SRM INSTITUTE OF SCIENCE AND TECHNOLOGY

FACULTY OF ENGINEERING AND TECHNOLOGY

SCHOOL OF COMPUTER SCIENCE ENGINEERING

QUESTION BANK

Degree & Branch : B.TECH- CSE& Specialization

Course Outcomes
CO1: Exhibit knowledge on text preprocessing techniques and perform word level analysis.
CO2: Illustrate approaches to syntax analysis including probabilistic context-free grammars
CO3: Apply approaches to semantics and discourse nalysis in NLP.
CO4: Develop models using transfer learning approaches.
CO5: Implement applications that use Natural Language Processing approaches
 Part A
Q1. What is Natural Language Processing (NLP)?
A) A programming language for AI
B) A field of AI that enables machines to understand and generate human language
C) A software that translates human languages into binary code
D) A subset of cryptography
Answer: B) A field of AI that enables machines to understand and generate human
language
Q2. Which of the following is NOT an application of NLP?
A) Machine Translation
B) Speech Recognition
C) Image Processing
D) Chatbots
Answer: C) Image Processing
Q3. Which level of NLP deals with the meaning of words and sentences?
A) Morphological level
B) Syntactic level
C) Semantic level
D) Pragmatic level
Answer: C) Semantic level
Q4. In regular expressions, what does the symbol \d represent?
A) Any alphabetic character
B) Any whitespace character
C) Any digit (0-9)
D) Any special character
Answer: C) Any digit (0-9)
Q5. What is the main purpose of morphological analysis in NLP?
A) To analyze the meaning of sentences
B) To break words into their root forms and affixes
C) To detect grammatical errors
D) To perform image recognition
Answer: B) To break words into their root forms and affixes
Q6. Tokenization in NLP refers to:
A) Converting text into speech
B) Splitting text into smaller meaningful units like words or sentences
C) Removing stopwords from text
D) Translating text into another language
Answer: B) Splitting text into smaller meaningful units like words or sentences
Q7. Which of the following is a stemming algorithm?
A) WordNet
B) Porter’s Algorithm
C) N-grams
D) TF-IDF
Answer: B) Porter’s Algorithm
Q8. What is the key difference between stemming and lemmatization?
A) Stemming provides the root form without considering meaning, while lemmatization
provides a linguistically correct base form
B) Stemming and lemmatization are the same
C) Lemmatization removes stopwords, while stemming does not
D) Stemming converts words into numbers
Answer: A) Stemming provides the root form without considering meaning, while
lemmatization provides a linguistically correct base form
Q9. What does Term Frequency (TF) measure in NLP?
A) The number of times a term appears in a document
B) The importance of a term across multiple documents
C) The similarity between two words
D) The number of documents containing a term
Answer: A) The number of times a term appears in a document
Q10. What is the main purpose of Inverse Document Frequency (IDF)?
A) To count the total number of words in a document
B) To reduce the weight of frequently occurring words across multiple documents
C) To increase the importance of common words
D) To translate text into another language
Answer: B) To reduce the weight of frequently occurring words across multiple
documents
Q11. What is the main advantage of using the TF-IDF model?
A) It completely removes stopwords
B) It gives higher importance to rare words and reduces the importance of common
words
C) It translates words into numerical values
D) It is faster than deep learning models
Answer: B) It gives higher importance to rare words and reduces the importance of
common words
Q12. Which of the following is NOT a part of speech (POS) category in NLP?
A) Noun
B) Verb
C) Convolutional Neural Network
D) Adjective
Answer: C) Convolutional Neural Network
Q13. Named Entity Recognition (NER) is used to identify:
A) The structure of a sentence
B) Important entities such as names, locations, and organizations in text
C) The root form of a word
D) The frequency of words in a document
Answer: B) Important entities such as names, locations, and organizations in text
Q14. What is the primary use of N-grams in NLP?
A) To identify grammatical errors
B) To split text into sentences
C) To model text sequences and predict the next word
D) To perform translation
Answer: C) To model text sequences and predict the next word
Q15. What is the main purpose of smoothing in NLP models?
A) To remove unnecessary words from a document
B) To handle zero probabilities in language models
C) To increase the complexity of text
D) To improve text-to-image conversion
Answer: B) To handle zero probabilities in language models
Q16. Which of the following regular expression patterns will correctly match an email
address format (assuming a basic email structure)?
A) [A-Za-z0-9]+@[A-Za-z]+\.[A-Za-z]{2,}
B) \d{10}@\w+\.\w{2,}
C) (\w+\.)?\w+@\w+\.\w+
D) .*@.*\..*
Answer: A) [A-Za-z0-9]+@[A-Za-z]+\.[A-Za-z]{2,}
Explanation: Option A correctly captures the general email format with a username, @
symbol, and domain name, while the others either allow incorrect formats or miss
necessary components.
Q17. Which of the following words would be considered a morpheme in the English
language?
A) Running
B) Unhappiness
C) Cat
D) All of the above
Answer: D) All of the above
Q18. Suppose you have a document containing 500 words, and the word "network"
appears 10 times. If "network" appears in 100 out of 10,000 documents in a corpus, what
is the TF-IDF score for "network"? (Use log base 10 for IDF calculation)
A) TF = 0.02, IDF = 2, TF-IDF = 0.04
B) TF = 0.02, IDF = 2, TF-IDF = 0.02
C) TF = 0.05, IDF = 2, TF-IDF = 0.1
D) TF = 0.05, IDF = 1, TF-IDF = 0.05
Answer: C) TF = 0.02, IDF = 2, TF-IDF = 0.04
Q19. In a Named Entity Recognition (NER) system, which of the following sentences
has incorrectly tagged entities?
A) "Elon Musk is the CEO of Tesla." → (Person: Elon Musk, Organization: Tesla)
B) "Amazon is located in South America." → (Organization: Amazon, Location: South
America)
C) "Barack Obama was the 44th President of the USA." → (Person: Barack Obama,
Title: President, Location: USA)
D) "Apple launched a new iPhone model." → (Organization: Apple, Product: iPhone)
Answer: B) "Amazon is located in South America."
Q20. Consider the bigram model for the sentence:
"The quick brown fox jumps over the lazy dog."
Which of the following bigram probabilities requires smoothing to avoid zero
probability?
A) P(jumps | fox)
B) P(over | jumps)
C) P(dog | lazy)
D) P(quick | apple)
Answer: D) P(quick | apple)

