Scribd Logo
Upload

Welcome to Scribd!

  • 68 views

    Mid-Term: Answer Point Value: 1.0 Points Answer Key: D

    Uploaded by

    Bindu K R
    This document contains a mid-term exam with multiple choice questions testing knowledge of natural language processing concepts. It is divided into 4 parts, with questions ranging from 1 to 25 points assessing topics like word forms, lexical representation, part-of-speech tagging, n-gram language models, hidden Markov models, and maximum likelihood estimation. The document records student responses to questions on linguistic phenomena, NLP applications, tokenization, morphological rules, finite state transducers, word classes, derivational vs. inflectional morphology, challenges in Indian languages, edit distance, POS tagging, Naive Bayes, and perplexity.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Mid-Term: Answer Point Value: 1.0 Points Answer Key: D

    Uploaded by

    Bindu K R
    68 views12 pages
    This document contains a mid-term exam with multiple choice questions testing knowledge of natural language processing concepts. It is divided into 4 parts, with questions ranging from 1 to 25 points assessing topics like word forms, lexical representation, part-of-speech tagging, n-gram language models, hidden Markov models, and maximum likelihood estimation. The document records student responses to questions on linguistic phenomena, NLP applications, tokenization, morphological rules, finite state transducers, word classes, derivational vs. inflectional morphology, challenges in Indian languages, edit distance, POS tagging, Naive Bayes, and perplexity.

    Original Title

    Mid-Term202207224546

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document contains a mid-term exam with multiple choice questions testing knowledge of natural language processing concepts. It is divided into 4 parts, with questions ranging from 1 to 25 points assessing topics like word forms, lexical representation, part-of-speech tagging, n-gram language models, hidden Markov models, and maximum likelihood estimation. The document records student responses to questions on linguistic phenomena, NLP applications, tokenization, morphological rules, finite state transducers, word classes, derivational vs. inflectional morphology, challenges in Indian languages, edit distance, POS tagging, Naive Bayes, and perplexity.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    68 views12 pages

    Mid-Term: Answer Point Value: 1.0 Points Answer Key: D

    Uploaded by

    Bindu K R
    This document contains a mid-term exam with multiple choice questions testing knowledge of natural language processing concepts. It is divided into 4 parts, with questions ranging from 1 to 25 points assessing topics like word forms, lexical representation, part-of-speech tagging, n-gram language models, hidden Markov models, and maximum likelihood estimation. The document records student responses to questions on linguistic phenomena, NLP applications, tokenization, morphological rules, finite state transducers, word classes, derivational vs. inflectional morphology, challenges in Indian languages, edit distance, POS tagging, Naive Bayes, and perplexity.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 12

    Name: _________________________

    Score: ______ / ______

    Mid-Term

    Part 1: Part-1

    4
    The linguists phenomena in which the same word appears in different word forms
    (POS category) are called
    A. Hapology

    B. Heterograph

    C. Homology

    D. Homograph

    Answer Point Value: 1.0 points


    Answer Key: D

    11
    Which among the following is NOT coming under the scope of NLP application
    A. Automatic grammar correction

    B. Human to computer interaction

    C. Creating new artificial language

    D. Customer care chat-bots

    E. Identifying errors in computer program

    F. Extracting data from JSON file

    Answer Point Value: 1.0 points


    Answer Key: C,E,F
    7
    What would be the total number of tokens for the following sentence after word
    level tokenization and character level tokenization: "All good things come to an
    end."
    A. 8,25

    B. 8, 31

    C. 7, 25

    D. 7,31

    Answer Point Value: 1.0 points


    Answer Key: D

    10
    Which of the following can be used to effectively represent morphological
    rules[CO1,L2,1]
    A. Directed Acyclic graphs

    B. Finite state transducers

    C. Context free grammars

    D. If-else rule

    Answer Point Value: 1.0 points


    Answer Key: B

    8
    FST is [CO1,L2,1]
    A. Mealy machine

    B. Moore machine

    C. Push-down automaton

    D. Turing machine

    Answer Point Value: 1.0 points


    Answer Key: A
    2
    Which among the following will be the correct lexical representation of the
    surface form "dog" [CO1,L2,1]
    A. dog+V+Present

