Hi guys, I was trying to find out the reasons for regression failures using the machine learning feature of Kibana, but I was only able to detect the anomalies, that too based on the count only.
I want it to analyze certain string fields and their values to detect the reason behind failures. Is it possible to customize the model this way in Kibana or not?
If it's possible, please tell me how to do that or attach any links available.
We have a log message categorizaton feature which will allow you to analyse counts of categories or rare categories from log messages. This works best on machine written log messages - meaning log messages typically written by a developer for the purpose of system troubleshooting.
https://www.elastic.co/guide/en/machine-learning/current/ml-configuring-categories.html
Hope this helps with your use case.
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.