Machine defect prediction based on a signature
AssigneeOracle International Corporation
InventorsManoj Mathen, Raghothama Vijendran
Life of this patent
Estimating the term…
Abstract
Methods, system, and computer readable medium are presented for predicting defects using a machine learning component based on a generated signature. A trained machine learning component that has been trained with historic data that represents a series of events that occurred within a plurality of heterogeneous systems over a plurality of periods of change for the heterogeneous systems can be received. A base signature for a first heterogeneous system that includes a first mix of modules can be compared to a current signature for the first heterogeneous system to identify one or more irregularities. The trained machine learning component can predict one or more defects for the first heterogeneous system based on the identified irregularity.
Claims20
A method for predicting defects using a machine learning component based on a generated signature, the method comprising:
- receiving a trained machine learning component that has been trained with historic data that represents a series of events that occurred within a plurality of heterogeneous systems over a plurality of periods of change for the heterogeneous systems, wherein the heterogeneous systems comprise different mixes of modules;
- comparing a base signature for a first heterogeneous system that comprises a first mix of modules to a current signature for the first heterogeneous system to identify irregularities, wherein the base signature represents a baseline series of events that occurred within one or more the plurality of heterogeneous systems over one or more periods of change, the current signature represents a runtime series of events that occur within the first heterogeneous system, and at least one irregularity is identified when an event from the base signature is missing from the current signature or a time duration for an event of the base signature is different from a time duration for a corresponding event of the current signature;
- and
- predicting, using the trained machine learning component, one or more defects for the first heterogeneous system based on the identified one or more irregularities.
A system for predicting defects using a machine learning component based on a generated signature, the system comprising:
- a processor and memory, wherein the processor is configured to:
- receive a trained machine learning component that has been trained with historic data that represents a series of events that occurred within a plurality of heterogeneous systems over a plurality of periods of change for the heterogeneous systems, wherein the heterogeneous systems comprise different mixes of modules;
- compare a base signature for a first heterogeneous system that comprises a first mix of modules to a current signature for the first heterogeneous system to identify irregularities, wherein the base signature represents a baseline series of events that occurred within one or more of the plurality of heterogeneous systems over one or more periods of change, the current signature represents a runtime series of events that occur within the first heterogeneous system, and at least one irregularity is identified when an event from the base signature is missing from the current signature or a time duration for an event of the base signature is different from a time duration for a corresponding event of the current signature;
- and
- predict, using the trained machine learning component, one or more defects for the first heterogeneous system based on the identified one or more irregularities.
A non-transitory computer-readable medium having instructions stored thereon that, when executed by a processor, cause the processor to predict defects using a machine learning component based on a generated signature, the instructions causing the processor to:
- receive a trained machine learning component that has been trained with historic data that represents a series of events that occurred within a plurality of heterogeneous systems over a plurality of periods of change for the heterogeneous systems, wherein the heterogeneous systems comprise different mixes of modules;
- compare a base signature for a first heterogeneous system that comprises a first mix of modules to a current signature for the first heterogeneous system to identify irregularities, wherein the base signature represents a baseline series of events that occurred within one or more of the plurality of heterogeneous systems over one or more periods of change, the current signature represents a runtime series of events that occur within the first heterogeneous system, and at least one irregularity is identified when an event from the base signature is missing from the current signature or a time duration for an event of the base signature is different from a time duration for a corresponding event of the current signature;
- and
- predict, using the trained machine learning component, one or more defects for the first heterogeneous system based on the identified one or more irregularities.
Dependencies are inferred from the claim text.