wayground logo

Free Printable Worksheets

NEW

Font size

S
M
L
XL
Worksheets

NLP CLASS TEST - 2

Total questions: 65

Worksheet time: 34mins

Name
Class
Date
1.

ELIZA mainly worked using:

a)

Deep neural networks

b)

Semantic parsing

c)

World knowledge

d)

Pattern matching

2.

ELIZA gave the illusion of intelligence because:

a)

It understood emotions

b)

It reasoned logically

c)

It had memory

d)

Replies sounded relevant

3.

ELIZA did NOT actually understand:

a)

Grammar

b)

Turn-taking

c)

Dialogue flow

d)

Meaning of words

4.

ELIZA rules mainly used:

a)

Parsing trees

b)

Knowledge graphs

c)

Probabilities

d)

IF–THEN patterns

5.

In NLP, words are treated as:

a)

Emotions

b)

Concepts

c)

Sentences

d)

Data units

6.

Large language models usually treat punctuation as:

a)

Noise

b)

Ignored

c)

Errors

d)

Separate tokens

7.

An utterance is:

a)

Only a written sentence

b)

Grammar rule

c)

Paragraph

d)

Continuous spoken language

8.

“uh” and “um” are examples of:

a)

Prefixes

b)

Tokens

c)

Corpus

d)

Fillers

9.

Speech disfluencies include:

a)

Only pauses

b)

Only repetition

c)

Only grammar errors

d)

False starts and fillers

10.

A corpus is:

a)

One document

b)

Grammar rule set

c)

Vocabulary list

d)

Large collection of text/speech

11.

Word types refer to:

a)

All tokens

b)

Word positions

c)

Sentences

d)

Unique words

12.

Vocabulary size equals:

a)

Tokens

b)

Sentences

c)

Characters

d)

Number of word types

13.

Word instances count:

a)

Unique words

b)

Characters

c)

Lines

d)

Every occurrence

14.

Morphemes are:

a)

Letters

b)

Sounds

c)

Sentences

d)

Minimal meaning units

15.

“un-happy” consists of:

a)

Two characters

b)

Two tokens

c)

Two words

d)

Two morphemes

16.

Free morphemes can:

a)

Attach only

b)

Change tense

c)

Modify grammar

d)

Stand alone

17.

“-ed” is a:

a)

Root

b)

Free morpheme

c)

Word

d)

Bound morpheme

18.

Inflectional morphemes change:

a)

Word class

b)

Meaning completely

c)

Vocabulary

d)

Grammar form

19.

Derivational morphemes:

a)

Only pluralize

b)

Only tense

c)

Never change class

d)

Create new words

20.

Tokenization means:

a)

Parsing grammar

b)

Translating

c)

Tagging

d)

Breaking text into units

21.

Tokenization errors affect:

a)

Only parsing

b)

Storage

c)

Fonts

d)

All NLP tasks

22.

Tokenization depends mainly on:

a)

CPU speed

b)

Dataset size

c)

Randomness

d)

Task and language

23.

Regex stands for:

a)

Rule generator

b)

Grammar engine

c)

Parser

d)

Regular expression

24.

Regex is used to:

a)

Translate

b)

Learn meaning

c)

Classify sentiment

d)

Match patterns

25.

Pattern cat will match:

a)

catalog

b)

cats

c)

scatter

d)

cat

26.

Wildcard . matches:

a)

Nothing

b)

Digits only

c)

Letters only

d)

Any single character

27.

Regex [aeiou] matches:

a)

Any word

b)

Numbers

c)

Punctuation

d)

A vowel

28.

[0-9]+ is used to match:

a)

Names

b)

Symbols

c)

Emails

d)

Numbers

29.

[a-zA-Z]+ matches:

a)

Digits

b)

Symbols

c)

Punctuation

d)

Alphabetic words

30.

Tokenization is:

a)

Universal

b)

Fixed

c)

Random

d)

Design choice

31.

A filler may help speech recognition by:

a)

Ending sentences

b)

Creating noise

c)

Removing words

d)

Predicting restarts

32.

Fragment “main-” is:

a)

Token

b)

Word type

c)

Root

d)

Disfluency

33.

Regex for punctuation [.,!?] matches:

a)

Words

b)

Numbers

c)

Emails

d)

Symbols

34.

Tokenization is foundation of:

a)

Hardware

b)

Storage

c)

Displays

d)

NLP pipelines

35.

Corpus examples include:

a)

RAM

b)

CPU

c)

Token

d)

News articles

36.

Word “cats” contains:

a)

One morpheme

b)

Four tokens

c)

Letters only

d)

Two morphemes

37.

“teacher” shows:

a)

Inflection only

b)

Prefixing

c)

Plurality

d)

Derivation

38.

