8

What adjustments do I need to make when I have extreme values in the confusion matrix as stated above i.e sensitivity = 1 and specificity = 0?

Moses
  • 247
  • 7
  • 9
    It implies that the model is always predicting positives. What do you mean by adjustments? – Laksan Nathan Jun 02 '21 at 12:01
  • I thought at first that the model may not correctly classifying the predictions and may need need to do a few tweaks – Moses Jun 02 '21 at 12:24

1 Answers1

13

Sensitivity $= 1$ means you had some true positives and no false negatives: all actual cases were correctly predicted as positive

Specificity $= 0$ means you had some false positives and no true negatives: all actual non-cases were incorrectly predicted as positive

So having both of these means that everything was predicted to be positive, whether it was an actual case or not

You might want to adjust your predictions so some are predicted positive and some negative. How you do this depends on how you are predicting

Henry
  • 30,848
  • 1
  • 63
  • 107