Perfect separation
pairs = 0.9, 1 0.8, 1 0.7, 1 0.6, 0 0.5, 0
Result: AUC 1.00
ROC-AUC Calculator
ROC-AUC summarizes how well a scoring model separates positives from negatives across every threshold: the probability that a random positive example is scored above a random negative one. Paste one "score,label" pair per line (label 0 or 1) and get the AUC, plus the positive/negative counts and Gini coefficient. Tied scores are handled correctly.
Enter your values and press Calculate — the result appears here.
Worked examples
One tied score
pairs = 0.8, 1 0.7, 1 0.7, 0 0.5, 0
Result: AUC 0.875 (tie at 0.7 counts half)
Random-ish ranking
pairs = 0.9, 1 0.6, 0 0.5, 1 0.4, 0
Result: AUC 0.75
Frequently asked questions
What does AUC mean intuitively?
AUC is the probability that a randomly chosen positive example is scored higher than a randomly chosen negative one. 1.0 means perfect separation, 0.5 means the model ranks no better than a coin flip, and 0 means it is perfectly inverted.
How are tied scores handled?
When a positive and a negative have the same score, the pair contributes 0.5 instead of 1.0. This matches the average-rank method used by statistical packages and keeps AUC within [0, 1].
What is Gini?
Gini = 2 × AUC − 1, a common rescaling in credit risk and marketing: 0 for a random model, 1 for a perfect one. It carries exactly the same information as AUC.
📚 Learn More
Cybersecurity
Running LLMs on CPU: What Actually Matters?
CPU inference speed depends primarily on memory bandwidth and model size—not CPU cores. A well-quan…
CybersecuritySingle-Agent vs Multi-Agent: The Core Difference
Multi-agent systems split complex tasks across specialized agents — planner, researcher, coder, tes…
AI & Machine LearningWhat Is Numerical Precision?
Quantization reduces model size by using lower-precision numbers (4-bit instead of 32-bit). A 7B mo…
AI & Machine LearningWhy Traditional Logs Are Not Enough
AI agents execute tool loops, make multiple model calls, and change state across iterations. Tradit…
CybersecurityThe Core Comparison
Key Takeaway Prompt engineering controls what you ask. Context engineering controls what the model …
CybersecurityCategory 1: Foundational Patterns
Key Takeaway Prompt patterns are reusable templates for common AI tasks. Mastering 15 core patterns…
🔧 Related Tools
Precision Calculator
Compute precision — the share of positive predictions that were actually correct.
Try it now →Recall Calculator
Compute recall (sensitivity) — the share of actual positives the model managed to find.
Try it now →Specificity Calculator
Compute specificity — the share of actual negatives the model correctly identified.
Try it now →Accuracy Calculator
Calculate classification accuracy from true/false positives and negatives.
Try it now →💬 Discuss on BestWordz Community
Join the conversation about Machine Learning, roc auc, auc calculator on the BestWordz Community forum.
Visit Forum →💬 Discuss ROC-AUC Calculator
Have questions about ROC-AUC Calculator? Join the BestWordz Community.
Copied!