Part-B
1. Explain the different applications of Natural Language Processing.
2. Discuss the various levels of NLP with suitable examples.
3. What are Regular Expressions? Explain their role in NLP with examples.
4. Compare and contrast stemming and lemmatization with examples.
5. Describe the process of tokenization in NLP.
6. Explain how Term Frequency (TF) and Inverse Document Frequency (IDF) are used for feature
extraction.
7. Discuss the concept of Named Entity Recognition (NER) and its importance in NLP.
8. Define and explain the role of Parts of Speech (POS) tagging in NLP.
9. What is N-grams? How are they used in text analysis?
10. Explain the need for smoothing in language modeling and describe any two smoothing
techniques.
Part-C
1. Explain the different stages of Natural Language Processing (NLP) with examples.
2. Describe the importance of feature extraction in NLP and how TF-IDF is used in text
classification.
3. Explain the process of Named Entity Recognition (NER) with suitable examples and its real-
world applications.
4. Discuss in detail the significance of Parts of Speech (POS) tagging and the different techniques
used for POS tagging.
5. What is morphological analysis? Explain different types of morphological processing with
examples.
6. How are N-grams used in NLP? Explain their advantages and disadvantages with real-world
applications.
7. Describe the role of smoothing in NLP.
8. Explain Laplace smoothing techniques with suitable example.
9. Discuss different types of text preprocessing techniques in NLP with appropriate examples.
10. Explain the working of TF-IDF in text representation and how it improves text classification.
11. What are the challenges in NLP, and how can they be addressed using modern techniques?