Python for Machine Learning - The Complete Beginners Course - Hierarchical Clustering

Hierarchical clustering is a method of creating clusters with a predetermined order from top to bottom. It is similar to how files and folders are organized on a hard disk. This clustering method allows us to see the relationships between subclusters. There are two main strategies: divisive, which is a top-down approach starting with a large cluster and breaking it into smaller ones, and agglomerative, which is a bottom-up approach where each observation starts in its own cluster and clusters are merged as they move up the hierarchy.

What is the main characteristic of hierarchical clustering?

• It involves creating clusters with a predetermined order.

• It creates clusters with no specific order.

• It is unrelated to the organization of files and folders.

• It only uses a bottom-up approach.

Which of the following best describes divisive clustering?

• It does not involve any hierarchy.

• It begins with a large cluster and divides it into smaller ones.

• It is a bottom-up approach.

• It starts with individual observations and merges them.

In divisive clustering, what is the initial step?

• Start with individual observations.

• Merge clusters into a larger one.

• Ignore the hierarchy.

• Begin with a large cluster and break it down.

How does agglomerative clustering differ from divisive clustering?

• It starts with a large cluster and divides it.

• It is a top-down approach.

• Each observation starts in its own cluster and clusters are merged.

• It does not involve merging clusters.

What is the initial condition in agglomerative clustering?

• All observations are in a single large cluster.

• Each observation starts in its own cluster.

• Clusters are divided into smaller ones.

• There is no initial condition.