INSTITUTE OF AERONAUTICAL ENGINEERING

(Autonomous)
Dundigal, Hyderabad - 500 043

COMPUTER SCIENCE AND ENGINEERING(AI & ML)

QUESTION BANK

Department COMPUTER SCIENCE AND ENGINEERING(AI & ML)

Course Title NATURAL LANGUAGE PROCESSING
Course Code ACAC13
Program B.Tech
Semester VI CSE(AI & ML)
Course Type Core
Regulation IARE - UG20
Theory Practical
Course Structure Lecture Tutorials Credits Laboratory Credits
3 - 3 - -
Course Coordinator Dr. M.Nagaraju, Assistant Professor

COURSE OBJECTIVES:
The students will try to learn:

I The concepts and techniques of Natural language Processing for analyzing words
based on Morphology and CORPUS
II The mathematical foundations, Probability theory with Linguistic essentials such
as syntactic andsemantic analysis of text.
III The applications of statistical learning methods and cutting-edge research models
from deep learning.

COURSE OUTCOMES:
After successful completion of the course, students should be able to:
CO 1 Remember the knowledge of complex language behaviour in terms Remember
of phonetics, morphology etc
CO 2 Understand the semantics and pragmatics for text processing Understand
CO 3 Apply the CORPUS linguistics to compile and analyze the texts Apply
based on digestive approach (Text Corpus Method).
CO 4 Understand various statistical approaches to machine translation Understand
for a given natural language.
CO 5 Apply Part-of-speech (POS) tagging for a given natural language Apply
and suitable modeling technique based on the structure.
 CO 6 Apply the state of the art algorithms and techniques for text-based Apply
processing of natural language with respect to morphology

QUESTION BANK:
Q.No QUESTION Taxonomy
How does this subsume CO’s
the level
MODULE I
CLASSICAL APPROACH TO NLP
PART A-PROBLEM SOLVING AND CRITICAL THINKING QUESTIONS
1 How can NLP be used to Understand The learner can Recall how CO 1
enhance educational and NLP can be used in solving
research applications, and linguistic problems and
what are some potential Understand how it change
uses and limitations? human living.
2 Explain any three use Understand The learner will Recall how CO 1
cases of NLP in respective NLP can solve real time
to its applications? problems and Understand
the complex types of use
cases
3 What are the stages in the Understand The learner will Recall how CO 1
lifecycle of a natural NLP pipeline can explain
language processing different stages of NLP and
(NLP) project? Understand how nlp can
solve NLP projects.
4 How can NLP be used to Remember The learner can able to Find CO 1
improve communication how NLP is used to
between humans and communicate with machines.
machines, and what are
some of the potential
benefits and drawbacks of
this technology?
5 Explain about various Understand The learner can able to CO 1
components of Recall how to conversational
conversational agents and agents are helpful and
how they work? Understand how it solves
the problem of
communication systems.
6 What is meant by data Understand The learner can able to CO 1
augmentation? What are Recall the process of
some of the ways in which augmentatio and
data augmentation can be Understand the different
done in NLP projects? types of augmentations.

Page 2
7 How can data be obtained Remember The learner can Recall how CO 1
for NLP projects? NLP projects can perform
the classification based on
the features of its data points
and the bank application.
8 How can NLP algorithms Remember The learner will Recall how CO 1
be optimized to improve NLP used in real time
accuracy and reduce bias? applications and how they
can be used to improve
projects efficiency.
9 How can NLP be used to Apply The learner will Recall the CO 1
analyze large amounts of concerned of thinking of NLP
text data, and what are systems and Understand
some best practices for the application in the NLP
doing so effectively? system.
10 Create a text classification Apply The learner can Recall CO 1
system using the publicly several issues that come
available dataset and across while discussion
explain the step by step specific mechanisms to
procedure? classify knowledge and
Understand the real-world
applications.
PART-B LONG ANSWER QUESTIONS
1 Justify the statement Apply The learner can Recall the CO 1
”Ambiguity is the knowledge about the
capability of desirable properties and
understanding in more Understand how that
than one way” with ambiguity can be removed.
suitable examples?
2 What is ambiguity? Apply The learner can Recall CO 1
Explain different types of different types of ambiguities
ambiguities with Understand the various
examples? characteristics of NLP and
Apply in generating
matching sentences.
3 Explain how NLP models Understand The learner can Recall the CO 1
and algorithms can resolve detail knowledge about the
the disambiguation ? models of NLP and
Understand how that are
useful in solving complex
problems.

