Granted Patent B2

US 12,321,911 · App. 18/197,618 · Granted Jun 3, 2025

Inventors: Christian Anderson (San Francisco, CA); Stanislas Polu (San Francisco, CA)

Assignee: STRIPE INC.

G06Q20/12G06Q20/027G06Q20/227G06Q20/405

Abstract

Claims (76)

1. A computer implemented method performed by a commerce platform system, comprising:

establishing, by a server system, first communications channels using a first set of application programming interfaces (APIs) that integrate the server system with a plurality of payment type provider platforms that provide a plurality of payment types;

receiving, by the server system, a request from a third-party server via a second communication channel, wherein the request comprises account data associated with an account, and wherein the third-party server and the server system communicate via a second set of APIs over the second communication channel to provide the third-party server access to the plurality of payment types;

generating a query comprising at least a portion of the account data;

executing the query at the plurality of payment type provider platforms;

receiving responses to the query from the plurality of payment type provider platforms, the response indicating whether at least a portion of the account data is associated with information in one or more data resources of the plurality of payment type provider platforms;

identifying available payment types for the account based on the responses;

executing a machine learning model to determine a relevancy score of each of the available payment types with respect to the account data, the relevancy score based at least in part on a probability of the account using each available payment type to complete a transaction, wherein the probability is calculated by the machine learning model;

ranking the available payment types based on the relevancy scores;

selecting, as one or more recommended payment types, available payment types having relevancy scores equal to or greater than a threshold or rank;

communicating, by the server system, the one or more recommended payment types to the third-party server causing the third-party server to provide the one or more recommended payment types within a user interface;

receiving, at the server system, a selection of one of the one or more recommended payment types; and

completing, by the server system, an electronic payment using the selected one of the one or more recommended payment types.

2. The method of claim 1 , wherein the machine learning model comprises one or more different machine learning models that each predict probability of the customer using that payment type, and wherein each of the one or more different machine learning models is constructed at varying levels of specificity relative to customer information and/or merchant information.

3. The method of claim 2 , wherein the varying levels of specificity comprise one or more of:

an individual customer level,

a cohort level in which customers sharing similar characteristics are grouped, a merchant level, and

a merchant cohort level in which merchants sharing similar characteristics are grouped.

4. The method of claim 1 , further comprising:

detecting, by the commerce platform system, a selected payment type to complete the transaction with the third-party server;

in response to determining, by the server system, that the selected payment type is associated with a low probability of being used by the account, determining that the transaction is likely to be fraudulent, and

performing one or more fraud protection operations.

5. The method of claim 4 , further comprising:

when the transaction is determined to likely be fraudulent, performing, by the server system, an authentication of the account prior to completion of the transaction.

6. The method of claim 4 , wherein ranking the plurality of payment types based on the value of the determined probabilities comprises:

assigning to each of the ranked plurality of payment types selection power index values that indicates a probability of said each of the ranked plurality of payment types being selected for the transaction;

using a selection power index value of the selected payment type to determine whether the selected payment type is associated with the low probability of selection.

7. The method of claim 6 , wherein the power index values are inversely related to the probability of fraud for the transaction.

8. The method of claim 1 , wherein the request is received from the third-party server in response to the transaction being initiated or completed at the third-party server that requests processing of a payment for the transaction.

9. The method of claim 8 , further comprising:

completing, by the server system, payment processing using a payment type selected for completion of the transaction.

10. The method of claim 1 , wherein the account comprises a customer data that identifies a customer, and the query determines which payment type provider platforms the customer has a relationship with based on which payment type provider platforms also store the customer data.

11. The method of claim 1 , wherein the account comprises a set of customer data for a customer, the customer data associated with parameters of the transaction, and the query determines which data from the set of customer data are stored by the payment type provider platforms, the method further comprising:

comparing information of the customer in the query to information of the customer in the one or more data resources to identify at least a threshold match; and

selecting payment types, associated with the information of the customer in the one or more data resources, that meet the threshold match, as the payment types that are relevant to the customer.

12. The method of claim 1 , wherein executing the query at the payment type provider platforms comprises:

hashing customer information to generate a hash value representative of the customer information, the hash value comprised in the query; and

comparing the hash value in the query to hashed customer information in the one or more data resources of the payment type provider platforms to identity the relevant payment types.

