Patent · US Active

Diagnosing and remediating errors using visual error signatures

US11314609B2 · kind B2 · utility

3Cited by
2References
20Claims
0Family size

Assignee

Inventor

Key dates

Filing dateOct 24, 2019
Grant dateApr 26, 2022
Priority date
Expiry dateSep 25, 2040

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06F2201/88
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

A method includes detecting an error that has occurred in one or more assets of an enterprise system and generating a visual error signature of the detected error, the visual error signature comprising at least a portion of a graph-based visualization of operation of the assets. The method also includes providing the generated visual error signature for the detected error as input to a machine learning model and utilizing the machine learning model to classify the visual error signature for the detected error as belonging to at least a given one of a plurality of error classes, the machine learning model being trained using historical visual error signatures for previously-detected errors. The method further includes identifying at least one action taken to remediate each of one or more previously-detected errors of the given error class and remediating the detected error utilizing one or more of the identified actions.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.