Page 3
4 What is state machine? Understand The learner will Recall CO 1
Explain different exploring various state
variations of state machines and Understand
machines? the different types of NLP
state machines.
5 What is Formal Understand The learner can Recall the CO 1
Rule-Based Systems? benefits of Rule-based
Explain the advantage of systems and Understand
using probabilitic models the implementing them in
in solving the problems of solving ambiguity problems.
ambiguity?
6 Explain in detail how the Understand The learner can Recall the CO 1
turing test is conducted importance of turing test and
and also the need of the Understand how that can
test? improve the machines
capability.
7 What is the need of Understand The learner will Recall how CO 1
knowledge in speech and to examine the need of
language processing? language processing and
Understand how it is
necessary to solve the
problem.
8 What is the need of Understand The learner will Recall the CO 1
understanding linguistics information about the
in NLP? linguistics and Understand
the frame language problems.
9 Explain in detail about Understand the learner can Recall the CO 1
any five models and need of NLP models and
algorithms that are the Understand some
key insights of NLP? mechanisms for implementing
the algorithms to create
solutions to new problems.
10 Explain how AI can play a Understand The learner can Recall four CO 1
role in sentiment analysis different examples of
of human beings? language analysis and
Understand the various
language analysis schemes.
11 Why are human languages Understand The learner will Recall the CO 1
complicated for a importance of human
computer to understand? language understanding and
Explain. Understand how to
describe the basic mechanism
of machine understanding.

Page 4
12 Differentiate between Understand The learner will CO 1
Human Language and Recalldifferent computer
Computer Language? languages and Understand
the differences between the
languages.
13 Mention some applications Understand The learner can Recall the CO 1
of Natural Language importance of NLP and
Processing and explain Understand how that can
each? support variety of things in
NLP applications.
14 What is the significance of Understand The learner can Recall the CO 1
converting the text into a significance of NLP and
common case? Understand how to perform
the text conversion.
15 Explain how NLP is useful Understand The learner can Recall to CO 1
in Customer Support know various methods of
System? representing NLP and
Understand how to use it
in real time applications.
16 Explain the different Understand The learner will Recall how CO 1
overlaping fields of the NLP-based techniques
Natural Language are differentiated for the
Processing? problems and Understand
the techniques are used to
resolve the problems.
17 How speech and language Understand The learner will Recall the CO 1
processing can consider to speech and language
create an intelligent agent processing and Understand
like HAL? how to solve the NLP
problems with respect to
intelligent agents.
18 List any two real-life Understand The learner will Recall the CO 1
applications of Natural benefits of NLP,
Language Processing? Understand the real-time
applications and to solve
language understanding
problems
19 What is NLTK and how it Understand The learner will Recall the CO 1
is useful in processing benefits while using NLP
human language? toolkits, Understand how
to solve complex problems
using NLP and solve human
language processing

Page 5
20 Explain in detail all the Understand The learner can Recall how CO1
components of NLP algorithms can be developed
Pipeline? for a specific NLP problem,
Understand the various
components and solve the
text processing problems.

