MCQs

Disclaimer:

Educational Use Only:

The website links provided are to be used strictly for educational and knowledge-gaining purposes only.

No Misuse:

Do not misuse the links or the information contained therein for any commercial, illegal, or unethical activities.

Respect for Websites and Businesses:

Respect the intellectual property and business interests of the websites and their owners. Do not attempt to alter, copy, or redistribute content without proper authorization.

Acknowledge Value of Knowledge:

Recognize and value the knowledge and resources shared by these websites. Use the information responsibly and ethically.

No Endorsement:

The provision of these links does not imply endorsement of the linked sites, their content, or their products and services. Users are responsible for evaluating the suitability and accuracy of the information.

User Responsibility:

By accessing these links, you agree to use the information at your own risk and to comply with all applicable laws and website terms of use.

Overall MCQs for Reference

What is an n-gram in NLP?
a. Sequence of n characters
b. Sequence of n words
c. Division of text by sentence
d. Unique identifier of words
Answer: b. Sequence of n wordsgeeksforgeeks+1
What is a unigram?
a. 1 unique document
b. Sequence of 1 word
c. Sequence of 2 words
d. Sequence of n words
Answer: b. Sequence of 1 wordtowardsdatascience+1
What is a bigram?
a. Sequence of 1 word
b. Sequence of 3 words
c. Sequence of 2 words
d. Sequence of n words
Answer: c. Sequence of 2 wordslunartech+1
In a trigram model, how many previous words are considered to predict the next word?
a. 1
b. 2
c. 3
d. n
Answer: b. 2geeksforgeeks+1
What’s the primary use of n-gram models?
a. Data encryption
b. Language modeling
c. Image recognition
d. Speech synthesis
Answer: b. Language modelinglunartech+1
Which application commonly uses n-gram models?
a. Machine translation
b. Face detection
c. Data normalization
d. Video compression
Answer: a. Machine translationgeeksforgeeks+1
What is the main weakness of high-order n-gram models?
a. Simple calculation
b. High memory and data requirements
c. Fast convergence
d. Poor readability
Answer: b. High memory and data requirementsstanford+1
What does an n-gram language model predict?
a. The next sentence
b. The next word
c. The previous paragraph
d. Image category
Answer: b. The next wordstanford+1
Which linguistic unit does a character-level n-gram model use?
a. Syllables
b. Sentences
c. Characters
d. Documents
Answer: c. Charactersstanford+1
Which of the following is NOT a use case for n-grams in NLP?
a. Spelling correction
b. Word prediction
c. Parsing diagrams
d. Sentence completion
Answer: c. Parsing diagramsscribd
N-grams consider which element for modeling language?
a. Only the middle word
b. Context of previous (n-1) words
c. Only punctuation
d. Topic of document
Answer: b. Context of previous (n-1) wordslunartech+1
What is a common limitation of n-gram models?
a. Too complex
b. Lack of long-range context
c. Requires visual data
d. Cannot handle numbers
Answer: b. Lack of long-range contextlunartech+1
What is the minimum value of n in an n-gram?
a. 0
b. 1
c. 2
d. Unlimited
Answer: b. 1geeksforgeeks
Computing the probability of a sentence using n-grams applies the:
a. Chain rule of probability
b. Bayes theorem
c. Gradient descent
d. Eigen decomposition
Answer: a. Chain rule of probabilityyoutube
Higher-order n-grams tend to:
a. Reduce fluency
b. Increase sparsity
c. Decrease accuracy
d. Ignore punctuation
Answer: b. Increase sparsitystanford+1
What is “smoothing” in n-gram models?
a. Removing punctuation
b. Adjusting probabilities for unseen n-grams
c. Adding noise
d. Data normalization
Answer: b. Adjusting probabilities for unseen n-gramslunartech+1
What is a practical application of bigrams?
a. Image segmentation
b. Keyword extraction
c. Audio compression
d. Data backup
Answer: b. Keyword extractiongeeksforgeeks+1
How are n-grams typically generated from text?
a. By random sampling
b. By sliding window technique
c. By decision trees
d. By clustering
Answer: b. By sliding window techniquelunartech
In what language modeling scenario are trigrams often preferred over bigrams?
a. When text is very short
b. When context is important
c. When building inverted index
d. When sequence is unordered
Answer: b. When context is importantgeeksforgeeks+1
Which of the following is an example of a trigram from the phrase “the quick brown fox”?
a. the quick
b. quick brown
c. the quick brown
d. brown fox
Answer: c. the quick browndev+1
Why do n-gram models use start- and end-of-sentence tokens?
a. For better grammar correction
b. To appropriately model word transitions at sentence boundaries
c. For spell checking only
d. For sentiment analysis
Answer: b. To appropriately model word transitions at sentence boundariesstanford+1
What happens to data requirements when increasing the value of n in n-gram models?
a. Data requirement decreases
b. Data requirement stays the same
c. Data requirement increases
d. Data is not required
Answer: c. Data requirement increasesstanford+1
In the sentence “AI is awesome”, what is a possible bigram?
a. is awesome
b. AI
c. awesome is
d. AI awesome
Answer: a. is awesometowardsdatascience+1
Which of the following is a use case for character n-grams?
a. Language identification
b. Video summarization
c. Neural network training
d. Hardware design
Answer: a. Language identificationgeeksforgeeks+1
