Confusion matrix in ML Python - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is a confusion matrix in machine learning?

A confusion matrix is a table that shows the performance of a classification model by comparing actual labels with predicted labels. It helps us see how many predictions were correct or wrong for each class.

beginner

What do the terms True Positive (TP), False Positive (FP), True Negative (TN), and False Negative (FN) mean in a confusion matrix?

TP: Model correctly predicted positive.
FP: Model predicted positive but it was negative.
TN: Model correctly predicted negative.
FN: Model predicted negative but it was positive.

intermediate

How can a confusion matrix help in understanding model errors?

It shows exactly where the model makes mistakes by counting wrong predictions for each class. This helps us know if the model confuses certain classes more than others.

intermediate

What is the difference between accuracy and precision using confusion matrix terms?

Accuracy = (TP + TN) / Total predictions, shows overall correctness.
Precision = TP / (TP + FP), shows how many predicted positives are actually positive.

advanced

Why is a confusion matrix especially useful for imbalanced datasets?

Because accuracy can be misleading when classes are imbalanced, the confusion matrix shows detailed counts of each type of error, helping us understand model performance on minority classes.

What does the cell in the confusion matrix representing False Negative (FN) indicate?

A Model predicted negative but actual was positive

B Model predicted negative and actual was negative

C Model predicted positive and actual was positive

D Model predicted positive but actual was negative

Which metric can be directly calculated from a confusion matrix?

A Learning rate

B Accuracy

C Number of features

D Epoch count

If a confusion matrix has high False Positives, what does it mean?

A Model predicts all cases correctly

B Model often misses positive cases

C Model often wrongly predicts positive when it is negative

D Model has no errors

Why might accuracy be misleading on imbalanced datasets?

A Because it ignores the number of features

B Because it only counts True Positives

C Because it depends on the learning rate

D Because it can be high even if the model ignores minority class

Which of these is NOT part of a confusion matrix?

A Learning Rate

B True Positive

C False Negative

D True Negative

Explain what a confusion matrix is and how it helps evaluate a classification model.

Describe the four parts of a confusion matrix and what each part means.