Which evaluation metric is commonly used for binary classification?

The Area Under the ROC Curve (AUC-ROC) is a widely recognized evaluation metric for binary classification tasks. This metric provides a way to evaluate the performance of a classification model at various threshold settings. The ROC curve itself plots the true positive rate (sensitivity) against the false positive rate (1-specificity) for different classification thresholds.

The AUC value ranges from 0 to 1, where a value of 0.5 suggests no discriminative ability (similar to random guessing), and a value of 1 indicates perfect classification. This makes it particularly valuable for assessing the trade-offs between sensitivity and specificity, allowing practitioners to understand the model's performance comprehensively.

In contrast, other options like Mean Squared Error are primarily used in regression contexts to measure the average of the squares of the errors. The Silhouette Score is a measure of how well clusters are formed in clustering tasks, not applicable to binary classification. Adjusted R-Squared is used for regression models to indicate the proportion of variance explained by the independent variables, rather than evaluating classification models. Therefore, the AUC-ROC remains the most suitable metric for binary classification evaluations.