How Does Class Imbalance Affect Sensitivity and Specificity?
Sensitivity, or the true positive rate, measures how effectively a model identifies positive cases (e.g., individuals with a disease). Specificity, or the true negative rate, measures how effectively a model identifies negative cases (e.g., individuals without the disease). In a class-imbalanced dataset, a model might perform well on the majority class, leading to high specificity but low sensitivity. This imbalance can result in many false negatives, where actual disease cases are missed, rendering the model unreliable for public health decisions.
