. For example, many customers request information regarding the factors that had the most impact on an individual's risk score. Thus, in one embodiment the computing system 100 selects one or more adverse action codes that are indicative of reasons that a particular score was assigned to an individual. In certain embodiments, the assignment of an individual to a particular segment may be a factor that was relevant in arriving at the risk score for the individual. Thus, in one embodiment, one or more adverse action codes provided to a customer may be related to the assignment of the individual to a particular segment, or to particular segments in the segment hierarchy. [Column 20, Lines 11-15] After each of the segments in the segmentation hierarchy are considered for allotment analysis, the method continues from block 1650 to a block 1670, where the adverse action codes allotted to various segments are generated and provided to the customer).
However, Robida does not explicitly disclose: the tree-based decision model using one or more factors to determine a fraud score for a transaction; by subtracting the probability risk associated with a parent node from the probability risk associated with a respective child node; associated with the one or more factors used to determine the fraud score for the transaction.
Xiong teaches, in an analogous system: the tree-based decision model using one or more factors to determine a fraud score for a transaction ([0045] Use of the GBRT allows for more stable threes that have better performance than conventionally used random forest (RF) techniques. Regression trees may be used instead of decision trees. The regression trees may output different probabilities for fraud for different rules (e.g., instead of binary “yes” or “no” decisions), as represented by the weighting factor associated with each rule. In some embodiments, the probability of fraud from applying one or more applicable rules may be defined with respect to a “fraud score” that is determined as a function of the weighting factors associated with each applicable rule);
associated with the one or more factors used to determine the fraud score for the transaction ([0045] Use of the GBRT allows for more stable threes that have better performance than conventionally used random forest (RF) techniques. Regression trees may be used instead of decision trees. The regression trees may output different probabilities for fraud for different rules (e.g., instead of binary “yes” or “no” decisions), as represented by the weighting factor associated with each rule. In some embodiments, the probability of fraud from applying one or more applicable rules may be defined with respect to a “fraud score” that is determined as a function of the weighting factors associated with each applicable rule).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the risk determination system of Robida to incorporate the teachings of Xiong to use the tree-based decision model using one or more factors to determine a fraud score for a transaction; associated with the one or more factors used to determine the fraud score for the transaction. One would have been motivated to do this modification because doing so would give the benefit of outputting different probabilities for fraud for different rules as taught by Xiong [0045].
Moore teaches, in an analogous system: by subtracting the probability risk associated with a parent node from the probability risk associated with a respective child node ([Page 87, Last Paragraph] Imagine that all values of all attributes in the dataset are independent random binary variables, taking value 2 with probability p and taking value 1 with probability 1p. Then the further p is from 0:5, the smaller we can expect the ADtree to be. This is because, on average, the less common value of a Vary node will match fraction min(p; 1 p) of its parent's records. And, on average, the number of records matched at the kth level of the tree will be R(min(p; 1 p))k. Thus, the maximum level in the tree at which we may and a node matching one or more records is approximately b(log2 R)=(log2 q)c, where q = min(p; 1 p). And so the total number of nodes in the tree is approximately).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combined teachings of Robida and Xiong to incorporate the teachings of Moore to use subtracting the probability risk associated with a parent node from the probability risk associated with a respective child node. One would have been motivated to do this modification because doing so would give the benefit of bringing enormous savings in memory as taught by Moore [Page 88, Paragraph 1].
Regarding claim 8:
The system of Robida, Xiong, and Moore teaches: The computer-implemented method of Claim 7 (as shown above).
However, Robida does not explicitly disclose: wherein the tree-based decision model comprises at least one of: a random forest model or a gradient boosted model.
Xiong teaches, in an analogous system: wherein the tree-based decision model comprises at least one of: a random forest model or a gradient boosted model ([0012] FIG. 4 shows an example of a gradient boosting regression tree (GBRT) according to an embodiment of the present invention. [0049] Gradient boosting may be applied to generate multiple regression trees, such as in a sequential manner to maximize accuracy).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the computer-implemented method of Robida to incorporate the teachings of Xiong wherein the tree-based decision model comprises at least one of: a random forest model or a gradient boosted model. One would have been motivated to do this modification because doing so would give the benefit of generating multiple regression trees, such as in a sequential manner to maximize accuracy as taught by Xiong [0049].
Regarding claim 10:
The system of Robida, Xiong, and Moore teaches: The computer-implemented method of Claim 7 (as shown above).
