Which clustering method starts with each data point as its own cluster and merges similar clusters?

Which clustering method starts with each data point as its own cluster and merges similar clusters?

• A. Density-based Clustering
• B. Divisive Clustering
• C. Agglomerative Clustering
• D. DBSCAN

Answer: (C)

This item tests agglomerative clustering, a bottom-up hierarchical approach. In this method you start with every data point as its own cluster and then repeatedly merge the pair of clusters that are most similar (or closest) according to a chosen linkage criterion, until you reach a stopping rule such as a desired number of clusters or a distance threshold. The idea is to build up larger clusters by successively combining the smallest, most similar pieces. Divisive clustering, by contrast, begins with all points in a single cluster and splits it into smaller groups, which is the opposite process. Density-based methods like DBSCAN group points based on density of the data space rather than merging clusters; they form clusters around dense regions and may leave points as noise, rather than progressively merging singleton clusters.

This item tests agglomerative clustering, a bottom-up hierarchical approach. In this method you start with every data point as its own cluster and then repeatedly merge the pair of clusters that are most similar (or closest) according to a chosen linkage criterion, until you reach a stopping rule such as a desired number of clusters or a distance threshold. The idea is to build up larger clusters by successively combining the smallest, most similar pieces.

Divisive clustering, by contrast, begins with all points in a single cluster and splits it into smaller groups, which is the opposite process. Density-based methods like DBSCAN group points based on density of the data space rather than merging clusters; they form clusters around dense regions and may leave points as noise, rather than progressively merging singleton clusters.