📏 Micro AUC

Micro AUC#

AUC pooled across all classes by aggregating decisions before averaging.

Important

✨ AI-generated content. This page was written with the assistance of an AI language model and is provided as a learning aid. Despite careful review, it may still contain mistakes, omissions, or out-of-date information. Whether you are new to the topic, a team lead, or a senior practitioner, treat it as a starting point rather than an authoritative reference: read it critically and independently verify anything you act on (code, commands, figures, and factual claims) against official documentation and primary sources before relying on it.

What it is#

Micro AUC is the multiclass / multilabel AUROC that pools every class’s predictions into one global ROC curve — aggregating all true-positive and false-positive counts across classes, then computing a single AUC. Because it counts every prediction equally, frequent classes contribute more.

Micro vs macro#

Where macro AUC averages per-class AUCs with equal weight, micro AUC is effectively weighted by prevalence — a rare class with few samples barely moves it. Micro answers how well does the model do on the average prediction?, macro on the average class?

When to use it#

Micro AUC suits imbalanced multiclass problems when you care about overall performance dominated by common classes, and it matches how a multilabel system is scored (over the flattened label matrix). Report it beside macro to expose class-size effects.

Theme: Classification & Averaging Metrics · All terminology

Hint

Mind map — connected ideas

Macro AUC · One-vs-Rest (OvR) · Multiclass Classification · ROC-AUC (Receiver Operating Characteristic – Area Under Curve, = AUROC) · Multiclass AUROC · Partial AUC (pAUC)

Hint

More in Classification & Averaging Metrics

Accuracy · AUC (Area Under the Curve) · Average Precision (AP) · Binary Classification · Classification Probability · Discriminatory Power · F1-score · Gini Coefficient · Harmonic Mean · Log Loss (also called Logarithmic Loss or Cross-Entropy Loss) · Macro AUC · Macro AUROC (Macro-Averaged AUROC) · Macro Averaging · Macro F1