Hierarchical clustering

Question 1

Hierarchical clustering algorithm operates in _______ fashion and why

Answer 1

Hierarchical clustering algorithms typically operate in a greedy fashion, making locally optimal choices at each step (merging the closest clusters or spitting the largest clusters) without reconsidering previous steps.

Question 2

Hierarchal clustering is __________-

Answer 2

divide and conquer clustering

Question 3

Another name of agglomerative clustering

Answer 3

Bottom up approach

Question 4

Another name of agglomerative clusetring

Answer 4

Top down approach

Question 5

Hierarchical clustering can be used for what or cant’s be used for what

Answer 5

Hierarchical clustering can be used for outlier detection but not for finding missing values (NA) or detected fake values.

Question 6

Hierarchical clustering is primarily used for ______ because ________

Answer 6

Hierarchical clustering is primarily used for exploration because it helps in understanding the natural grouping within data which can be very useful in exploratory data analysis.

Question 7

Hierarchical clustering is _________ visualization

Answer 7

Dendogram visualization

Question 8

In hierarchical clustering do we need to specify the number of clusters?

Answer 8

No need to specify the number of clusters in hierarchical clustering

Question 9

How hierarchical clustering provides flexibility or not

Answer 9

It allows you to choose the number of clusters by cutting the dendrogram at different levels, providing flexibility to explore the data at different granularities.

Question 10

Hierarchal clustering is deterministic or not

Answer 10

Hierarchal clustering is deterministic because it allows a fixed sequence of merging or splitting clusters based on defined criteria like distance

Question 11

Linkage (definition and types)

Answer 11

Linkage is how to link the clusters
Linkage techniques are two types: Single linkage and complete linkage

Question 12

Single linkage
* Another name
* Keyword
* Definition
* Formula

Answer 12

• Another name: Nearest neighbour method
• Keyword: shortest distance
• Definition: This linkage technique focused on the shortest distance between data points in each cluster.

Question 13

Complete linkage
* Another name
* Keyword
* Definition
* Formula

Answer 13

• Another name: Farthest neighbour method
• Keyword: longest distance
• Definition: This linkage technique focused on the longest distance between data points in each cluster.

Question 14

Agglomerative clustering keyword

Answer 14

Merging approach

Question 15

Agglomerative clustering use which linkage

Answer 15

can use any linkage
Single linkage or complete linkage

Question 16

Decisive clustering use which linkage

Answer 16

Decisive linkage use only complete linkage

Question 17

Remember point in agglomerative clustering problem

Answer 17

Question 18

Average linkage technique

Answer 18

Question 19

Decisive clustering keyword

Answer 19

Splitting approach

Question 20

How to do problem of decisive clustering

Answer 20

We create Minimal spanning tree (MST) based on dissimilar matrix

Question 21

Minimal spanning tree characteristics (4)

Answer 21

It is a connected tree
No loops/ no closed circuits in the tree.
Each data point(node) in the tree is visited atleast once.
If ‘n’ nodes are present in the tree, then (n-1) edges are present or formed in the tree.

Question 22

If there is n nodes in MST then

Answer 22

If ‘n’ nodes are present in the tree, then (n-1) edges are present or formed in the tree.

Question 23

Remember point in decisive clustering problem

Answer 23

Question 24

Explain number of levels in hierarchy in both agglomerative clustering and decisive clustering

Answer 24

Agglomerative clustering: If there are n observations, then there will be n-1 levels in the hierarchy.Since n−1 merges are required to combine n observations into a single cluster, the hierarchy has n−1 levels.

Decisive clustering: The number of levels depends on the way splits occur (e.g., binary splits may create more or fewer than n−1 levels).

Question 25

Ward’s method

Answer 25

Merging cluster way.
In this technique, we minimise the increase in variance when merging clusters. Repeat this process iteratively until all data points are in a single cluster or until reach the desired number of clusters.

Similarity of two clusters is based on the increase in squared error when two clusters are merged.
(Similar to group average if distance between points is distance squared).
Less susceptible to noise and outliers.
Biased towards global clusters.
Hierarchical analogue of k-means
(can be used to initialise k-means)

Question 26

In Divisive clustering, explain split clusters in terms of variance

Answer 26

Question 27

Who is more sensitive to outliers (single linkage/ multiple linkage)

Answer 27

Single linkage is more sensitive to outliers

Question 28

Single linkage characteristics

Answer 28

Question 29

Another name of multiple linkage

Answer 29

Complete linkage or maximum linkage

Question 30

Multiple linkage characteristics

Answer 30

Question 31

Which one is more expensive (Divisive/ agglomerative)

Answer 31

Divisive clustering is computationally more expensive than agglomerative clustering because it requires considering all possible splits at each.

Question 32

Which one uses more (Divisive / agglomerative)

Answer 32

Agglomerative is more commonly used whereas divisive is less commonly used due to its complexity.

Question 33

If we solve question of divisive clustering using k means, what should be the value of k

Answer 33

k=2

Question 34

Which is more efficient (k means/ hierarchal clustering)

Answer 34

k means