Adaptive learning system with a product configuration engine

• US 10,454,779 B2
• Filed: 08/26/2016
• Issued: 10/22/2019
• Est. Priority Date: 08/26/2016
• Status: Active Grant

• Current Assignee: PayPal, Inc. (PayPal Holdings, Inc.)
• Original Assignee: PayPal, Inc. (PayPal Holdings, Inc.)
• Inventors: Narasimhan, Srivathsan, Alagarsamy, Premkumar, Chatterjee, Avik
• Primary Examiner(s): Taylor, Nicholas R
• Assistant Examiner(s): Vang, Meng
• Application Number:
  US15/248,228
• Publication Number:
  US 20180062937A1
• Time in Patent Office: 1,152 Days
• Field of Search: 709224
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: H04L 41/0893 Assignment of logical groups to network elements
• CPC Class Codes: H04L 41/147 for predicting network behaviour
• CPC Class Codes: H04L 43/0876 Network utilisation, e.g. volume of load or congestion level

Abstract

The adaptive learning systems described herein may include machine-learning engines, product configuration engines, and/or various other components configured to improve the efficiency of processing transactions. The systems described herein may detect and/or predict declined transactions, token deficiencies, insufficient system capacities, and/or other system anomalies. As such, the system may perform operations to generate additional tokens associated with assets, provision various bin ranges, and/or share reserve capacities, and/or other operations. Thus, the system may improve system efficiencies, ensure reliability and operability across the system, and optimize the operations for successfully processing transactions.

Claims

1. A system, comprising:

a non-transitory memory; and

one or more hardware processors coupled to the non-transitory memory and configured to execute instructions from the non-transitory memory to cause the system to perform operations comprising;

determining token traffic in a network based at least on analyzing one or more token requests transmitted over the network during a period of time;

determining a token depletion rate based at least on the determined token traffic in the network and a number of tokens issued in response to token requests during the period of time;

generating a plurality of tokens in a token database based at least on the token depletion rate;

detecting one or more transaction processing anomalies in the network based on analyzing the token traffic;

generating a transaction processing policy based on the one or more detected transaction processing anomalies, wherein the transaction processing policy comprises an intermediate matrix that represents policy requirements in a binary form;

receiving, from a user device, a transaction request comprising a set of transaction data;

retrieving a token from the token database for the transaction request;

modifying the set of transaction data in the transaction request based on the generated transaction processing policy by applying the intermediate matrix to the transaction data; and

causing the transaction request to be processed based on the modified set of transaction data and the retrieved token.

2. The system of claim 1, wherein the plurality of generated tokens is a first plurality of tokens associated with a first bank identification number (BIN) range, and wherein the operations further comprise:

detecting a low token indication based at least on the token depletion rate;

provisioning a second BIN range for a second plurality of tokens based at least on the low token indication; and

generating the second plurality of tokens in the token database based at least on the second BIN range.

3. The system of claim 1, wherein the plurality of tokens is a first plurality of tokens associated with a first bank identification number (BIN) range, wherein the transaction processing policy is a first transaction processing policy, and wherein the operations further comprise:

activating a second BIN range for a second plurality of tokens in response to detecting a low token indication based on the token depletion rate; and

generating a second transaction processing policy based at least on the second BIN range.

4. The system of claim 1, wherein detecting the one or more transaction processing anomalies comprises:

identifying one or more declined transactions in the token traffic; and

determining a cause for the one or more declined transactions.

5. The system of claim 1, wherein the operations further comprise:

determining an operating capacity associated with processing transaction requests;

determining a reserve capacity associated with the operating capacity based at least on the one or more transaction processing anomalies detected in the network; and

sharing the reserve capacity with a plurality of sites of the system for processing the transaction requests.

6. The system of claim 1, wherein the modified set of transaction data consists of a subset of the set of transaction data.

7. The system of claim 1, wherein the operations further comprise:

determining a first matrix based on the set of transaction data; and

performing one or more matrix operations with the first matrix and the intermediate matrix to generate a second matrix representing the modified set of transaction processing data, wherein the transaction request is caused to be processed based in part on the second matrix.

8. A non-transitory machine-readable medium having stored thereon machine-readable instructions which, when executed by one or more processors, cause a machine to perform operations comprising:

receiving, from a user device over a network, a transaction request comprising a set of transaction data;

determining token traffic in the network based at least on analyzing one or more token requests transmitted over the network during a period of time prior to receiving the transaction request;

accessing a token database comprising a plurality of tokens generated based on a token depletion rate, wherein the token depletion rate is based at least on the determined token traffic in the network and a number of tokens issued in response to token requests over the period of time;

retrieving a token from the token database for the transaction request;

accessing a transaction processing policy that was generated based on one or more transaction processing anomalies in the network, wherein the one or more transaction processing anomalies are detected based on analyzing the token traffic, and wherein the transaction processing policy comprises an intermediate matrix that represents policy requirements in a binary form;

modifying the set of transaction data in the transaction request based on the generated transaction processing policy by applying the intermediate matrix to the set of transaction data; and

causing the transaction request to be processed based on the modified set of transaction data and the retrieved token.