Robida further teaches: wherein an absolute value of the difference in the probability of risk between child nodes and respective parent nodes is used to update the probability of risk ([Column 20, Lines 34-52] FIG. 17 is one embodiment of a flowchart illustrating an exemplary process of allocating adverse action codes to various segments in a segment hierarchy. FIG. 17 also includes an example of application of the general formulas described in the flowchart using exemplary data related to an exemplary individual. In the example illustrated in FIG. 17, it is assumed that the highest final risk score possible for an individual is 100, the penalty for being assigned to the previous bankruptcy segment 410 (FIG. 7) is 20, and the penalty for assignment to the higher bankruptcy risk segment 610 is 15. Thus, in the example discussed with reference to FIG. 17, for an individual assigned to the higher bankruptcy risk segment 610, the total possible final risk score is 65. For purposes of example, an individual assigned to the higher bankruptcy risk segment 610 and having a final score of 50, for example, having 15 points deducted for reasons other than being assigned to the higher bankruptcy risk segment 610, is discussed with reference to the adverse action code allotment method. Note: All the numbers shown are absolute numbers as they are all positive).
Regarding claim 11:
The system of Robida, Xiong, and Moore teaches: The computer-implemented method of Claim 7 (as shown above).
Robida further teaches: determining a next decision node associated with a next greatest difference in risk, the next greatest difference in risk being less than the greatest difference in risk and greater than other differences in risk that are associated with other decision nodes ([Column 21, Lines 8-18] Moving to a block 1730, an adverse action code related to being assigned to a subgroup, or segment configured as a child of the previous bankruptcy segment, is allotted if the ratio of the penalty for assignment to the particular subgroup to the difference in the highest available final risk score and the actual final risk score is larger than a predetermined ratio. In the example of FIG. 17, the penalty for assignment to the higher bankruptcy risk segment 610 is 15 and a difference between the highest final risk score and the actual final risk score is 50 (for example, 100-50=50). Accordingly, the determined ratio is 30%);
calculating the adverse action code associated with the next decision node ([Column 21, Lines 18-25] In this example, if the ratio is between 12.5% and 37.5%, one adverse action code is allotted to indicate segmentation to the subgroup; and if the ratio is greater than 37.5%, two adverse action codes are allotted to indicate segmentation to the subgroup. Using the exemplary figures provided herein, the ratio is 30% and, thus, one adverse action code is allotted for indicating segmentation to the higher bankruptcy risk segment 610).
However, Robida does not explicitly disclose: and associating the calculated adverse action code with a next contributing factor of the one or more factors affecting the fraud score for the transaction.
Xiong further teaches, in an analogous system: and associating the calculated adverse action code with a next contributing factor of the one or more factors affecting the fraud score for the transaction ([0045] Use of the GBRT allows for more stable threes that have better performance than conventionally used random forest (RF) techniques. Regression trees may be used instead of decision trees. The regression trees may output different probabilities for fraud for different rules (e.g., instead of binary “yes” or “no” decisions), as represented by the weighting factor associated with each rule. In some embodiments, the probability of fraud from applying one or more applicable rules may be defined with respect to a “fraud score” that is determined as a function of the weighting factors associated with each applicable rule. [0046] For example, in some embodiments, the fraud score for a set of variable values of a transaction when applying rule 0, rule 3 and rule 10 (e.g., as determined by the variable values applied to a regression tree) may be given by: Fraud score=1/1+e.sup.(−(w0*r0)−(w3*r3)−(w10*r10)+bias) [0047] Here, (w0*r0) is the weight of the rule 0, (w3*r3) is the weight of the rule 3, (w10*r10) is the weight of a rule 10, and bias is a (e.g., optional) constant value output from the machine learning algorithm. Similarly, the fraud score equation may be applied to different sets of rules and their associated values, and may be defined by the equation: Fraud Score=1/1+e.sup.(−(Σwi)+bias)).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the computer-implemented method of Robida to incorporate the teachings of Xiong and associate the calculated adverse action code with a next contributing factor of the one or more factors affecting the fraud score for the transaction. One would have been motivated to do this modification because doing so would give the benefit of the fraud score equation being applied to different sets of rules and their associated values as taught by Xiong [0047].
