G-Mean Score (GMS) — Permetrics 2.0.0 documentation

G-mean is a performance metric in the field of machine learning and specifically in binary classification problems. It is a balanced version of the geometric mean, which is calculated as the square root of the product of true positive rate (TPR) and true negative rate (TNR) also known as sensitivity and specificity, respectively. The G-mean is a commonly used metric to evaluate the performance of a classifier in imbalanced datasets where one class has a much higher number of samples than the other. It provides a balanced view of the model’s performance as it penalizes low values of TPR and TNR in a single score. The G-mean score provides a balanced evaluation of a classifier’s performance by considering both the positive and negative classes. The formula for the G-mean score is given by where TPR (True Positive Rate) is defined as and TNR (True Negative Rate) is defined as with TP (True Positives) as the number of instances that are correctly classified as positive, TN (True Negatives)

G-mean is a performance metric in the field of machine learning and specifically in binary classification problems. It is a balanced version of the geometric mean, which is calculated as the square root of the product of true positive rate (TPR) and true negative rate (TNR) also known as sensitivity and specificity, respectively. The G-mean is a commonly used metric to evaluate the performance of a classifier in imbalanced datasets where one class has a much higher number of samples than the other. It provides a balanced view of the model’s performance as it penalizes low values of TPR and TNR