Page 6
 PART-C SHORT ANSWER QUESTIONS
1 What is Natural Language Understand The learner can Recall how CO 1
Processing and explain its NLP makes computers to
various applications? understand and
Understand how it change
various applications.
2 How the regular definition Understand The learner will Recall how CO 1
of NLP is used? NLU performs creative and
Understand the complex
types of cognitive activities
such as translation, writing
texts etc.
3 What is the difference Understand The learner will Recall how CO 1
between Natural Language different NLP and NLU are,
Processing and Natural and Understand how
Language Understanding? especially it is different in the
case of chatbots.
4 How Natural Language Remember The learner can able to Find CO 1
Understanding and how the NLP can be used to
Natural Language generate communication.
Generation are differ?
5 Natural Language Understand The learner can able to CO 1
Processing is all about a Recall how to know the
new interdisciplinary field. process of determining
Justify the statement? whether the computer is so
efficient and Understand
how it exhibits the behaviour
of a human.
6 What is a Conversational Understand The learner can able to CO 1
Agent and how it is useful Recall the need to have an
in humans and machine interaction to accomplish a
communications? goal and Understand the
key capabilities.
7 What do you mean by Understand The learner can Recall how CO 1
web search and explain AI algorithms can perform
Web-based Questioning the questioning systems
Answering System? based on the human queries
and Understand the
web-based application.

Page 7
8 What is a Turing Test and Understand The learner will Recall how CO 1
why it is suggested as an the representation of
empirical test of knowledge to specific problem
determing compputers statements and Understand
ability? how they can be framed and
apply turing test.
9 Explain Cross-Language Understand The learner will Recall the CO 1
Information Retrieval and concerned of thinking of
Translation Services? conversational agents while
retrieving the knowledge and
Understand the application
in the NLP system.
10 Explain Text Analysis in Understand The learner can Recall CO 1
Natural Language several issues that come
Processing? across while discussion
specific mechanisms to
represent knowledge and
Understand the real-world
applications.
11 What is the difference Understand The learner will Recall the CO 1
between syntatic and various components of
semantic analysis? production system including
the descriptions and
Understand various
NLP-based problems.
12 Explain the term Understand The learner can Recall the CO 1
Computational Linguistics need of compuational
in linguistics? linguistics and Understand
its importance.
13 What is speech Understand The learner will Recall how CO 1
recognition and why it is language processing is
useful in natural language necessary and Understand
processing? when there is no longer
guaranteed to find the best
answer and finds a very good
answer.
14 Explain in detail the Understand The learner can able to CO 1
functionality of the Recall how HAL maps the
conversational agent problem state descriptions
HAL? and Understand how to
measure desirability using
conversational agents.

Page 8
15 What are the well known Understand The learner first Recall the CO 1
algorithms for handling how the problem of
non-probabilistic tasks? computing the expression is
broken into smaller problem,
and Understand how each
of which can then solve by
using a small collection of
specific rules.
16 Explain the difference Understand The learner will able to CO 1
between regular grammer Recall one class of grammer
and context free grammer? and Understand the
differences between context
free and regular grammer.
17 Define the term dynamic Understand The learner will able to CO 1
programming in NLP? Recall one class of
production system and
Understand the different
NLP programs.
18 How probabilitic models Understand The learner will able to CO 1
are useful in natural Recall one class of
language processing? probability models and
Understand how they are
different with other
processing systems
19 What do you mean by Understand The learner will able to CO 1
word sence Recall one class of
disambiguation? disambiguation and
Understand the differences
between several
disambiguations.
20 Why there is a ambiguity Understand The learner get to Recall CO 1
in understanding natural the different ambiguities and
language? Understand through the
NLP.

Page 9
 MODULE II
REGULAR EXPRESSIONS, TEXT NORMALIZATION
PART-A PROBLEM SOLVING AND CRITICAL THINKING QUESTIONS
1 You created a document Apply The learner to Recall how CO 2
term matrix on the input the models can be used and
data of 20K documents for Understand that are used
a Machine learning model. to solve real-time
Which of the following can applications.
be used to reduce the
dimensions of data?
Explain with an example.
2 How can regular Apply The learner able to Recall CO 2
expressions be used to how specific information
extract specific types of processing techniques are
information from text used and Understand how
data, and what are some they are Applied to make
best practices for doing so best practices effectively.
effectively?
3 How can regular Apply The learner will Recall how CO 2
expressions be optimized the RE is more powerful in
to improve accuracy and representing the search
reduce errors when patterns and Understand
processing large amounts and Apply the basics of text
of text data? processing.
4 How do regular Apply The learner will Recall how CO 2
expressions differ from to prove that the above
other NLP techniques, statement is true in terms of
such as machine learning possible world and
algorithms, and what are Understand and Apply the
some situations where one NLP techniques.
approach may be more
effective than the other?
5 How can regular Apply The learner must Recall the CO 2
expressions be used to certainity in the RE model
identify patterns or trends and Understand and
in large datasets, and Apply the possibilities while
what are some of the extracting some tokens.
potential applications of
this technology in fields
such as marketing or
finance?