Regarding claim 12:
Robida teaches: Non-transitory computer readable medium storing computer executable instructions thereon, the computer executable instructions when executed cause a risk determination system to at least ([Column 4, Lines 38-47] The computing system 100 includes, for example, a personal computer that is IBM, Macintosh, or Linux/Unix compatible. In one embodiment, the exemplary computing system 100 includes a central processing unit ("CPU") 105, which may include a conventional microprocessor. The computing system 100 further includes a memory 130, such as random access memory ("RAM") for temporary storage of information and a read only memory ("ROM") for permanent storage of information, and a mass storage device 120, such as a hard drive, diskette, or optical media storage device);
assign each decision node in a tree-based decision model to an adverse action code ([Column 20, Lines 16-27] Although the embodiment of FIG. 16 begins the process of allocating adverse action codes at the final segment to which the individual is assigned and moves upward through the segmentation hierarchy, it is understood that the process of allocating adverse action codes to segments may be performed in the opposite direction, or in any other order. In one embodiment, adverse action code allotment begins at the first segmentation level, with the entire population segment 310 (FIG. 7), for example, and then moves to the children nodes, such as to the previous bankruptcy segment 410, then to the higher risk segment 510, and then to the higher bankruptcy risk segment 610),
assign a probability of risk for each adverse action code ([Column 18, Lines 62-67] For example, an adverse action code for an individual assigned to the higher bankruptcy risk segment 610 (FIG. 7) may indicate that the individual was assigned to the higher bankruptcy risk segment. Additionally, the individual assigned to the higher bankruptcy risk segment 610 may also receive an adverse action code indicating that [column 19, Lines 1-2] the individual was assigned to a higher risk segment, for example, the higher risk segment 510);
determine, for the tree-based decision model, a difference in the probability of risk between child nodes and respective parent nodes beginning with a cell in the tree-based decision model associated with the probability of risk for the transaction ([Column 6, Lines 7-15] A segmentation structure may include multiple segments arranged in a tree configuration, wherein certain segments are parents, or children, of other segments. A segment hierarchy includes the segment to which an individual is assigned and each of the parent segments to the assigned segment. FIG. 7, described in detail below, illustrates a segmentation structure having multiple levels of segments to which individuals may be assigned. [Column 18, Lines 13-19] In certain embodiments, adverse action code may indicate that a final risk score is less than the maximum partly because of the segment, or segment hierarchy, to which the individual was assigned. However, for different individuals, the actual affect of being assigned in a particular segment or in a segment hierarchy on the final risk score may be significantly different. [Column 20, Lines 6-11] Thus, the process of determining a percentage drop of the final risk score due to a penalty for assignment to a particular segment and allotment of adverse action codes based on the determined percentage may be performed for each segment in the segmentation hierarchy for the individual. [Column 20, Lines 56-67] Continuing to a block 1720, an adverse action code related to being assigned to the previous bankruptcy segment is allotted if the ratio of the penalty for assignment to the previous bankruptcy segment to the difference between the highest available final risk score and the actual final risk score is larger than a predetermined ratio. In the example of FIG. 17, the penalty for assignment to the previous bankruptcy segment is 20 and the difference between the highest final risk score and the actual final risk score is 50 (for example, 100-50=50). Thus, the determined ratio is 40%. In this example, one adverse action code is allotted to indicate segmentation to the previous bankruptcy segment if the ratio is greater than [Column 21, Lines 1-3] 12.5%. Because the determined ratio of 40% is greater than 12.5%, an adverse action code is assigned to indicate segmentation to the previous bankruptcy segment);
for each of the decision nodes having an adverse action code, update the probability of risk based on the difference ([Column 20, Lines 34-52] FIG. 17 is one embodiment of a flowchart illustrating an exemplary process of allocating adverse action codes to various segments in a segment hierarchy. FIG. 17 also includes an example of application of the general formulas described in the flowchart using exemplary data related to an exemplary individual. In the example illustrated in FIG. 17, it is assumed that the highest final risk score possible for an individual is 100, the penalty for being assigned to the previous bankruptcy segment 410 (FIG. 7) is 20, and the penalty for assignment to the higher bankruptcy risk segment 610 is 15. Thus, in the example discussed with reference to FIG. 17, for an individual assigned to the higher bankruptcy risk segment 610, the total possible final risk score is 65. For purposes of example, an individual assigned to the higher bankruptcy risk segment 610 and having a final score of 50, for example, having 15 points deducted for reasons other than being assigned to the higher bankruptcy risk segment 610, is discussed with reference to the adverse action code allotment method);
determine a parent decision node associated with a greatest positive difference in risk ([Column 18, Lines 16-23] However, for different individuals, the actual affect of being assigned in a particular segment or in a segment hierarchy on the final risk score may be significantly different. For example, for a first individual, assignment to lower bankruptcy risk segment 620 (FIG. 7) may have had a larger percentage impact on the individuals final risk score than for a second individual that was also assigned to the lower bankruptcy risk segment 620);