9. The non-transitory machine readable medium of claim 8, wherein the plurality of generated tokens is a first plurality of tokens associated with a first bank identification number (BIN) range, and wherein the operations further comprise:

detecting a low token indication based at least on the token depletion rate;

provisioning a second BIN range for a second plurality of tokens based at least on the low token indication; and

generating the second plurality of tokens in the token database based at least on the second BIN range.

10. The non-transitory machine readable medium of claim 8, wherein the plurality of tokens is a first plurality of tokens associated with a first bank identification number (BIN) range, wherein the transaction processing policy is a first transaction processing policy, and wherein the operations further comprise:

activating a second BIN range for a second plurality of tokens in response to detecting a low token indication based on the token depletion rate; and

generating a second transaction policy based at least on the second _bin BIN range.

11. The non-transitory machine readable medium of claim 8, wherein the operations further comprise:

determining a first matrix based on the set of transaction data; and

performing one or more matrix operations with the first matrix and the intermediate matrix to generate a second matrix representing the modified set of transaction processing data, wherein the transaction request is caused to be processed based in part on the second matrix.

12. The non-transitory machine readable medium of claim 8, wherein the operations further comprise:

determining that the one or more transaction processing anomalies detected in the network are cause by a shortage of operating capacity.

13. The non-transitory machine readable medium of claim 8, wherein the operations further comprise:

In response to determining that the one or more transaction processing anomalies is caused by a shortage of operating capacity, determining an operating capacity associated with processing transactions;

determining a reserve capacity associated with the operating capacity based at least on the one or more processing anomalies detected in the network; and

sharing the reserve capacity with a plurality of sites for processing the transactions.

14. The non-transitory machine-readable medium of claim 8, wherein the operations further comprises:

identifying one or more declined transactions in the token traffic; and

determining a cause for the one or more declined transactions, wherein the transaction processing policy is generated based in part on the cause.

15. A computer-implemented method, comprising:

determining, by a transaction processing system comprising one or more hardware processors, token traffic in a network based at least on analyzing one or more token requests transmitted over the network during a period of time;

determining, by the transaction processing system, a token depletion rate based at least on the token traffic and on a number of tokens issued in response to token requests during the period of time;

generating, by the transaction processing system, a plurality of tokens in a token database based at least on the token depletion rate;

detecting, by the transaction processing system, one or more transaction processing anomalies in the network based on analyzing the token traffic;

generating, by the transaction processing system, a transaction processing policy based on the one or more detected transaction processing anomalies, wherein the transaction processing policy comprises an intermediate matrix that represents policy requirements in a binary form;

receiving, by the transaction processing system from a user device, a transaction request comprising a set of transaction data;

retrieving, by the transaction processing system, a token from the token database for the transaction request;

modifying, by the transaction processing system, the set of transaction data in the transaction request based on the generated transaction processing policy by applying the intermediate matrix to the set of transaction data; and

processing, by the transaction processing system, the transaction request based on the modified set of transaction data and the retrieved token.

16. The method of claim 15, wherein the plurality of generated tokens is a first plurality of tokens associated with a first bin bank identification number (BIN) range, and wherein the method further comprises:

detecting a low token indication based at least on the token depletion rate;

provisioning a second BIN range for a second plurality of tokens based at least on the low token indication; and

generating the second plurality of tokens in the token database based at least on the second BIN range.

17. The method of claim 15, wherein the plurality of tokens is a first plurality of tokens associated with a first bank identification number (BIN) range, wherein the transaction processing policy is a first transaction processing policy, and wherein the method further comprises:

activating a second BIN range for a second plurality of tokens in response to detecting a low token indication based on the token depletion rate; and

generating a second transaction processing policy based at least on the second BIN range.

18. The method of claim 15, further comprising:

identifying one or more declined transactions in the token traffic; and

determining a cause for the one or more declined transactions, wherein the transaction processing policy is generated based in part on the cause.

19. The method of claim 15, wherein modifying the set of transaction data comprises removing a portion of the set of transaction data.

20. The method of claim 15, further comprising:

determining a first matrix based on the set of transaction data; and

performing one or more matrix operations with the first matrix and the intermediate matrix to generate a second matrix representing the modified set of transaction processing data, wherein the transaction request is caused to be processed based in part on the second matrix.