What is model evaluation in machine learning?

• Building a new model

• Testing how well a model performs

• Storing data for training

• Drawing graphs for fun

Which metric is commonly used to evaluate classification models?

• Mean Squared Error

• Accuracy

• Time Taken

• File Size

What is accuracy?

• Number of correct predictions divided by total predictions

• Number of wrong predictions

• Speed of the model

• Amount of data used

What does a confusion matrix show?

• The confusion in the model

• True and false predictions for classification

• Training time

• Data errors

In a confusion matrix, what are True Positives (TP)?

• Wrong predictions marked as positive

• Correct positive predictions

• Negative predictions

• Errors in code

Which of these is NOT a model evaluation metric?

• Precision

• Recall

• Accuracy

• Compression

What is precision?

• True positives divided by all predicted positives

• False positives divided by total predictions

• All positives divided by true positives

• Speed of prediction

What is recall?

• True positives divided by all actual positives

• Number of incorrect results

• Total number of predictions

• Accuracy of negatives

A model with high recall but low precision means:

• It predicts too few positives

• It makes very accurate predictions

• It predicts many positives, but many are wrong

• It doesn’t work

What is F1 Score used for?

• Comparing models

• Ignoring results

• Combining precision and recall

• Calculating speed