Hierarchical clustering in ML Python - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

Understanding linkage methods in hierarchical clustering

Which linkage method in hierarchical clustering merges clusters based on the maximum distance between points in the two clusters?

A Average linkage

B Single linkage

C Complete linkage

D Ward linkage

❓ Predict Output

intermediate

Output of dendrogram linkage matrix

What is the output of the following Python code snippet using scipy for hierarchical clustering linkage matrix?

ML Python

from scipy.cluster.hierarchy import linkage
import numpy as np
X = np.array([[1, 2], [3, 4], [5, 6]])
Z = linkage(X, method='single')
print(Z)

A

[[0. 1. 2.82842712 2.]
 [2. 3. 4.24264069 2.]
 [4. 5. 5.65685425 3.]]

B

[[0. 1. 2.82842712 2.]
 [2. 3. 2.82842712 2.]
 [4. 5. 4.24264069 3.]]

C

[[0. 1. 2.82842712 2.]
 [2. 3. 2.82842712 2.]
 [4. 5. 5.65685425 3.]]

D

[[0. 1. 1.41421356 2.]
 [2. 3. 2.82842712 2.]
 [4. 5. 4.24264069 3.]]

❓ Model Choice

advanced

Choosing hierarchical clustering for data type

Which type of dataset is hierarchical clustering most suitable for compared to k-means clustering?

A Data with non-globular clusters and unknown number of clusters

B High-dimensional sparse data with many irrelevant features

C Large datasets with millions of samples

D Data with clearly separated spherical clusters and known cluster count

❓ Hyperparameter

advanced

Effect of distance metric on hierarchical clustering

How does changing the distance metric from Euclidean to Manhattan affect hierarchical clustering results?

A Changing distance metric only affects k-means, not hierarchical clustering

B Clusters will always be identical regardless of distance metric

C Manhattan distance causes hierarchical clustering to fail with an error

D Clusters may differ because Manhattan distance measures distance along axes, affecting cluster shape

❓ Metrics

expert

Interpreting cophenetic correlation coefficient

What does a cophenetic correlation coefficient close to 1 indicate about a hierarchical clustering result?

A The clustering perfectly preserves the pairwise distances between original points

B The clustering poorly represents the original data distances

C The clustering has exactly one cluster containing all points

D The clustering uses single linkage method