Forum Discussion
Anomaly detection - raw score meaning / calculation
Hi - when I run the anmoaly detection analytics tool, I get a raw score for each anomaly. How is this raw score calculated?
Hi Anonymous ,
At present, although it is very complicated, the algorithm explanation given in the official document is the most standard. In visual, over time, adding anomaly detection will automatically make anomalies and expected value ranges appear in the chart.
If I find a concise and clear explanation later, I will notify you in time so that you have a further understanding of row score.
Best Regards,
Henry
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.
3 Replies
v-henryk-mstfCommunity Support
Hi Anonymous ,
It seems a bit difficult to understand the calculation logic of row score in specific anmoaly detection. You can refer to the following technical blogs, hoping to help you.
related document links:
Overview of SR-CNN algorithm in Azure Anomaly Detector - Microsoft Tech Community
Anomaly detection tutorial - Power BI | Microsoft Docs
Best Regards,
Henry
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.- v-henryk-mstf
Hi Anonymous ,
At present, although it is very complicated, the algorithm explanation given in the official document is the most standard. In visual, over time, adding anomaly detection will automatically make anomalies and expected value ranges appear in the chart.
If I find a concise and clear explanation later, I will notify you in time so that you have a further understanding of row score.
Best Regards,
Henry
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.