Question 35
Domain 2: Describe fundamental principles of machine learning on AzureWhich is not a metric for clustering models?
Correct answer: A
Explanation
Mean Absolute Error is a regression metric, measuring the average absolute difference between predicted and actual values. Clustering is evaluated with metrics like silhouette score, Davies-Bouldin index, or within-cluster sum of squares, which assess group cohesion and separation rather than prediction error.
Why each option is right or wrong
A. Mean Absolute Error
Mean Absolute Error is defined as an error measure for supervised prediction tasks, calculated as the average of |y - ŷ| across observations, so it belongs to regression/forecasting evaluation rather than clustering. Clustering quality is instead assessed with internal indices such as silhouette score, Davies–Bouldin index, or within-cluster sum of squares, which evaluate cohesion and separation of unlabeled groups rather than pointwise prediction error.
B. Average Distance to Cluster Center
C. Average Distance to Other Center
D. Number of Points