I have been playing with Elastic Machine Learning capabilities and I do know if I use things like grok, I might achieve something similar. But it's not as convenient as Logz.io makes it.
This being said I would really appreciate to listen to your thoughts of how something similar could be achieved.
Slightly old video - now there is a dedicated job wizard that assists in the configuration of such a job.
What is shown in the video here is to find anomalies in patterns in log files, but of course, the side-effect is that the process of categorization also produces information about number of unique categorizes per data source, etc: Get categories API | Elasticsearch Guide [7.15] | Elastic
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.