Regex helps mainly with:

a)

Semantics

b)

Pragmatics

c)

Reasoning

d)

Preprocessing

39.

In ELIZA, sadness was:

a)

Understood

b)

Reasoned

c)

Stored

d)

Pattern-matched

40.

Sentence vs utterance differs in:

a)

Meaning

b)

Tokenization

c)

Grammar

d)

Speech vs writing

41.

Counting punctuation depends on:

a)

Font

b)

Storage

c)

Language only

d)

Task

42.

Morpheme study is called:

a)

Syntax

b)

Phonology

c)

Pragmatics

d)

Morphology

43.

Word instance example “the” twice counts as:

a)

One

b)

Two types

c)

Zero

d)

Two instances

44.

Regex email example contains:

a)

Only letters

b)

Spaces

c)

Slashes

d)

@ symbol

45.

Disfluencies may sometimes be:

a)

Removed always

b)

Ignored fully

c)

Forbidden

d)

Kept for prediction

46.

Tokenization precedes:

a)

Printing

b)

Saving

c)

Display

d)

Sentiment analysis

47.

Pattern-based chatbots rely on:

a)

Knowledge bases

b)

Planning

c)

Learning

d)

Rules

48.

Corpus vocabulary is denoted by:

a)

T

b)

C

c)

N

d)

V

49.

Regex describes:

a)

Exact sentence

b)

Meaning

c)

Tree

d)

String patterns

50.

Bound morphemes cannot:

a)

Change grammar

b)

Modify meaning

c)

Add tense

d)

Stand alone

51.

Tokenization example “don’t → do + n’t” shows:

a)

Lemmatization

b)

Parsing

c)

Stemming

d)

Task-dependent split

52.

Utterances include:

a)

Only paragraphs

b)

Only books

c)

Only essays

d)

Single words

53.

NLP pipeline always begins with:

a)

Translation

b)

Parsing

c)

Learning

d)

Tokens

54.

Regex used for searching text is:

a)

Random

b)

Statistical

c)

Neural

d)

Rule-based

55.

“play → played” uses:

a)

Derivation

b)

Prefix

c)

Root change

d)

Inflection

56.

“happy → unhappy” shows:

a)

Inflection

b)

Tokenization

c)

Plurality

d)

Derivation

57.

Speech systems must handle:

a)

Only silence

b)

Only grammar

c)

Only tokens

d)

Disfluencies

58.

Corpus size affects:

a)

Fonts

b)

Grammar

c)

Punctuation

d)

Vocabulary

59.

Morphemes differ from characters because they:

a)

Are letters

b)

Are sounds

c)

Are punctuation

d)

Carry meaning

60.

Regex [a-z]+@[a-z]+\.com matches:

a)

URL

b)

Number

c)

Email

d)

Date

61.

The Kleene closure of a language L, written as L*, represents:

a)

Zero or more concatenations of strings from L

b)

Only one repetition of strings in L

c)

All strings of length exactly two from L

d)

All strings formed by infinite repetition

62.

The Kleene positive closure of a language L, written as L⁺, represents:

a)

Zero or more repetitions

b)

One or more concatenations of strings from L

c)

Only the empty string

d)

At most two repetitions

63.

If L = {ab}, then L⁺ contains:

a)

b

b)

ε

c)

ab

d)

ab, abab, ababab, …

64.

Tokenization is described as a design choice in NLP mainly because:

a)

It depends on the language, application, and task requirements

b)

All languages follow identical grammar rules

c)

Computers cannot store raw text

d)

Tokens must always be characters

65.

Which of the following tokenizations best reflects a task-dependent NLP design choice as described in the notes?

a)

don’t → don + t

b)

New-York → New-York

c)

don’t → do + n’t

d)

don’t → dont

Similar Resources on Wayground

MICROCONTROLLER CAT-III_2025_EVEN

60 questions

MICROCONTROLLER CAT-III_2025_EVEN

quiz

Worksheet

•

University

Kiến trúc máy tính

62 questions

Kiến trúc máy tính

quiz

Worksheet

•

University

C Programming Functions and Storage Classes

60 questions

C Programming Functions and Storage Classes

quiz

Worksheet

•

University

EMGT PRACTICE

60 questions

EMGT PRACTICE

quiz

Worksheet

•

University

Introduction to Geotechnical Engineering

66 questions

Introduction to Geotechnical Engineering

quiz

Worksheet

•

University

Semiconductor and Diode Self Learning Activity 1

60 questions

Semiconductor and Diode Self Learning Activity 1

quiz

Worksheet

•

University

python_string_set_dict

60 questions

python_string_set_dict

quiz

Worksheet

•

University

Basic Web Technologies Quiz

65 questions

Basic Web Technologies Quiz

quiz

Worksheet

•

University