Model Evaluation Checkpoint

1. Accuracy is defined as:

Correct predictions / Total predictions True positives / Total positives Precision * Recall 1 - Error rate

2. Precision answers the question:

Of predicted positives, how many are correct? Of actual positives, how many were found? What percentage is correct? How many mistakes were made?

3. Recall answers the question:

Of actual positives, how many were found? Of predicted positives, how many are correct? What is the error rate? How balanced is the data?

4. F1 score is the:

Harmonic mean of precision and recall Arithmetic mean of precision and recall Maximum of precision and recall Product of precision and recall

5. A confusion matrix has how many elements for binary classification?