    B. dog+N+PL

    C. dog+N+SG

    D. dog+VB+SG

    Answer Point Value: 1.0 points


    Answer Key: C

    1
    Knowledge of the relationship of meaning to the goals and intentions of the
    speaker is [CO1,L2,1]
    A. Morphology

    B. Pragmatics

    C. Discourse

    D. Semantics

    Answer Point Value: 1.0 points


    Answer Key: C

    5
    The word "upcoming" is [CO1,L2,1]
    A. a derivational morphology and contains 3 morphemes

    B. both inflection and derivational morphology, contains 3 morphemes

    C. an inflectional morphology and contains 2 morphemes

    D. an inflectional morphology and contains 2 morphemes

    Answer Point Value: 1.0 points


    Answer Key: B

    6
    Which of the following is an example for closed word class
    A. Nouns

    B. Interrogative words

    C. Verb

    D. Proper nouns

    Answer Point Value: 1.0 points


    Answer Key: B
    3
    Computerize, playing, Computation are examples for
    A. Derivational, inflectional, inflectional

    B. Derivational, Derivational, inflectional

    C. Derivational, inflectional, Derivational

    D. Inflectional, Inflectional, Derivational

    Answer Point Value: 1.0 points


    Answer Key: A

    9
    What is the main challenge in morphological analysis of Indian languages
    [CO1,L2,1]
    A. Removing stopwords

    B. Multiple affixes

    C. Removing punctuations

    D. Sandhi split

    Answer Point Value: 1.0 points


    Answer Key: D
    Part 2: Part-2

    13
    Assume that the edit distance algorithm uses following costs: insertion-1,
    deletion-1 and substitution-2. Now calculate the distance between the strings
    "cookery" and "bakery", number of insertions, deletions and substitutions
    A. 5, 2, 1,1

    B. 5, 0, 1, 2

    C. 3, 1, 0, 1

    D. 3, 0, 1, 1

    Answer Point Value: 2.0 points


    Answer Key: B

    12
    Assume that the edit distance algorithm uses following costs: insertion-1,
    deletion-1 and substitution-1. Now calculate the distance between the strings
    "saturday" and "sunday", number of insertions, deletions and substitutions
    A. 3, 1, 1,1

    B. 3, 0, 2, 1

    C. 4, 2, 0, 1

    D. 5, 1, 2, 0

    Answer Point Value: 2.0 points


    Answer Key: B
    Part 3: Part-3

    18
    Consider the following setence: "Ram went to by fruits in the supermarket of the
    mall, while his younger brother went to play games in the gaming section of the
    mall. What is the probability of P(to|went)?
    A. 0.5

    B. 0

    C. 1

    D. 0.001

    Answer Point Value: 1.0 points


    Answer Key: C

    22
    Which among the following is correct in case of a bigram model [CO2, L2,2]
    A. P(A,B,C,D) = P(A)P(B|A)P(C|B,A)P(D|A,B,C)

    B. P(A,B,C,D) = P(A)P(B|A)P(C|B)P(D|C)

    C. P(A,B,C,D) = P(A)P(B)P(C)P(D)

    D. P(A,B,C,D) = P(A)P(B)P(C|A)P(D|B)

    Answer Point Value: 1.0 points


    Answer Key: B

    16
    If we are assuming that likelihood of particular tag follows a trigram model, which
    among the following is correct [CO2, L2, 2]
    A. P(ti|t1...tn) = P(ti|ti-1)

    B. P(ti|t1...ti-1) = P(ti|ti-1,t-2)

    C. P(ti|t1...tn)= P(ti|ti-1,ti-2)

    D. P(ti|ti...tn)=P(ti|ti-1)

    Answer Point Value: 1.0 points


    Answer Key: B
    15
    Consider the following statements and select the correct option: Assertion (A): It
    is not necessary to use the denominator (common for all classes) while
    computing probabilities using Naïve Bayes formula Reason (R): In Naïve Bayes, a
    reasonably accurate rank ordering of probability values is required for classifying
    a new record [CO2, L2, 1]
    A. A is incorrect and R is correct

