Abstract
Networks have increased in size and complexity such that the number of events occurring each day has grown drastically. Techniques of this proposal provide for the ability to infer candidates for causal relationships—in some cases, with confidence. In particular, a novel machine learning (ML) based system is described that provides for the ability to narrow-down candidate temporal patterns that may potentially explain an event of interest (e.g., a network outage). The system is trainable with a human in the loop and is highly effective even with minimal amount of prior training.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Goloubew, Dmitry; Benoussaid, Nassim; Pignataro, Carlos M.; and Yashin, Vladimir, "SYSTEM FOR EXPERT-ASSISTED CAUSAL INFERENCE FOR RANKING EVENTS OF INTEREST IN NETWORKS", Technical Disclosure Commons, (January 24, 2021)
https://www.tdcommons.org/dpubs_series/3989