Precision, recall interpretation

Precision can be seen as a measure of quality, and recall as a measure of quantity. Higher precision means that an algorithm returns more relevant results than irrelevant ones, and high recall means that an algorithm returns most of the relevant results (whether or not irrelevant ones are also returned).

1. What is a good precision and recall score?
2. What does a precision-recall curve tell you?
3. What is an acceptable F1 score?
4. How do you interpret an F score?

What is a good precision and recall score?

High precision relates to the low false positive rate. We have got 0.788 precision which is pretty good. Recall (Sensitivity) - Recall is the ratio of correctly predicted positive observations to the all observations in actual class - yes.

What does a precision-recall curve tell you?

Precision-Recall curves summarize the trade-off between the true positive rate and the positive predictive value for a predictive model using different probability thresholds.

What is an acceptable F1 score?

An F1 score is considered perfect when it's 1 , while the model is a total failure when it's 0 . Remember: All models are wrong, but some are useful. That is, all models will generate some false negatives, some false positives, and possibly both.

How do you interpret an F score?

If you get a large f value (one that is bigger than the F critical value found in a table), it means something is significant, while a small p value means all your results are significant. The F statistic just compares the joint effect of all the variables together.