Answer: F1-score

The F1-score is the preferred metric for evaluating multiclass classification models with the MulticlassClassificationEvaluator. It effectively balances precision and recall, making it ideal for assessing models where both the identification of true positives and the avoidance of false positives are crucial. This is particularly beneficial in scenarios with imbalanced class distributions. Other metrics like R-squared and Mean Squared Error (MSE) are more suited for regression tasks, while ROC AUC, though useful for binary classification, is less commonly used for multiclass problems without adaptation. Thus, the F1-score stands out as the most appropriate choice for multiclass classification evaluation.

Author: LeetQuiz Editorial Team