AI engineering glossary · Evaluation metrics
What is g-mean?
G-mean (also called geometric mean score): The geometric mean of recall and specificity: sqrt(TPR * TNR). Like balanced accuracy it treats both classes equally, but it drops sharply if either rate is very low, so it punishes a model that ignores the minority class. It is common in imbalanced-learning research.
G-mean: an example
With recall 0.75 and specificity 0.99, G-mean = sqrt(0.75 * 0.99), about 0.86.
Related terms
More Evaluation metrics terms
- Gini coefficient
- Hamming loss
- Hit rate
- Human evaluation
- Isotonic calibration
- Jaccard index
- Kendall's tau
- Kolmogorov-Smirnov statistic
Where you learn it
G-mean is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.