What is specificity in machine learning?

Specificity is a measure in machine learning using which we can calculate the performance of a machine learning model that solves classification problems. Specificity determines how well a machine learning model can predict true negatives. Before we understand specificity in machine learning, we need to understand a few terms. They are True Positive, True Negative, False Positive, and False Negative.

True Positive (TP): True Positives (TP) are the output labels that are predicted to be true, and they are actually true.

True Negative (TN): True Negatives (TN) are the output labels that are predicted to be false, and they are actually false.

False Positive (FP): False Positives (FP) are the output labels that are predicted to be true, but they are actually false.

False Negative (FN): False Negatives (FN) are the output labels that are predicted to be false, but they are actually true.

Specificity in machine learning is defined as:

Specificity is also called true negative rate or selectivity.

How to calculate specificity using sklearn in Python?

We can use the following Python code to calculate specificity using sklearn.

from sklearn.metrics import recall_score

y_true = [True, False, True, True, False, False, False, False, True, True]
y_pred = [False, False, True, True, False, False, True, False, True, False]

specificity = recall_score(y_true, y_pred, pos_label=0)
print("Specificity or True Negative Rate: ", specificity)

Please note that we are using the recall_score() function with the argument pos_label=0. When this pos_label argument is zero, the function returns the true negative rate or specificity. y_true here indicates the true values of the target variable. And y_pred indicates the predicted values of the target variable.

The output of the given program will be:

Specificity or True Negative Rate:  0.8