Back to results
US 2025/0165458 A1Published application
Resource validation systems and methods
AssigneeCapital One Services, LLC
InventorsCarel de Bruyn, James Anthonyraj, David Duong
Life of this patent
Estimating the term…
Abstract
Techniques for improved data and/or resource validation provide a real-time validation of data and resources. A resource may include an aggregation of related data into a single unique unit. The validation may be performed based on one or more events or triggers, such as on the occurrence of a data change, deployment of an application, when the resource(s) are utilized, etc. The validation may be on demand (e.g., based on one or more events and/or triggers) and/or manually (e.g., via a call via an application program interface by a user).
Claims20
A computer-implemented method comprising:
- determining training data comprising:
- a history of validation requests, wherein each historical validation request identifies one or more resources to be aggregated and validated;
- and
- an indication of validation results corresponding to each historical validation request:
- generating a trained machine learning model by training, using the training data, a machine learning model to perform validation of input aggregated resources;
- tracking metrics by monitoring activity of serverless applications;
- generating, based on determining that the metrics satisfy a threshold corresponding to a change in status of one or more of the serverless applications, a resource request;
- aggregating, based on the resource request, data of one or more resources to generate an aggregated resource;
- distributing, based on a validator identifier, the aggregated resource to one or more registered validators associated with the validator identifier;
- validating the aggregated resource by:
- providing, to the trained machine learning model, at least a portion of the aggregated resource;
- and
- determining, based on output of the trained machine learning model, validation results;
- and
- writing the validation results to a results database.
12 more elements
A computing device comprising:
- one or more processors;
- and
- memory storing instructions that, when executed by the one or more processors, configure the computing device to:
- determine training data comprising:
- a history of validation requests, wherein each historical validation request identifies one or more resources to be aggregated and validated;
- and
- an indication of validation results corresponding to each historical validation request;
- generating a trained machine learning model by training, using the training data, a machine learning model to perform validation of input aggregated resources;
- tracking metrics by monitoring activity of serverless applications;
- generating, based on determining that the metrics satisfy a threshold corresponding to a change in status of one or more of the serverless applications, a resource request;
- aggregate, based on the resource request, data of one or more resources to generate an aggregated resource;
- distribute, based on a validator identifier, the aggregated resource to one or more registered validators associated with the validator identifier;
- validate the aggregated resource by:
- providing, to the trained machine learning model, at least a portion of the aggregated resource;
- and
- determining, based on output of the trained machine learning model, validation results;
- and
- generate a notification based on the validation results.
15 more elements
One or more non-transitory media storing instructions that, when executed, cause a computing device to:
- determine training data comprising:
- a history of validation requests. wherein each historical validation request identifies one or more resources to be aggregated and validated;
- and
- an indication of validation results corresponding to each historical validation request;
- generate a trained machine learning model by training, using the training data, a machine learning model to perform validation of input aggregated resources;
- track metrics by monitoring activity of serverless applications;
- generate, based on determining that the metrics satisfy a threshold corresponding to a change in status of one or more of the serverless applications, a resource request;
- aggregate, based on the resource request, data of one or more resources to generate an aggregated resource;
- distribute, based on a characteristic of the aggregated resource, the aggregated resource to one or more registered validators;
- validate, by the one or more registered validators, the aggregated resource by:
- providing, to the trained machine learning model, at least a portion of the aggregated resource;
- and
- determining, based on output of the trained machine learning model, validation results;
- and
- generate a notification based on the validation results.
12 more elements
Dependencies are inferred from the claim text.