Machine learning techniques for providing enriched root causes based on machine-generated data
US10600002B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Oct 17, 2017 |
| Grant date | Mar 24, 2020 |
| Priority date | — |
| Expiry date | Apr 16, 2038 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06Q10/20
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
A method and system for providing an enriched root cause of an incident using machine-generated textual data. The method includes extracting, from a dataset including machine-generated textual data for a monitored environment, a plurality of features related to a root cause of an incident in the monitored environment; generating a suitability score for each of a plurality of insights with respect to the incident based on the extracted features and a suitability model, wherein the suitability model is created based on a training set including a plurality of training inputs and a plurality of training outputs, wherein each training output corresponds to at least one of the plurality of training inputs; and selecting at least one suitable insight based on the generated suitability scores.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.