Question 15
Domain 2 — Data, Machine Learning, and Model DevelopmentWhich of the following metrics is typically used to evaluate classification models?
Correct answer: C
Explanation
Accuracy is a standard classification metric because it measures the proportion of predictions the model gets right. Classification models are commonly evaluated with metrics such as accuracy, precision, recall, and F1 score, and accuracy is the most general measure of overall performance.
Why each option is right or wrong
A. Mean Absolute Error
B. Root Mean Squared Error
C. Accuracy
Accuracy is the standard overall metric for classification tasks because it is defined as \((TP + TN) / (TP + TN + FP + FN)\), i.e., the proportion of all predictions that are correct. In a binary or multiclass classification setting, this directly measures how often the model assigns the right label, unlike regression metrics such as MSE or MAE, which are not appropriate for discrete class outcomes.
D. R-squared