Questions about Splunk Machine Learning Toolkit.

[mpaw]

I'm new to Splunk Machine Learning Toolkit and still deciding what model and algorithm should be used in predicting system outages. It would be a big help if someone can assist me with my inquiries below. Thanks in advance!

1. What are the pros and cons of each model and algorithm?
2. What are the most popular models and algorithms being used in predicting system outages?
3. Is it possible to use multiple data input criteria in machine learning? Example: Row1: Appdynamics business transaction error + Row2: SCOM memory utilization 95% spike Combination of Row1+2 = system outage

[gjanders · Accepted solution]

You could refer to the MLTK welcome page or the detailed MLTK algorithms page for algorithm information but it sounds like you are new to machine learning.

Perhaps look into workshops if they are available https://events.splunk.com/IT_Workshop_May2020 or documentation/tutorials online, this is a complex space but the assistants in the MLTK would be a place to start