What should be the implementation approach for logstash and tensorflow (and/or any other machine learning framework) for unsupervised learning on syslog data?

(Hara Acharya) #1

I have a syslog monitoring system using elk.
It works fine and matches the patterns correctly.

What I would like to achieve now is:

  • Implement unsupervised machine learning on the syslog data.

Is the below approach workable or is there any other better way I should think of ?

  • input syslog text file to logstash -> output as csv file without applying any filter-> use csv out file to be processed outside elk using any machine learning framework
    And then input the same file back in logstash to be feeded to elasticsearch for indexing.
    Is this possible and feasible ?

Please suggest.

(system) #2

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