determine the adverse action code associated with the determined parent decision node ([Column 18, Lines 24-25] Thus, providing an adverse action code related to segmentation of the first individual may be appropriate. [Column 18, Lines 29-33] Accordingly, described herein with respect to FIGS. 15-17 are exemplary methods of allotting adverse action codes related to segmentation of an individual based on the relevance of the segmentation decision on the final risk score assigned to the individual);
associate the determined adverse action code with a highest contributing factor of the one or more factors affecting the fraud score for the transaction ([Column 6, Lines 20-25] For example, many customers request information regarding the factors that had the most impact on an individual's risk score. Thus, in one embodiment the computing system 100 selects one or more adverse action codes that are indicative of reasons that a particular score was assigned to an individual. [Column 18, Lines 53-57] The allotment of adverse action codes for various levels of a segmentation hierarchy may be determined based on several factors, such as the relative impact of assignments to each level of the segment hierarchy had on the final risk score for the individual);
and populate one or more templates with adverse action code information ([Column 19, Lines 59-63] Accordingly, after allotment of adverse action codes to the higher bankruptcy risk segment 610, the computing device 100 determines at block 1650 that additional parent groups in the segment hierarchy are present and additional adverse action code allotment should be considered);
receive at a client computing device the populated one or more templates; and display the adverse action code information on a user interface associated with the client computing device ([Column 4, Lines 66 & 67] The exemplary computing system 100 includes one or more commonly available input/output (I/O) devices and [Column 5, Lines 1-4] interfaces 110, such as a keyboard, mouse, touchpad, and printer. In one embodiment, the I/O devices and interfaces 110 include one or more display device, such as a monitor, that allows the visual presentation of data to a user. [Column 6, Lines 18-32] After assigning a score to an individual, the computing system 100 may also select and provide reasons related to why the individual was assigned a particular score. For example, many customers request information regarding the factors that had the most impact on an individual's risk score. Thus, in one embodiment the computing system 100 selects one or more adverse action codes that are indicative of reasons that a particular score was assigned to an individual. In certain embodiments, the assignment of an individual to a particular segment may be a factor that was relevant in arriving at the risk score for the individual. Thus, in one embodiment, one or more adverse action codes provided to a customer may be related to the assignment of the individual to a particular segment, or to particular segments in the segment hierarchy. [Column 20, Lines 11-15] After each of the segments in the segmentation hierarchy are considered for allotment analysis, the method continues from block 1650 to a block 1670, where the adverse action codes allotted to various segments are generated and provided to the customer).
However, Robida does not explicitly disclose: the tree-based decision model using one or more factors to determine a fraud score for a transaction; by subtracting the probability risk associated with a parent node from the probability risk associated with a respective child node; associated with the one or more factors used to determine the fraud score for the transaction.
Xiong teaches, in an analogous system: the tree-based decision model using one or more factors to determine a fraud score for a transaction ([0045] Use of the GBRT allows for more stable threes that have better performance than conventionally used random forest (RF) techniques. Regression trees may be used instead of decision trees. The regression trees may output different probabilities for fraud for different rules (e.g., instead of binary “yes” or “no” decisions), as represented by the weighting factor associated with each rule. In some embodiments, the probability of fraud from applying one or more applicable rules may be defined with respect to a “fraud score” that is determined as a function of the weighting factors associated with each applicable rule);
associated with the one or more factors used to determine the fraud score for the transaction ([0045] Use of the GBRT allows for more stable threes that have better performance than conventionally used random forest (RF) techniques. Regression trees may be used instead of decision trees. The regression trees may output different probabilities for fraud for different rules (e.g., instead of binary “yes” or “no” decisions), as represented by the weighting factor associated with each rule. In some embodiments, the probability of fraud from applying one or more applicable rules may be defined with respect to a “fraud score” that is determined as a function of the weighting factors associated with each applicable rule).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the risk determination system of Robida to incorporate the teachings of Xiong to use the tree-based decision model using one or more factors to determine a fraud score for a transaction; associated with the one or more factors used to determine the fraud score for the transaction. One would have been motivated to do this modification because doing so would give the benefit of outputting different probabilities for fraud for different rules as taught by Xiong [0045].
