wayground logo

Free Printable Worksheets

NEW

Font size

S
M
L
XL
Worksheets

MCQs on Introduction to Clustering & Partitional Methods

Total questions: 60

Worksheet time: 30mins

Name
Class
Date
1.

What is the main goal of clustering?

a)

Classification of data

b)

Prediction of outcomes

c)

Grouping similar data objects

d)

Data compression

2.

A cluster is defined as a collection of data objects that are

a)

Randomly distributed

b)

Similar within a cluster and dissimilar to other clusters

c)

Identical in nature

d)

Always numerical

3.

Clustering is widely used in all EXCEPT

a)

Image processing

b)

Marketing research

c)

Crime analysis

d)

Compiler design

4.

Which metric is fundamental for clustering algorithms?

a)

Sorting order

b)

Distance or similarity

c)

Frequency count

d)

Probability value

5.

Clustering numeric data primarily uses

a)

Logical operators

b)

Geometric properties

c)

Lexicographic ordering

d)

Hash functions

6.

Why is clustering categorical data difficult?

a)

Too many values

b)

Distance function is not well-defined

c)

Requires normalization

d)

Needs labeled data

7.

Which of the following is NOT a stage of clustering process?

a)

Data collection

b)

Clustering tendency

c)

Validation

d)

Classification

8.

Scalability is a requirement of clustering algorithms because

a)

Data sets are very small

b)

Data sets are very large

c)

Algorithms are simple

d)

Clusters are predefined

9.

Which clustering paradigm forms a nested grouping of data?

a)

Partitioning

b)

Density-based

c)

Hierarchical

d)

Grid-based

10.

The graphical representation of hierarchical clustering is called

a)

Scatter plot

b)

Histogram

c)

Dendrogram

d)

Decision tree

11.

Agglomerative clustering follows which approach?

a)

Top-down

b)

Bottom-up

c)

Randomized

d)

Parallel

12.

In agglomerative clustering, initially

a)

One cluster exists

b)

All objects are in the same cluster

c)

Each object forms a separate cluster

d)

Clusters are predefined

13.

Divisive clustering follows which approach?

a)

Bottom-up

b)

Top-down

c)

Greedy

d)

Randomized

14.

DIANA is an example of

a)

Agglomerative clustering

b)

Partitional clustering

c)

Divisive clustering

d)

Density-based clustering

15.

A major disadvantage of hierarchical clustering is

a)

High accuracy

b)

Cannot handle numeric data

c)

Once merged or split, it cannot be undone

d)

Requires predefined k

16.

Partitional clustering divides data into

a)

Variable number of clusters

b)

Predefined number of clusters

c)

Overlapping clusters

d)

Hierarchical clusters

17.

The quality of partitional clustering is improved by

a)

Sorting data

b)

Swapping data points

c)

Removing attributes

d)

Increasing dimensions

18.

Which of the following is a partitional clustering method?

a)

DIANA

b)

AGNES

c)

K-Means

d)

BIRCH

19.

In partitional clustering, prototypes are

a)

Always random

b)

Never updated

c)

Iteratively recomputed

d)

Fixed throughout

20.

The number of ways to partition n objects into k clusters is

a)

n!

b)

k!

c)

kn/k!k^n / k!

d)

nkn^k

21.

In K-Means, each cluster is represented by

a)

Median

b)

Mode

c)

Mean

d)

Boundary

22.

The centroid in K-Means is

a)

A random object

b)

The mean of objects in the cluster

c)

The farthest object

d)

Always an actual data object

23.

K-Means algorithm is sensitive to

a)

Noise only

b)

Duplicate data

c)

Outliers

d)

Missing values

24.

Which step assigns objects to nearest centroid?

a)

Initialization

b)

Computation

c)

Assignment

d)

Validation

25.

K-Means stops when

a)

Maximum iterations reached

b)

No change in cluster assignment

c)

Error becomes zero

d)

Data is sorted

26.

K-Medoid represents each cluster using

a)

Mean value

b)

Median value

c)

One actual object

d)

Virtual point

27.

PAM stands for

a)

Partition and Merge

b)

Partition Around Medoids

c)

Parallel Analysis Model

d)

Pattern Association Method

28.

PAM works effectively for

a)

Very large datasets

b)

Streaming data

c)

Small datasets

d)

High-dimensional sparse data

29.

PAM improves clustering quality by

a)

Sampling

b)

Swapping medoids

c)

Merging clusters

d)

Normalization

30.

PAM has how many main modules?

a)

One

b)

Two

c)

Three

d)

Four

31.

