All MCQs

Complete linkage considers the farthest members of two clusters.
It favors compact groups with smaller diameters.

A narrow bridge of points can connect otherwise distinct groups.
This may create long, loosely connected clusters.

Single linkage merges clusters based on their nearest members.
It can recover elongated shapes but is prone to chaining.

A dendrogram is a tree whose branch heights reflect dissimilarity.
Cutting it at a chosen level produces a clustering.

Divisive clustering recursively splits broad groups into smaller clusters.
It builds the hierarchy from the top downward.

Agglomerative clustering repeatedly merges the closest clusters.
The process forms a bottom-up hierarchy.

An empty cluster has no points from which to compute a mean.
Algorithms handle this by relocating or replacing the centroid.

Absolute deviations and median centers reduce the influence of extremes.
This can produce more robust cluster locations.

K-medians minimizes an L1-style objective rather than squared error.
Medians are less sensitive to extreme values than means.

Values near one indicate a well-matched cluster assignment.
Values near or below zero suggest overlap or possible misassignment.

Within-cluster error always falls with more clusters.
The elbow marks where additional clusters yield diminishing improvement.

A distant observation can pull a centroid away from the main cluster.
Robust alternatives or outlier treatment may be needed.