US 11,902,269 B1Granted

Systems and methods for reducing traffic related to network operations

AssigneeCapital One Services, LLC

InventorsSunil Pradhan Sharma, Ravikanth Kompella, Rajendra Prasad Mokshagundam

Life of this patent

Estimating the term…

Loading patent…

Abstract

In some embodiments, reducing network traffic related to network operations may be facilitated. In some embodiments, information for an operation comprising a message to authenticate the operation may be received from a client device. A machine learning model trained on information regarding a plurality of historical operation and corresponding execution result may be obtained, where the plurality of historical operations were executed on a client device of a same type as the client device. Using the machine learning model, the information for the operation may be processed to predict an execution result for authenticating the operation. The execution result may be transmitted to the client device to prevent execution of the operation in response to the execution result indicating that authenticating the operation will be unsuccessful.

Claims20

Dependent claims

A token for reducing network traffic related to network operations, comprising:

one or more processors executing computer program instructions that, when executed, cause operations comprising:
receiving, from a client device, information for an operation comprising an authentication instruction for the operation and a value indicative of a priority for the operation;
in response to the value satisfying a predetermined threshold value, obtaining information regarding a plurality of historical operations and corresponding execution results, wherein the information regarding the plurality of historical operations comprises a corresponding authentication instruction and priority value for each historical operation, wherein the plurality of historical operations were executed on a device of a same type as the client device;
training a machine learning model using the information regarding the plurality of historical operations and the corresponding execution results to generate an updated machine learning model;
using the updated machine learning model, processing the information for the operation to predict an execution result for the authentication instruction for the operation;
and
in response to the execution result indicating that the authentication instruction will be unsuccessful, transmitting the execution result to the client device to prevent execution of the operation.

4 more elements

A method, comprising:

receiving, at a token, from a client device, information for an operation comprising a message to authenticate the operation and a value indicative of a priority of the operation;
in response to the value satisfying a predetermined threshold value, obtaining, at the token, a machine learning model trained on information regarding a plurality of historical operations and corresponding execution results, wherein the information regarding the plurality of historical operations comprises (i) a corresponding message to authenticate a historical operation and (ii) a value associated with the historical operation indicating a priority of the historical operation of the plurality of historical operations, and wherein the plurality of historical operations were executed on a device of a same type as the client device;
using the machine learning model, processing the information for the operation to predict an execution result for authenticating the operation;
and
in response to the execution result indicating that authenticating the operation will be unsuccessful, transmitting, from the token, the execution result to the client device to prevent execution of the operation.

2 more elements

One or more non-transitory computer-readable media comprising instructions that, when executed by one or more processors, cause operations comprising:

receiving, at a token, from a client device, information for an operation comprising a message to authenticate the operation and a value indicative of a priority of the operation;
in response to the value satisfying a predetermined threshold value, obtaining, at the token, a machine learning model trained on information regarding a plurality of historical operations and corresponding execution results, wherein the information regarding the plurality of historical operations comprises (i) a corresponding message to authenticate a historical operation and (ii) a value associated with the historical operation indicating a priority of the historical operation of the plurality of historical operations, and wherein the plurality of historical operations were executed on a device of a same type as the client device;
using the machine learning model, processing the information for the operation to predict an execution result for authenticating the operation;
and
in response to the execution result indicating that authenticating the operation will be unsuccessful, transmitting, from the token, the execution result to the client device to prevent execution of the operation.

2 more elements

Dependencies are inferred from the claim text.