CLARA is designed to handle

a)

Small datasets

b)

Numerical data only

c)

Larger datasets than PAM

d)

Hierarchical data

32.

CLARA applies PAM on

a)

Entire dataset

b)

Random clusters

c)

Sample of the dataset

d)

Sorted data

33.

A weakness of CLARA is

a)

High computation cost

b)

Biased sampling

c)

Inability to handle numeric data

d)

Sensitivity to outliers

34.

CLARANS is based on

a)

Exhaustive search

b)

Randomized search

c)

Greedy search

d)

Sequential search

35.

CLARANS improves scalability by

a)

Using full dataset

b)

Avoiding medoids

c)

Sampling neighbors dynamically

d)

Fixed clustering

36.

CLARANS stops when

a)

All objects are clustered

b)

Best global solution is found

c)

A fixed number of local optima are found

d)

Error becomes zero

37.

Which algorithm is more scalable?

a)

PAM

b)

K-Means

c)

CLARA

d)

CLARANS

38.

Which method uses mean as cluster center?

a)

K-Medoid

b)

PAM

c)

K-Means

d)

CLARA

39.

Which algorithm cannot undo merge or split decisions?

a)

K-Means

b)

PAM

c)

CLARA

d)

Hierarchical clustering

40.

Clustering tendency helps to

a)

Label data

b)

Decide number of clusters

c)

Measure similarity

d)

Store data

41.

Clustering is a type of learning where

a)

Supervised learning

b)

Unsupervised learning

c)

Semi-supervised learning

d)

Reinforcement learning

42.

Which of the following best describes clustering?

a)

Assigning labels based on prior knowledge

b)

Discovering hidden patterns without class labels

c)

Predicting future values

d)

Reducing dimensionality

43.

In clustering, similarity between objects is generally measured using

a)

Classification rules

b)

Distance functions

c)

Decision boundaries

d)

Probability distributions

44.

Which of the following is NOT a characteristic of a good clustering?

a)

High intra-cluster similarity

b)

Low inter-cluster similarity

c)

Random grouping of objects

d)

Compact clusters

45.

Which data type is easiest to cluster using traditional distance measures?

a)

Categorical

b)

Text

c)

Numerical

d)

Boolean

46.

One major challenge in clustering large databases is

a)

Availability of labels

b)

High dimensionality

c)

Small dataset size

d)

Lack of algorithms

47.

Which factor strongly affects the outcome of clustering?

a)

Data ordering

b)

Choice of distance metric

c)

File format

d)

Storage device

48.

Clustering differs from classification because clustering

a)

Uses training data

b)

Requires class labels

c)

Does not require predefined classes

d)

Is less accurate

49.

The value of k in K-Means represents

a)

Number of attributes

b)

Number of iterations

c)

Number of clusters

d)

Number of data points

50.

Choosing an inappropriate value of k may lead to

a)

Perfect clustering

b)

Overfitting or underfitting

c)

Faster convergence

d)

Reduced computation

51.

Which distance measure is most commonly used in K-Means?

a)

Manhattan distance

b)

Euclidean distance

c)

Cosine similarity

d)

Jaccard coefficient

52.

Initial selection of centroids in K-Means affects

a)

Data storage

b)

Execution speed only

c)

Final clustering result

d)

Input format

53.

K-Means may converge to different results because

a)

It is deterministic

b)

It uses random initialization

c)

It sorts data first

d)

It uses categorical data

54.

Which of the following is a limitation of K-Means?

a)

Works only for hierarchical data

b)

Cannot handle numerical data

c)

Assumes spherical-shaped clusters

d)

Requires labeled data

55.

K-Means performs poorly when clusters

a)

Are well separated

b)

Have similar sizes

c)

Are non-convex or irregular in shape

d)

Are compact

56.

Which situation makes K-Means unsuitable?

a)

Numeric attributes

b)

Presence of outliers

c)

Low dimensional data

d)

Small dataset size

57.

The objective function minimized by K-Means is

a)

Maximum distance between clusters

b)

Sum of squared distances within clusters

c)

Number of clusters

d)

Number of iterations

58.

In K-Means, reassignment of objects continues until

a)

Maximum error is reached

b)

No object changes its cluster

c)

Data is normalized

d)

All clusters have equal size

59.

K-Means requires which type of input attributes?

a)

Categorical only

b)

Ordinal only

c)

Numerical only

d)

Mixed type

60.

Which statement about K-Means is TRUE?

a)

Cluster centers must be actual data points

b)

Clusters can overlap

c)

The algorithm guarantees global optimum

d)

Results depend on initial centroids