Moore teaches, in an analogous system: by subtracting the probability risk associated with a parent node from the probability risk associated with a respective child node ([Page 87, Last Paragraph] Imagine that all values of all attributes in the dataset are independent random binary variables, taking value 2 with probability p and taking value 1 with probability 1p. Then the further p is from 0:5, the smaller we can expect the ADtree to be. This is because, on average, the less common value of a Vary node will match fraction min(p; 1 p) of its parent's records. And, on average, the number of records matched at the kth level of the tree will be R(min(p; 1 p))k. Thus, the maximum level in the tree at which we may and a node matching one or more records is approximately b(log2 R)=(log2 q)c, where q = min(p; 1 p). And so the total number of nodes in the tree is approximately).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combined teachings of Robida and Xiong to incorporate the teachings of Moore to use subtracting the probability risk associated with a parent node from the probability risk associated with a respective child node. One would have been motivated to do this modification because doing so would give the benefit of bringing enormous savings in memory as taught by Moore [Page 88, Paragraph 1].
Regarding claim 13:
The system of Robida, Xiong, and Moore teaches: The non-transitory computer readable medium of Claim 12 (as shown above).
However, Robida does not explicitly disclose: wherein the tree-based decision model comprises at least one of: a random forest model or a gradient boosted model.
Xiong teaches, in an analogous system: wherein the tree-based decision model comprises at least one of: a random forest model or a gradient boosted model ([0012] FIG. 4 shows an example of a gradient boosting regression tree (GBRT) according to an embodiment of the present invention. [0049] Gradient boosting may be applied to generate multiple regression trees, such as in a sequential manner to maximize accuracy).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the system of Robida to incorporate the teachings of Xiong wherein the tree-based decision model comprises at least one of: a random forest model or a gradient boosted model. One would have been motivated to do this modification because doing so would give the benefit of generating multiple regression trees, such as in a sequential manner to maximize accuracy as taught by Xiong [0049].
Regarding claim 15:
The system of Robida, Xiong, and Moore teaches: The non-transitory computer readable medium of Claim 12 (as shown above).
Robida further teaches: wherein an absolute value of the difference in the probability of risk between child nodes and respective parent nodes is used to update the probability of risk ([Column 20, Lines 34-52] FIG. 17 is one embodiment of a flowchart illustrating an exemplary process of allocating adverse action codes to various segments in a segment hierarchy. FIG. 17 also includes an example of application of the general formulas described in the flowchart using exemplary data related to an exemplary individual. In the example illustrated in FIG. 17, it is assumed that the highest final risk score possible for an individual is 100, the penalty for being assigned to the previous bankruptcy segment 410 (FIG. 7) is 20, and the penalty for assignment to the higher bankruptcy risk segment 610 is 15. Thus, in the example discussed with reference to FIG. 17, for an individual assigned to the higher bankruptcy risk segment 610, the total possible final risk score is 65. For purposes of example, an individual assigned to the higher bankruptcy risk segment 610 and having a final score of 50, for example, having 15 points deducted for reasons other than being assigned to the higher bankruptcy risk segment 610, is discussed with reference to the adverse action code allotment method. Note: All the numbers shown are absolute numbers as they are all positive).
Regarding claim 16:
The system of Robida, Xiong, and Moore teaches: The non-transitory computer readable medium of Claim 12 (as shown above).
Robida further teaches: wherein the computer executable instructions when executed further cause the risk determination system to at least: determine a next decision node associated with a next greatest difference in risk, the next greatest difference in risk being less than the greatest difference in risk and greater than other differences in risk that are associated with other decision nodes ([Column 21, Lines 8-18] Moving to a block 1730, an adverse action code related to being assigned to a subgroup, or segment configured as a child of the previous bankruptcy segment, is allotted if the ratio of the penalty for assignment to the particular subgroup to the difference in the highest available final risk score and the actual final risk score is larger than a predetermined ratio. In the example of FIG. 17, the penalty for assignment to the higher bankruptcy risk segment 610 is 15 and a difference between the highest final risk score and the actual final risk score is 50 (for example, 100-50=50). Accordingly, the determined ratio is 30%);
calculate the adverse action code associated with the next decision node ([Column 21, Lines 18-25] In this example, if the ratio is between 12.5% and 37.5%, one adverse action code is allotted to indicate segmentation to the subgroup; and if the ratio is greater than 37.5%, two adverse action codes are allotted to indicate segmentation to the subgroup. Using the exemplary figures provided herein, the ratio is 30% and, thus, one adverse action code is allotted for indicating segmentation to the higher bankruptcy risk segment 610).
