Hierarchical Agglomerative Clustering Concepts

The video tutorial introduces hierarchical agglomerative clustering (HAC), a bottom-up approach to clustering where individual items are iteratively merged into clusters until a single cluster is formed. It explains three linkage methods: single, complete, and average linkage, and describes how dendrograms visually represent the clustering process. The tutorial provides a detailed example of single linkage clustering, demonstrating the calculation of distances and the formation of clusters using a distance matrix.

What is the primary approach used in Hierarchical Agglomerative Clustering?

• Top-down

• Bottom-up

• Sideways

• Random

What structure is used to represent hierarchical clustering?

• Matrix

• Dendrogram

• Tree

• Graph

Which linkage method considers the distance between the closest members of two clusters?

• Weighted linkage

• Complete linkage

• Average linkage

• Single linkage

What formula is used to calculate the distance in single linkage?

• Jaccard index

• Manhattan distance

• Euclidean distance

• Cosine similarity

In the example, which two points form the first cluster using single linkage?

• P4 and P5

• P1 and P2

• P2 and P5

• P3 and P6

What is the next step after forming the initial cluster in single linkage?

• Calculate the average distance

• Recalculate the distance matrix

• Stop the process

• Form a new cluster with all points

What is the final outcome of the hierarchical agglomerative clustering process?

• A list of individual items

• Multiple small clusters

• A random assortment of clusters

• A single cluster containing all items