Balanced accuracy
Balanced accuracy: The average of recall computed separately for every class, useful when class supports are highly imbalanced. In binary classification it averages sensitivity and specificity. It ignores precision and should therefore be read alongside false positives. This perspective connects the definition with practical behaviour in artificial intelligence systems.
In italiano: Balanced AccuracyLegge V · Overfitting, bias e collasso dei modelliIn binary classification it averages sensitivity and specificity. It ignores precision and should therefore be read alongside false positives.