However, Robida does not explicitly disclose: and associate the calculated adverse action code with a next contributing factor of the one or more factors affecting the fraud score for the transaction.
Xiong further teaches, in an analogous system: and associate the calculated adverse action code with a next contributing factor of the one or more factors affecting the fraud score for the transaction ([0045] Use of the GBRT allows for more stable threes that have better performance than conventionally used random forest (RF) techniques. Regression trees may be used instead of decision trees. The regression trees may output different probabilities for fraud for different rules (e.g., instead of binary “yes” or “no” decisions), as represented by the weighting factor associated with each rule. In some embodiments, the probability of fraud from applying one or more applicable rules may be defined with respect to a “fraud score” that is determined as a function of the weighting factors associated with each applicable rule. [0046] For example, in some embodiments, the fraud score for a set of variable values of a transaction when applying rule 0, rule 3 and rule 10 (e.g., as determined by the variable values applied to a regression tree) may be given by: Fraud score=1/1+e.sup.(−(w0*r0)−(w3*r3)−(w10*r10)+bias) [0047] Here, (w0*r0) is the weight of the rule 0, (w3*r3) is the weight of the rule 3, (w10*r10) is the weight of a rule 10, and bias is a (e.g., optional) constant value output from the machine learning algorithm. Similarly, the fraud score equation may be applied to different sets of rules and their associated values, and may be defined by the equation: Fraud Score=1/1+e.sup.(−(Σwi)+bias)).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the system of Robida to incorporate the teachings of Xiong and associate the calculated adverse action code with a next contributing factor of the one or more factors affecting the fraud score for the transaction. One would have been motivated to do this modification because doing so would give the benefit of the fraud score equation being applied to different sets of rules and their associated values as taught by Xiong [0047].
Claims 4, 9, and 14 are rejected under 35 U.S.C. 103 as being unpatentable over Robida et al (US 7711636 B2) in view of Xiong et al (US 20160127319 A1) and Moore et al (Cached Sufficient Statistics for Efficient Machine Learning with Large Datasets, 1998) and further in view of Johnson et al (US 20100223211 A1).
Regarding claim 4:
The system of Robida, Xiong, and Moore teaches: The risk determination system of Claim 2 (as shown above).
However, the system of Robida, Xiong, and Moore does not explicitly disclose: wherein parent decision nodes associated with differences in the probability of risk that are equal to zero or negative are not considered for the determining a parent decision node associated with the greatest difference in risk.
Johnson further teaches, in an analogous system: wherein parent decision nodes associated with differences in the probability of risk that are equal to zero or negative are not considered for the determining a parent decision node associated with the greatest difference in risk ([Page 14] Computing Score Reasons Distances [0266] The end user can choose to compute distances based on maximum scores (distance=score weight-maximum score) or user-specified baseline scores (distance=score weight-baseline score). However, if the end user allows score weight expressions, the end user is limited to baseline scores. At runtime, the score reasons distances method returns score reasons using a summed distance or maximum distance. [0267] It is noted that only reason codes that have a positive distance are returned. Reason codes with zero or negative distances are not returned).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combined teachings of Robida, Xiong, and Moore to incorporate the teachings of Johnson wherein parent decision nodes associated with differences in the probability of risk that are equal to zero or negative are not considered for the determining a parent decision node associated with the greatest difference in risk. One would have been motivated to do this modification because doing so would give the benefit of the end user can choose to compute distances based on maximum scores as taught by Johnson [0266].
Regarding claim 9:
The system of Robida, Xiong, and Moore teaches: The computer-implemented method of Claim 7 (as shown above).
However, the system of Robida, Xiong, and Moore does not explicitly disclose: wherein parent decision nodes associated with differences in the probability of risk that are equal to zero or negative are not considered for determining a parent decision node associated with the greatest difference in risk.
