Identification of distinguishable anomalies extracted from real time data streams
US9996409B2 · kind B2 · utility
7Cited by
3References
21Claims
0Family size
Assignee
Inventors
Key dates
| Filing date | Mar 28, 2016 |
| Grant date | Jun 12, 2018 |
| Priority date | — |
| Expiry date | Jul 2, 2036 |
Classification
- Technology area (CPC H)Electricity
- CPC primaryH04L43/06
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
A big data processing system includes a workload trimming function that separates out from among a set of identified anomalies, those that are clearly outliers, rather than ones residing within clusters of anomalies as mapped within an anomalies distribution space. The outlier anomalies are not subjected to a computationally-intensive anomalies aggregating process and thus, processing resources are conserved.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.