US 12,596,926 B2Granted

Systems and methods for adjusting data processing components for non-operational targets

AssigneeCapital One Services, LLC

InventorsAamer Charania, Abhisek Jana, Jiankun Liu, Behrouz Saghafi Khadem

Life of this patent

Estimating the term…

Loading patent…

Abstract

Systems and methods for adjusting data processing components. In some aspects, the systems and methods include training a first machine learning model using a similarity graph generated based on training entries to predict whether a target system related to a node in the similarity graph will be non-operational within a future period of time, processing using the trained first machine learning model an updated similarity graph generated based on training and inference entries to predict for each node for the inference entries whether a target system related to the node will be non-operational within the future period of time, processing using a second machine learning model predictions and associated inference entries to predict that a target system related to a node for an entry will be non-operational within the future period of time, and adjusting data processing components related to the target system.

Claims20

Dependent claims

A system for adjusting data processing components related to a target system predicted to be non-operational within a future period of time, the system comprising:

one or more processors;
and
a non-transitory, computer-readable medium comprising instructions that, when executed by the one or more processors, cause operations comprising:
receiving an imbalanced dataset comprising a plurality of data entries in tabular form, each data entry related to a target system and including one or more features for the target system for a past period of time, the plurality of data entries comprising training entries and inference entries;
determining a similarity metric for each pair of the plurality of data entries;
generating a similarity graph based on the training entries, the similarity graph connecting a node for a training entry to another node for another training entry with a similarity metric that satisfies a threshold;
training a graph neural network for predicting whether a target system related to a node in the similarity graph will be non-operational within a future period of time;
generating an updated similarity graph based on the plurality of data entries including the training entries and the inference entries, the updated similarity graph connecting a node for a data entry to another node for another data entry with a similarity metric that satisfies the threshold;
processing, using the trained graph neural network, each node for the inference entries in the updated similarity graph to generate predictions indicating whether the target system related to the node will be non-operational within the future period of time;
providing the predictions for nodes in the updated similarity graph and associated data entries to a machine learning model to generate an output indicating whether the target system related to the node for a data entry will be non-operational within the future period of time, wherein the machine learning model comprises a gradient boosting machine or a logistic regression classifier;
and
in response to the machine learning model generating the output indicating that the target system related to the node for a data entry will be non-operational within the future period of time, adjusting computing resources related to the target system.

9 more elements

A method comprising:

receiving an imbalanced dataset comprising of a plurality of data entries, each data entry related to a target system and including one or more features for the target system for a past period of time, the plurality of data entries comprising first entries and second entries;
generating a similarity graph based on the first entries;
training a first machine learning model for predicting whether a target system related to a node in the similarity graph will be non-operational within a future period of time;
generating an updated similarity graph based on the plurality of data entries including the first entries and the second entries;
processing, using the trained first machine learning model, each node for the second entries in the updated similarity graph to generate predictions indicating whether the target system related to the node will be non-operational within the future period of time;
providing the predictions for nodes in the updated similarity graph and associated data entries to a second machine learning model to generate an output indicating whether the target system related to the node for a data entry will be non-operational within the future period of time;
and
based on the second machine learning model generating the output indicating that the target system related to the node for a data entry will be non-operational within the future period of time, adjusting data processing components related to the target system.

5 more elements

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

receiving an imbalanced dataset comprising of a plurality of data entries, each data entry related to a target system and including one or more features for the target system for a past period of time, the plurality of data entries comprising first entries and second entries;
generating a similarity graph based on the first entries;
training a first machine learning model for predicting whether a target system related to a node in the similarity graph will be non-operational within a future period of time;
generating an updated similarity graph based on the plurality of data entries including the first entries and the second entries;
processing, using the trained first machine learning model, each node for the second entries in the updated similarity graph to generate predictions indicating whether the target system related to the node will be non-operational within the future period of time;
providing the predictions for nodes in the updated similarity graph and associated data entries to a second machine learning model to generate an output indicating whether the target system related to the node for a data entry will be non-operational within the future period of time;
and
based on the second machine learning model generating the output indicating that the target system related to the node for a data entry will be non-operational within the future period of time, adjusting data processing components related to the target system.

5 more elements

Dependencies are inferred from the claim text.