Natural language processing basics quiz

Question 1

What does NLP stand for, and what is its core goal in computer science?

Accepted Answer

Natural Language Processing; enabling computers to understand and generate human language

Answer

Neural Logic Programming; building logic circuits

Answer

Network Layer Protocol; routing internet packets

Answer

Numeric List Processing; sorting numbers

Question 2

What is 'tokenization' in NLP?

Accepted Answer

Breaking text into smaller units like words or subwords

Answer

Encrypting text into secret codes

Answer

Translating text between languages

Answer

Removing all punctuation from text

Question 3

What are 'stop words' and why are they sometimes removed?

Accepted Answer

Common low-information words such as 'the' and 'is' often removed before analysis

Answer

Words that end a sentence

Answer

Misspelled words that must be corrected

Answer

Foreign words in a document

Question 4

What is the difference between stemming and lemmatization?

Accepted Answer

Stemming crudely chops endings; lemmatization returns a valid dictionary base form

Answer

They are identical processes with different names

Answer

Stemming uses grammar rules; lemmatization just removes suffixes

Answer

Lemmatization always produces non-words; stemming produces real words

Question 5

What is the 'Bag of Words' (BoW) model?

Accepted Answer

A representation of text as word counts that ignores word order

Answer

A model that keeps full word order and grammar

Answer

A neural network for translation

Answer

A method to compress text into a single number

Question 6

What does TF-IDF measure, and why is it useful?

Accepted Answer

How important a word is, balancing its frequency in a document against its rarity across all documents

Answer

The total number of words in a document

Answer

The grammatical role of each word

Answer

The sentiment score of a sentence

Question 7

What is a word embedding, such as those produced by Word2Vec?

Accepted Answer

A dense numeric vector capturing a word's meaning so similar words are close together

Answer

A unique random ID assigned to each word

Answer

A dictionary definition stored as text

Answer

The frequency count of a word in a corpus

Question 8

Why is one-hot encoding considered inefficient for representing large vocabularies?

Accepted Answer

It produces high-dimensional sparse vectors and captures no similarity between words

Answer

It is too compact to store enough information

Answer

It only works for numbers, not text

Answer

It loses the spelling of words

Question 9

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

Accepted Answer

Labeling each word with its grammatical category like noun or verb

Answer

Counting how many parts a document has

Answer

Splitting a paragraph into sentences

Answer

Detecting the language of the text

Question 10

What is Named Entity Recognition (NER)?

Accepted Answer

Identifying and classifying entities like people, places, and organizations in text

Answer

Recognizing the sentiment of a review

Answer

Naming variables in a program

Answer

Detecting spelling errors

Question 11

What is sentiment analysis?

Accepted Answer

Determining the emotional tone or opinion expressed in text

Answer

Counting the words in a sentence

Answer

Translating text into another language

Answer

Summarizing a long document

Question 12

What is an n-gram in NLP?

Accepted Answer

A contiguous sequence of n words or characters from text

Answer

A grammar rule for sentences

Answer

A type of neural network

Answer

A measure of document length

Question 13

What is a 'corpus' in NLP?

Accepted Answer

A large collection of text used to train and evaluate models

Answer

A single sentence used as an example

Answer

A grammar rulebook

Answer

A type of word embedding

Question 14

What does 'vocabulary' mean when building an NLP model?

Accepted Answer

The set of all unique tokens the model recognizes

Answer

The grammar rules of a language

Answer

The total number of documents

Answer

The list of stop words

Question 15

What is cosine similarity used for in NLP?

Accepted Answer

Measuring the angle between two vectors to judge similarity of text representations

Answer

Counting matching letters in two words

Answer

Calculating the length of a document

Answer

Sorting words alphabetically

Question 16

What is the difference between syntax and semantics in language?

Accepted Answer

Syntax is sentence structure/grammar; semantics is meaning

Answer

Syntax is meaning; semantics is grammar

Answer

They both refer only to spelling

Answer

Syntax is for speech; semantics is for text

Question 17

What is ambiguity in natural language, and why does it challenge NLP?

Accepted Answer

When a word or sentence has more than one possible meaning

Answer

When text contains spelling errors

Answer

When text is too long to process

Answer

When two languages are mixed

Question 18

What is text normalization and give two examples of normalization steps.

Accepted Answer

Converting text to a consistent form, e.g., lowercasing and removing punctuation

Answer

Translating text into English

Answer

Encrypting text for security

Answer

Adding grammar tags to words

Question 19

What is machine translation and which NLP architecture revolutionized it recently?

Accepted Answer

Translating text between languages; the Transformer architecture

Answer

Converting speech to text; the decision tree

Answer

Summarizing documents; the support vector machine

Answer

Detecting spam; the k-means algorithm

Question 20

In simple terms, what does the 'attention mechanism' do in modern NLP models?

Accepted Answer

It lets the model weigh which words are most relevant when producing each output

Answer

It deletes irrelevant words from the input

Answer

It counts how many times each word appears

Answer

It translates words one at a time in order

Question 21

What is a language model in NLP?

Accepted Answer

A model that assigns probabilities to word sequences and can predict the next word

Answer

A dictionary of word definitions

Answer

A list of grammar rules

Answer

A program that only counts letters

Question 22

What does it mean for a word to be 'out-of-vocabulary' (OOV)?

Accepted Answer

A word not present in the model's known vocabulary

Answer

A word that is grammatically incorrect

Answer

A word with multiple meanings

Answer

A word that appears too frequently

Question 23

What is subword tokenization (e.g., Byte Pair Encoding), and what problem does it solve?

Accepted Answer

Splitting words into smaller frequent pieces to handle rare/unknown words

Answer

Splitting text into sentences to improve speed

Answer

Removing prefixes and suffixes permanently

Answer

Merging all words into one token

Question 24

Why is word order important, and which simple model ignores it?

Accepted Answer

Word order carries meaning; the Bag of Words model ignores it

Answer

Word order is irrelevant; TF-IDF ignores it

Answer

Word order only matters in code; embeddings ignore it

Answer

Word order is preserved by all models

Question 25

What is text classification, and give one real-world example.

Accepted Answer

Assigning a category label to text; e.g., spam detection

Answer

Sorting words alphabetically; e.g., a dictionary

Answer

Translating text; e.g., a translator app

Answer

Counting characters; e.g., a word counter

Question 26

What is the role of a 'feature vector' when applying machine learning to text?

Accepted Answer

A numeric representation of text that ML algorithms can process

Answer

A picture of the text

Answer

A list of grammar rules

Answer

The original raw string itself

Question 27

What is the distributional hypothesis that underlies word embeddings?

Accepted Answer

Words appearing in similar contexts tend to have similar meanings

Answer

Words with similar spellings have similar meanings

Answer

Longer words carry more meaning

Answer

Word meaning depends only on its definition

Question 28

What is a chatbot's typical NLP pipeline at a high level?

Accepted Answer

It understands intent and entities, chooses a response, then generates a reply

Answer

It randomly picks replies from a list

Answer

It only translates the user's text

Answer

It stores text without processing it

Question 29

What is the difference between Natural Language Understanding (NLU) and Natural Language Generation (NLG)?

Accepted Answer

NLU understands/extracts meaning from input; NLG produces human-readable output

Answer

NLU generates text; NLG understands it

Answer

They are the same thing

Answer

NLU is for speech only; NLG is for text only

Question 30

Why might lowercasing text sometimes lose useful information?

Accepted Answer

It can erase signals like proper nouns and acronyms (e.g., 'Apple' vs 'apple')

Answer

It always improves accuracy with no downside

Answer

It changes the language of the text

Answer

It removes whole words from the text

Test yourself