wayground logo

Free Printable Worksheets

NEW

Font size

S
M
L
XL
Worksheets

Natural Language Processing Quiz

Total questions: 20

Worksheet time: 10mins

Name
Class
Date
1.

What is the main goal of Natural Language Processing (NLP)?

a)

Translate human language into computer code

b)

Create new programming languages

c)

Bridge the gap between human communication and machine understanding

d)

Improve image recognition accuracy

2.

Which of the following is NOT a core area of NLP?

a)

Speech recognition

b)

Machine translation

c)

Image segmentation

d)

Text processing

3.

Why is text preprocessing important in NLP?

a)

It removes valuable insights

b)

It makes the text harder to interpret

c)

It reduces data quality

d)

It cleans messy data and improves model accuracy

4.

What does tokenization do?

a)

Encrypts the text

b)

Splits text into smaller units like words or sentences

c)

Translates text into another language

d)

Summarizes long texts

5.

Which type of tokenization is used by models like BERT or GPT?

a)

Sentence tokenization

b)

Word tokenization

c)

Subword tokenization

d)

Character tokenization

6.

What is the risk of removing all stopwords during preprocessing?

a)

It makes training faster

b)

It increases vocabulary size

c)

It can remove important words like 'not' and change meaning

d)

It improves context understanding

7.

What is Part-of-Speech (POS) tagging?

a)

Removing unwanted characters from text

b)

Splitting text into subwords

c)

Assigning grammatical categories to each word

d)

Detecting spam messages

8.

What does Named Entity Recognition (NER) identify in text?

a)

Spelling errors

b)

Sentiment polarity

c)

Named entities like people, dates, and locations

d)

Text length

9.

What is the purpose of vectorization in NLP?

a)

Translate text into speech

b)

Convert text into numerical format for machine learning

c)

Compress text files

d)

Detect named entities

10.

Which method considers the importance of words across the entire corpus?

a)

Bag of Words

b)

Lemmatization

c)

Word Embeddings

d)

TF-IDF

11.

Why are word embeddings preferred over BoW?

a)

They reduce training time

b)

They are easier to compute

c)

They capture semantic meaning and word relationships

d)

They do not require any preprocessing

12.

Which traditional ML model is commonly used for binary text classification?

a)

Decision Trees

b)

Logistic Regression

c)

Naive Bayes

d)

SVM

13.

What makes Naive Bayes 'naive'?

a)

It ignores punctuation

b)

It assumes words are dependent

c)

It uses no training data

d)

It assumes word occurrences are independent

14.

What is a key advantage of Naive Bayes?

a)

Requires large datasets

b)

Performs poorly on text data

c)

Fast and effective on small datasets

d)

Always requires embeddings

15.

What is the main idea behind Support Vector Machines (SVM)?

a)

Predict sequences with high accuracy

b)

Find the optimal decision boundary that separates classes

c)

Randomly classify data

d)

Use Naive assumptions about text

16.

What are support vectors in SVM?

a)

Vectors that contain stopwords

b)

The points farthest from the hyperplane

c)

The closest data points to the decision boundary

d)

Words that represent vector embeddings

17.

What makes Random Forest more robust than a single decision tree?

a)

It always gives the same output

b)

It uses multiple trees and combines their outputs

c)

It removes the need for preprocessing

d)

It trains slower than trees

18.

What is a major limitation of basic RNNs?

a)

Too fast for training

b)

Limited accuracy on structured data

c)

Vanishing gradient problem with long sequences

d)

Inability to work with text

19.

What does the Forget Gate in LSTM do?

a)

Decides what information to discard from memory

b)

Stores the input

c)

Prevents dropout

d)

Tracks activation functions

20.

What is the main difference between GRU and LSTM?

a)

GRU has fewer gates and is computationally simpler

b)

LSTM is used only for images

c)

GRU requires labeled data

d)

LSTM cannot handle sequences