Johnson further teaches, in an analogous system: wherein parent decision nodes associated with differences in the probability of risk that are equal to zero or negative are not considered for determining a parent decision node associated with the greatest difference in risk ([Page 14] Computing Score Reasons Distances [0266] The end user can choose to compute distances based on maximum scores (distance=score weight-maximum score) or user-specified baseline scores (distance=score weight-baseline score). However, if the end user allows score weight expressions, the end user is limited to baseline scores. At runtime, the score reasons distances method returns score reasons using a summed distance or maximum distance. [0267] It is noted that only reason codes that have a positive distance are returned. Reason codes with zero or negative distances are not returned).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combined teachings of Robida, Xiong, and Moore to incorporate the teachings of Johnson wherein parent decision nodes associated with differences in the probability of risk that are equal to zero or negative are not considered for determining a parent decision node associated with the greatest difference in risk. One would have been motivated to do this modification because doing so would give the benefit of the end user can choose to compute distances based on maximum scores as taught by Johnson [0266].
Regarding claim 14:
The system of Robida, Xiong, and Moore teaches: The non-transitory computer readable medium of Claim 12 (as shown above).
However, the system of Robida, Xiong, and Moore does not explicitly disclose: wherein parent decision nodes associated with differences in the probability of risk that are equal to zero or negative are not considered for determining a parent decision node associated with the greatest difference in risk.
Johnson further teaches, in an analogous system: wherein parent decision nodes associated with differences in the probability of risk that are equal to zero or negative are not considered for determining a parent decision node associated with the greatest difference in risk ([Page 14] Computing Score Reasons Distances [0266] The end user can choose to compute distances based on maximum scores (distance=score weight-maximum score) or user-specified baseline scores (distance=score weight-baseline score). However, if the end user allows score weight expressions, the end user is limited to baseline scores. At runtime, the score reasons distances method returns score reasons using a summed distance or maximum distance. [0267] It is noted that only reason codes that have a positive distance are returned. Reason codes with zero or negative distances are not returned).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combined teachings of Robida, Xiong, and Moore to incorporate the teachings of Johnson wherein parent decision nodes associated with differences in the probability of risk that are equal to zero or negative are not considered for determining a parent decision node associated with the greatest difference in risk. One would have been motivated to do this modification because doing so would give the benefit of the end user can choose to compute distances based on maximum scores as taught by Johnson [0266].
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Wallace et al (Coding Decision Trees, 1993) discloses that there is an error in their derivation of message lengths, which fortunately has no effect on the final inference. We further suggest two improvements to their coding techniques, one removing an inefficiency in the description of non-binary trees, and one improving the coding of leaves. We argue that these improvements are superior to similarly motivated proposals in the original paper. Empirical tests confirm the good results reported by Quinlan and Rivest, and show our coding proposals to lead to useful improvements in the performance of the method.
Choi et al (Learning Latent Tree Graphical Models, 2010) discloses the problem of learning a latent tree graphical model where samples are available only from a subset of variables. We propose two consistent and computationally efficient algorithms for learning minimal latent trees, that is, trees without any redundant hidden nodes. Unlike many existing methods, the observed nodes (or variables) are not constrained to be leaf nodes. Our first algorithm, recursive grouping, builds the latent tree recursively by identifying sibling groups using so-called information distances. One of the main contributions of this work is our second algorithm, which we refer to as CLGrouping. CLGrouping starts with a pre-processing procedure in which a tree over the observed variables is constructed. This global step groups the observed nodes that are likely to be close to each other in the true latent tree, thereby guiding subsequent recursive grouping (or equivalent procedures) on much smaller subsets of variables. This results in more accurate and efficient learning of latent trees. We also present regularized versions of our algorithms that learn latent tree approximations of arbitrary distributions. We compare the proposed algorithms to other methods by performing extensive numerical experiments on various latent tree graphical models such as hidden Markov models and star graphs. In addition, we demonstrate the applicability of our methods on real-world datasets by modeling the dependency structure of monthly stock returns in the S&P index and of the words in the 20 newsgroups dataset.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to CHAITANYA RAMESH JAYAKUMAR whose telephone number is (571)272-3369. The examiner can normally be reached Mon-Fri 9am-1pm.
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, Omar Fernandez Rivas can be reached at (571)272-2589. 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.
/C.R.J./Examiner, Art Unit 2128
/OMAR F FERNANDEZ RIVAS/Supervisory Patent Examiner, Art Unit 2128
Prosecution Insights