1

I read about Bayes Naive classifier will be affected by imbalance data. If so, why still such high roc_auc_score of 0.96?

target Proportion
Diseased 0.85
Not diseased 0.15
Metric Result
Precision 0.71
Recall 0.94
F1-score 0.81
Accuracy 0.91
Wong
  • 33
  • 3
  • 1
    Statisticians do not see class imbalance as such a problem. It might be helpful if you say why you find the imbalance problematic. https://stats.stackexchange.com/questions/357466 https://www.fharrell.com/post/class-damage/ https://www.fharrell.com/post/classification/ https://stats.stackexchange.com/a/359936/247274 https://stats.stackexchange.com/questions/464636/ In particular, ROCAUC is a function of the probability outputs, so while it might not be as good as the strictly proper scoring rules discussed in the links, it does make use of the probability outputs. – Dave Mar 19 '22 at 20:49
  • What exactly and where did you read? – Tim Mar 19 '22 at 21:09

0 Answers0