Anomaly detection

We are trying to use anomaly detection from Machine Learning. But very little information is available on the internet regarding it. Can you please help us with the following?

Details of anomaly detection algorithm, which algorithm is getting used, how does it work etc?

  1. Also how is the severity calculated?
  2. What are the typical values in the anomalies displayed?
  3. Which algorithm gets used for detecting anomalies in univariate time series?

Any document throwing light on the above documents will help.

There's a fair amount of written information available:

Book on Elastic ML:

The definitive guide

Recent ML Webinars

Public ML Blogs:

See What Users Are Asking Questions About:

Background on “How it works” or the “Math behind ML":

30 Min Meetup w/ Steve Dodson:

2017 ElasticON:

2018 ElasticON:

1 Like

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.