Machine-learned entity function management
US11410085B1 · kind B1 · utility
Assignee
Inventors
Key dates
| Filing date | Apr 7, 2021 |
| Grant date | Aug 9, 2022 |
| Priority date | — |
| Expiry date | Apr 7, 2041 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06F18/2433
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
A system and a method are disclosed herein for machine-learned detection of outliers within entity functions. An entity management system uses machine learning to cluster data characterizing functions performed by entities, and determines one or more data clusters that are outliers. The system receives entity function data indicating a metric and a type, and provides the received data into a supervised machine learning model. The model is trained to apply a label to the entity function data, where the label indicates a classification of the data into a cluster (e.g., an outlier cluster). This outlier detection may inform the system's generation of a function monitor to guide rectifying action that addresses the detected outliers. The system may receive user input affirming or rejection the classification of the data into a particular cluster. The system may leverage the user input to retrain the supervised machine learning model.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.