I want to understand how Quicksight uses ML based anomaly feature behind the scenes and more specifically once it finds anomalies, how does Quicksight do contribution analysis to find the key contributors?
Hello @Mehroz, welcome to the QuickSight Community!
Honestly, a lot of these things are going to depend on what data you are analyzing in QuickSight, whether you are collecting data that displays a meaningful change over time, and how long you have been collecting it. I would highly recommend looking into the documentation AWS has for this feature, and just testing out the anomaly detection in an insight visual in an analysis to see what it is able to tell you.
Let me know if that helps! This is one of those things that will make a lot more sense if you test it out to see if it provides helpful information about your data.
Thanks @DylanM , I am trying to replicate the Quicksight ML-based anomaly detection in Sagemaker. The RCF implementation of sagemaker does not provide the explanation like it does in Quicksight. I tried to use Amazon Q for explanation of Key Driver in Quicksight and it told me that Quicksight use regression to find the top contributors but I did not find it in documentation.
Hello @Mehroz, I do feel like I read on a document somewhere the QuickSight anomaly detection uses regression, but I can’t remember where I saw that. I will see if I can find some other information to send over to you, but asking Amazon Q was a great idea!
Hello @Mehroz, I’ll link some documentation here. Let me know if there is anything else I can do to assist!