Popular Courses
Beginner – Linear Regression
Introduction
Beginner – Decision Trees
Introduction
Beginner – Neural Networks
Introduction

Sign in to Curvingo

or

New Account     Reset Password     

Plans

Details
Visualizing Performance
Introduction
Building ROC Curves
What Makes A Good ROC Curve?
AUC: Area Under the Curve
Considerations
Code with Python
Summary
Curvingo · Made in India

Share

Send us Feedback

Please let us know if you have any feedback and suggestions, or if you find any errors and bugs in our content.

Sorry, your message couldn’t be submitted. Please try again!

Thanks for your feedback!

Reset Progress

This will delete your progress and chat data for all chapters in this course, and cannot be undone!

Glossary

Select one of the keywords on the left…

Visualizing PerformanceWhat Makes A Good ROC Curve?

Reading time: ~5 min

The shape of an ROC curve shows how well the scores separate the two groups. Here is a smaller test with 8 patients and 8 healthy people. Drag the patients' scores left or right and complete the missions.

When the two groups get the same scores, every threshold catches patients and healthy people at the same rate, and the curve is the diagonal — no better than tossing a coin. When every patient outscores every healthy person, one threshold catches them all with no false alarm, and the curve runs up the left edge to the top-left corner. Real tests sit in between: the more the curve bulges towards the top-left, the better the test.

Name the Curves

Three tests, three curves.

Curve A belongs to , curve B to and curve C to .

Beware of perfection

Real tests are almost never perfect. If a model's curve hugs the top-left corner on the very first try, suspect a data leak — some information about the answers sneaked into the training data — before you celebrate.

Sina