13. A non-transitory computer storage medium having computer-executable instructions stored thereon that, when executed by one or more processors, causes the one or more processors to perform operations of a commerce platform system, the operations comprising:

establishing, by a server system, first communications channels using a first set of application programming interfaces (APIs) that integrate the server system with a plurality of payment type provider platforms that provide a plurality of payment types;

receiving, by the server system, a request from a third-party server via a second communication channel, wherein the request comprises account data associated with an account, and wherein the third-party server and the server system communicate via a second set of APIs over the second communication channel to provide the third-party server access to the plurality of payment types;

generating a query comprising at least a portion of the account data;

executing the query at the plurality of payment type provider platforms;

receiving responses to the query from the plurality of payment type provider platforms, the response indicating whether at least a portion of the account data is associated with information in one or more data resources of the plurality of payment type provider platforms;

identifying available payment types for the account based on the responses;

executing a machine learning model to determine a relevancy score of each of the available payment types with respect to the account data, the relevancy score based at least in part on a probability of the account using each available payment type to complete a transaction, wherein the probability is calculated by the machine learning model;

ranking the available payment types based on the relevancy scores;

selecting, as one or more recommended payment types, available payment types having relevancy scores equal to or greater than a threshold or rank;

communicating, by the server system, the one or more recommended payment types to the third-party server causing the third-party server to provide the one or more recommended payment types within a user interface;

receiving, at the server system, a selection of one of the one or more recommended payment types; and

completing, by the server system, an electronic payment using the selected one of the one or more recommended payment types.

14. The non-transitory computer storage medium of claim 13 , wherein the machine learning model comprises one or more different machine learning models that each predict probability of the customer using that payment type, and wherein each of the one or more different machine learning models is constructed at varying levels of specificity relative to customer information and/or merchant information.

15. The non-transitory computer storage medium of claim 14 , wherein the varying levels of specificity comprise one or more of:

an individual customer level,

a cohort level in which customers sharing similar characteristics are grouped, a merchant level, and

a merchant cohort level in which merchants sharing similar characteristics are grouped.

16. The non-transitory computer storage medium of claim 13 , further comprising:

detecting, by the commerce platform system, a selected payment type to complete the transaction with the third-party server;

in response to determining, by the server system, that the selected payment type is associated with a low probability of being used by the account, determining that the transaction is likely to be fraudulent, and

performing one or more fraud protection operations.

17. The non-transitory computer storage medium of claim 16 , further comprising: when the transaction is determined to likely be fraudulent, performing, by the server system, an authentication of the account prior to completion of the transaction.

18. A commerce platform system, comprising

a memory; and

one or more processors coupled with the memory configured to perform operations comprising:

establishing, by a server system, first communications channels using a first set of application programming interfaces (APIs) that integrate the server system with a plurality of payment type provider platforms that provide a plurality of payment types;

receiving, by the server system, a request from a third-party server via a second communication channel, wherein the request comprises account data associated with an account, and wherein the third-party server and the server system communicate via a second set of APIs over the second communication channel to provide the third-party server access to the plurality of payment types;

generating a query comprising at least a portion of the account data;

executing the query at the plurality of payment type provider platforms;

receiving responses to the query from the plurality of payment type provider platforms, the response indicating whether at least a portion of the account data is associated with information in one or more data resources of the plurality of payment type provider platforms;

identifying available payment types for the account based on the responses;

executing a machine learning model to determine a relevancy score of each of the available payment types with respect to the account data, the relevancy score based at least in part on a probability of the account using each available payment type to complete a transaction, wherein the probability is calculated by the machine learning model;

ranking the available payment types based on the relevancy scores;

selecting, as one or more recommended payment types, available payment types having relevancy scores equal to or greater than a threshold or rank;

communicating, by the server system, the one or more recommended payment types to the third-party server causing the third-party server to provide the one or more recommended payment types within a user interface;

receiving, at the server system, a selection of one of the one or more recommended payment types; and

completing, by the server system, an electronic payment using the selected one of the one or more recommended payment types.

Assignments (2)

Continuity (3)

References Cited (92)