How does n-gram size affect language model generalization?
a. Larger n generalizes better always
b. Very large n overfits to corpus
c. Smaller n always fails
d. Size does not matter
Answer: b. Very large n overfits to corpuslunartech+1
What is a major advantage of using unigrams for modeling?
a. Captures long dependencies
b. Simple and needs less data
c. Guarantees perfect prediction
d. Requires deep neural networks
Answer: b. Simple and needs less datastanford
Bigrams can help in:
a. Capturing word order
b. Ignoring context
c. Designing networks
d. Plotting graphs
Answer: a. Capturing word orderdev+1
In which field are n-gram models a fundamental baseline?
a. NLP
b. Robotics
c. Signal processing
d. Computer vision
Answer: a. NLPgeeksforgeeks+1
What is the major drawback of unigram models?
a. Poor performance with common words
b. Ignores context
c. Overfits the training set
d. Complex structure
Answer: b. Ignores contextstanford
What is “perplexity” in n-gram models?
a. Length of document
b. Measure of probability distribution prediction quality
c. Count of unique tokens
d. Number of errors
Answer: b. Measure of probability distribution prediction qualitystanford
What is the effect of Zipf’s law on n-gram frequency distributions?
a. Uniform distribution
b. Imposes a power law on word frequencies
c. Guarantees equal n-gram probabilities
d. No known effect
Answer: b. Imposes a power law on word frequenciesstanford
Which smoothing technique adds one to seen and unseen events?
a. Add-one (Laplace) smoothing
b. Backoff smoothing
c. Interpolation
d. Discounting
Answer: a. Add-one (Laplace) smoothingstanford
Why is “backoff” used in n-gram models?
a. It ignores probabilities
b. To use lower-order models when higher-order n-gram is not found
c. To increase training data
d. To reduce output vocabulary
Answer: b. To use lower-order models when higher-order n-gram is not foundstanford
The sparsity problem in n-gram models refers to:
a. Too many zeros in the n-gram counts
b. Too many sentences
c. Lack of punctuation
d. Oversampling
Answer: a. Too many zeros in the n-gram countslunartech+1
What’s one way to handle rare and unknown words in n-gram models?
a. Ignore them
b. Replace them with a special token “<UNK>”
c. Randomly shuffle words
d. Substitute with numbers
Answer: b. Replace them with a special token “<UNK>”lunartech
Which is usually better for sentence completion: bigram or unigram models?
a. Bigram
b. Unigram
c. Neither
d. Not applicable
Answer: a. Bigramgeeksforgeeks+1
How does the Markov assumption simplify n-gram models?
a. Considers only the next word
b. Considers only immediate history (n-1 previous words)
c. Considers all words in history
d. Ignores probability
Answer: b. Considers only immediate history (n-1 previous words)geeksforgeeks+1
Which task is often improved by using n-gram features?
a. Sentiment classification
b. Sorting numbers
c. File compression
d. Hardware acceleration
Answer: a. Sentiment classificationtowardsdatascience+1
Probabilities in n-gram models are estimated by:
a. Relative frequency of n-grams
b. Random sampling
c. Decision trees
d. Graph traversal
Answer: a. Relative frequency of n-gramslunartech+1
Which of the following would increase the computational cost of an n-gram model?
a. Lower n
b. Higher n
c. Ignoring tokens
d. Preprocessing
Answer: b. Higher nlunartech+1
What is a sentence boundary token used for?
a. To join words
b. To denote start and end of sentences
c. To count n-grams
d. To create punctuation
Answer: b. To denote start and end of sentenceslunartech+1
Which metric is often used to evaluate n-gram models?
a. Cross-entropy
b. Mean absolute error
c. RMSE
d. F1 score
Answer: a. Cross-entropystanford
How does the training corpus affect n-gram models?
a. More data leads to better n-gram estimation
b. Data has no effect
c. Too much data reduces accuracy
d. Data only affects output format
Answer: a. More data leads to better n-gram estimationlunartech+1
In a character-level 3-gram, how are n-grams formed from “machine”?
a. mac, ach, chi, hin, ine
b. ma, ch, in
c. m, a, c
d. mach, chin
Answer: a. mac, ach, chi, hin, inegeeksforgeeks+1
What is interpolation in the context of n-gram modeling?
a. Combining lower-order and higher-order probabilities
b. Ignoring missing data
c. Adding start and stop tokens
d. Sorting unigrams
Answer: a. Combining lower-order and higher-order probabilitiesstanford
Which of the following n-gram models captures the most context?
a. Unigram
b. Bigram
c. Trigram
d. 4-gram
Answer: d. 4-gramlunartech+1
How is an n-gram with n=1 related to the Bag of Words model?
a. Equivalent
b. Opposite
c. Complex
d. Nonlinear
Answer: a. Equivalentstanford
What is the consequence of having too large an n in n-gram models?
a. Overfitting and increased sparsity
b. Better generalization
c. Lower computation
d. Uniform probability
Answer: a. Overfitting and increased sparsitylunartech+1
In NLP, what does “tokenization” refer to?
a. Breaking text into sentences or words (tokens)
b. Creating new words
c. Training a neural network
d. Encrypting data
Answer: a. Breaking text into sentences or words (tokens)geeksforgeeks+1
Which modern method supersedes n-gram models in current state-of-the-art NLP?
a. Rule-based methods
b. Neural language models
c. Statistical charts
d. Syntax trees
Answer: b. Neural language modelsdev+1

Blog

Systemethic Technologies

Leave a Reply Cancel reply

Useful links

Other Terms & Policies

Contact Info