Expand my Community achievements bar.

SOLVED

Out-of-the-box Machine Learning capabilities

Avatar

Level 1

Hi everybody,

 

I was wondering what kind of Machine Learning capabilities come pre-configured with Adobe Analytics. Are there some ML options that require additional configuration? I'm trying to get a grasp as to whether this is more of a reporting tool, or a true analytics tool with built-in ML.

 

Thank you,

W

1 Accepted Solution

Avatar

Correct answer by
Community Advisor

The most obvious and common usage of ML in AA is anomaly detection. https://experienceleague.adobe.com/docs/analytics/analyze/analysis-workspace/virtual-analyst/anomaly...

If you use Attribution modelling a lot, then there's algorithmic attribution, which also uses ML: https://experienceleague.adobe.com/docs/analytics/analyze/analysis-workspace/attribution/algorithmic...

And then there's the Segment Comparison panel: https://experienceleague.adobe.com/docs/analytics/analyze/analysis-workspace/panels/segment-comparis...

So Adobe's approach to ML is that it doesn't expose the inner workings of machine learning nor let you configure what the machine should learn. Rather, they have built out fully functional tools that take advantage of ML internally.

If you want to design your own ML, you'll need to work it out on your own, e.g. through a cloud service like Azure, AWS, GCP, etc.

View solution in original post

2 Replies

Avatar

Correct answer by
Community Advisor

The most obvious and common usage of ML in AA is anomaly detection. https://experienceleague.adobe.com/docs/analytics/analyze/analysis-workspace/virtual-analyst/anomaly...

If you use Attribution modelling a lot, then there's algorithmic attribution, which also uses ML: https://experienceleague.adobe.com/docs/analytics/analyze/analysis-workspace/attribution/algorithmic...

And then there's the Segment Comparison panel: https://experienceleague.adobe.com/docs/analytics/analyze/analysis-workspace/panels/segment-comparis...

So Adobe's approach to ML is that it doesn't expose the inner workings of machine learning nor let you configure what the machine should learn. Rather, they have built out fully functional tools that take advantage of ML internally.

If you want to design your own ML, you'll need to work it out on your own, e.g. through a cloud service like Azure, AWS, GCP, etc.

Avatar

Level 10
Do any of the answers below answer your initial question? If so, can you select one of them as the correct answer? If none of the answers already provided answer your question, can you provide additional information to better help the community solve your question?