top of page
flag-16-512.png

Day 1

Consider that your dataset is biased. You want to compare the model performance. However, you are not concerned with the performance with the majority class. Your task is the correct detection of positive class. Which predictive curve will you use?

Day 1

Avenir Light is a clean and stylish font favored by designers. It's easy on the eyes and a great go to font for titles, paragraphs & more.

Precision Recall Curves are used to counter the effect of class imbalance.

There is a class imbalance in your dataset. However, a correct detection of a sample from a class which is not of your concern or is the majority class would be considered as high True Negative (TN) score. In this case, you'd be considering to get a better True Positive, False Positive & False Negative Score. Ignoring the TN will be possible only via Precision Recall Curves as neither Precision nor Recall consider TN in their calculations.

Credits :
bottom of page