# Xpack machine learning boundary **URL:** \ **Category:** Elasticsearch\ **Tags:** elastic-stack-machine-learning\ **Created:** [April 27, 2018, 9:52am UTC](https://discuss.elastic.co/t/xpack-machine-learning-boundary/129803 "2018-04-27T09:52:23Z")\ **Posts on this page:** 5\ **Page:** 1
**Author:** ![dpkkumar01](https://avatars.discourse-cdn.com/v4/letter/d/76d3ee/32.png) [@dpkkumar01](https://discuss.elastic.co/u/dpkkumar01)\ **Post date:** [April 27, 2018, 9:52am UTC](https://discuss.elastic.co/t/xpack-machine-learning-boundary/129803/1 "2018-04-27T09:52:23Z")
We have tested our system metrics (CPU, Memory,Disk IO........etc) with the new version of elasticsearch machine learning system(Version-6.2.3) As per your documentation we understood that the boundary conditions are calculated dynamically it will be wider in the beginning and get refined as more data comes in. we would like to know the details on the following, 1. Is it is possible to have a static threshold limit set as lower and upper boundary limit values? we want to apply this logic for some of our system metrics. 2. Is it possible to have a feedback mechanism incorporated into ES Anomaly system?. User will provide the feedback on the anomalies detected by ES is actually an anomaly or not. the ES machine should take this feedback input into consideration while predicting the future anomalies. Thanks in advance. ---
**Author:** ![warkolm](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/warkolm/32/39224_2.png) [@warkolm](https://discuss.elastic.co/u/warkolm)\ **Post date:** [April 27, 2018, 9:54am UTC](https://discuss.elastic.co/t/xpack-machine-learning-boundary/129803/2 "2018-04-27T09:54:43Z")
1 is better handled with Alerting as it can do static thresholds. 2 should be available very soon, but maybe someone with more knowledge can comment. ---
**Author:** ![Hendrik\_Muhs](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/hendrik_muhs/32/25802_2.png) [@Hendrik\_Muhs](https://discuss.elastic.co/u/Hendrik_Muhs)\ **Post date:** [April 27, 2018, 11:54am UTC](https://discuss.elastic.co/t/xpack-machine-learning-boundary/129803/3 "2018-04-27T11:54:00Z")
Regarding 2: We do not have a feedback-loop system yet. There are thoughts around this, but note that this is far from trivial, if you want to do it right, especially when it comes to combining user feedback into the ML models. Having that said: For special events that are known upfront, be it the Black Friday or a planned outage due to scheduled maintenance, we build special API's to suspend modeling for this periods: > **[Scheduled Events and the Amorous Anomaly | Elastic](https://www.elastic.co/blog/scheduled-events-and-the-amorous-anomaly-elasticsearch-machine-learning)** > > Machine Learning in X-Pack learns continuously adapting its model to the changing input, redefining what is normal and what is anomalous. This is great as behaviour changes over time but what happens ... ---
**Author:** ![warkolm](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/warkolm/32/39224_2.png) [@warkolm](https://discuss.elastic.co/u/warkolm)\ **Post date:** [April 27, 2018, 2:34pm UTC](https://discuss.elastic.co/t/xpack-machine-learning-boundary/129803/4 "2018-04-27T14:34:24Z")
> [@Hendrik\_Muhs](#): > > Having that said: For special events that are known upfront, be it the Black Friday or a planned outage due to scheduled maintenance, we build special API's to suspend modeling for this periods: That's what I was thinking ok, thanks 🙂 ---
**Author:** ![warkolm](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/warkolm/32/39224_2.png) [@warkolm](https://discuss.elastic.co/u/warkolm)\ **Post date:** [October 30, 2018, 10:21pm UTC](https://discuss.elastic.co/t/xpack-machine-learning-boundary/129803/5 "2018-10-30T22:21:33Z")