    B. A is correct and R is incorrect

    C. Both A and R are correct but R is not the correct explanation of A

    D. Both A and R are correct and R is the correct explanation of A

    Answer Point Value: 1.0 points


    Answer Key: D

    20
    Which among the following is a rulebased POS tagger [CO2, L2, 1]
    A. HMM tagger

    B. EngCG tagger

    C. Pen Tagger

    D. Brill Tagger

    Answer Point Value: 1.0 points


    Answer Key: B

    21
    Perplexity is inversly propotional to probability [CO2, L2, 1]
    A. True

    B. False

    Answer Point Value: 1.0 points


    Answer Key: A
    19
    Consider the following corpus containig 3 sentences. (1) Alice went to the cafe
    (2)Bob was waiting for Alice (3) Alice and Bob went to the museum. Assume we
    are not applying any preprocessing and stopword removal. What is the total
    number of unique bigrams for which the likelihood will be estimated? [CO2, L2, 2]
    A. 17

    B. 15

    C. 18

    D. 16

    Answer Point Value: 1.0 points


    Answer Key: A

    14
    Viterbi algorithm uses [CO2, L2, 1]
    A. Dynamic Programming

    B. Greedy approach

    C. Divide and conquer

    D. Recursion

    Answer Point Value: 1.0 points


    Answer Key: A

    17
    Which among the following is an extrinsic measure for language model
    evaluation? [CO2, L2, 1]
    A. Perplexity

    B. Cross entropy

    C. Accuracy

    D. Probability

    Answer Point Value: 1.0 points


    Answer Key: C
    Part 4: Part-4

    24
    Imagine that you start reading a poem and you have already read 10 times "is", 3
    times "love", 2 times "end", 1 times "that", 1 times "kind" and 1 times "enough".
    Total 18 tokens are read so far. How likely is it when the next token is "kind" by
    using MLE? How likely is it when the next token is "hate" by MLE
    A. 1/18, 1/18

    B. 1/9, 0

    C. 1/18, 0

    D. 1/9, 1/6

    Answer Point Value: 2.0 points


    Answer Key: C
    25
    Suppose you are in room quarantine due to primary contact of a covid patient.
    You are locked in a room and you want to find out how the weather outside only
    based on how you feel. Assume a first order hidden markove model with the
    states {sunny, foggy, rainy} with their transition probabilities as given in the
    figure-1
    (https://drive.google.com/file/d/1TwrLVt6ZvXRDLagcpJWwf5uS7K7AYKIl/view?usp
    =sharing).
    You either feel happy or grumpy. The emission probabilities of how you feel given
    the weather outside are shown below  figure-2
    (https://drive.google.com/file/d/1Ew6_JuQG_8b1J-
    4cInxC1p8b5PD2w3yh/view?usp=sharing).
    Let the initial probabilities of the weather being sunny, foggy and rainy on day 1
    be 0.5,0.3 and 0.2 respectively.
    Suppose you feel grumpy on day-1, happy on day-2 and happy again on day-3.
    What is the probability of all three days being sunny? [CO2, L2, 2]

    Attachments
    A.
    0.02016

    B.
    0.07526

    C.
    0.10976

    D.
    0.04704

    Answer Point Value: 2.0 points


    Answer Key: A
    23
    Suppose you are reading an article on IPL cricket. Till now you have read the
    words "score", "cricket", and "India" 5 times each, "catch",and "wicket" thrice
    each, "points", "company", "ground" and "won" once each. What is the maximum
    likelyhood estimate and add-one smoothned probabilities of reading "batsman"
    as the next word? [CO2, L3, 2]
    A. 0/25, 1/35

    B. 4/25, 4/25

    C. 0/25,1/26

    D. 0/25, 0/25

    Answer Point Value: 2.0 points


    Answer Key: A

    You might also like