1

Hi I'm a CS graduate student

I have a question for AI or data experts. I'm writing a paper

My dataset is time-series sensor data and anomaly (positive class) ratio is between 5% and 6%

you can see the picture below. I used classification_report in sklearn library

enter image description here

But, I'm confused as to what value to report in my evaluation section ...

I think it is reasonable to report f1-score with macro avg (0.40)

Is it ok?

Thank you for your explanation !

Dae-Young Park
  • 31
  • 1
  • Use the metric you see being used in similar work. // Statisticians prefer proper scoring rules that are not based on thresholds. Log loss and Brier score are two common examples. [You might be interested in this link.](https://stats.stackexchange.com/questions/357466/are-unbalanced-datasets-problematic-and-how-does-oversampling-purport-to-he) – Dave Jul 02 '21 at 01:57

0 Answers0