I have an imbalaced multi-label classification dataset
I tried these 2 models
First I used Bernoulli Naive Bayes algorithm that nativelly supports multi-label classification I got Micro F1-score of 45% and Macro F1 = 12%
Another model I tried was tweaked Random Forest, which gave me Micro F1= 61% and Macro F1=7%
I did the experiment using other similar datasets
I always got Micro F1 increases in my tweaked Random Forest. while Macro F1 always decreases!
what could be the reason that can cause Micro F1 increase and Macro F1 decrease?