Page 10
6 How can regular Apply The learner must Recall the CO 2
expressions be used to concepts of data
enhance the user communication strategies and
experience in applications Understand and Apply the
such as search engines or concept of chatbots
chatbots?
7 How can text Apply The learner will able to CO 2
normalization be used to Recall how to apply the
improve the accuracy of concept of normalization,
NLP models, particularly Understand the problem
in tasks such as sentiment statement while representing
analysis or named entity the NLP models and Apply
recognition? to solve NLP models.
8 How do different Apply The learn will Recall how to CO 2
languages or dialects apply the concept of text
impact the process of text normalization, Understand
normalization, and what how to use the same while
are some best practices for best practices and Apply to
dealing with these process text.
differences?
9 How can text Apply The learner will Recall how CO 2
normalization be used to to improve the ML
improve the performance performances, Understand
of machine learning how they used measure the
algorithms, and what are possibilities to prove the
some potential efficiency and Apply to solve
applications of this real world problem .
technology in fields such
as healthcare or finance?
10 How can text Apply The learner will Recall how CO 2
normalization be used to to improve the accessibility,
enhance the accessibility Understand the text
of digital content for normalization and Apply to
individuals with solve the probability of
disabilities, such as those accessibility.
who are visually impaired
or have dyslexia?
PART-B LONG ANSWER QUESTIONS
1 Explain the UNIX tools Understand The learner can Recall the CO 2
for word tokenization and knowledge of decision tree
normalization of English? and Understand about the
text normalization and
tokenization

Page 11
2 What is word Understand The learner will Recall the CO 2
tokenization? Explain the preprocessing steps and
need of considering Understand the difference
punctuations as a seperate between the word and
token during NLP sentence tokenization
applications?
3 Explain the need of Understand The learner can Recall the CO 2
Byte-Pair Encoding for difference between these
tokenization? terms and Understand the
basic principles of encoding.
4 Explain different Apply The learner can Recall the CO 2
tokenization schemes in knowledge about the
detail with suitable representation schemes and
example? Understand how the same
knowledge and Apply in
various tokenization schemes.
5 Explain the Understand The learner will Recall to CO 2
implementation of know about the algorithms
Byte-Pair Encoding related to feature selection
Algorithm? and Understand the
knowledge representation
using the algorithm
6 Explain in detail about Understand The learner will Recall to CO 2
different tokenization know different methods and
schemes in NLP? Understand the NLP
schemes.
7 What is Minimum Edit Understand The learner can Recall the CO 2
Distance and how it basic concepts of text
constitutes the similarity preprocessing and
about the strings? Understand to calculate
minimum edit distance
among the words.
8 Explain how dynamic Understand The learner will Recall to CO 2
programming can be used know the process which gains
to compute the shortest efficiency from the fact that
path to each state of the it operates on statements and
string conversion? Understand how they have
been converted to a very
convenient standard form.

Page 12
9 Explain Minimum Edit Understand The learner can Recall the CO 2
Distance algorithm in basic concepts of text
detail using the words preprocessing and
”intention” and Understand to calculate
”execution”? minimum edit distance
among the words.
10 Explain the term Finite Understand The learner will Recall know CO 2
State Automata in terms that the procedure is an
of NLP? iterative one and
Understand how to perform
FSA in NLP.
11 Explain regular language Understand The learner will Find to CO 2
and FSA? know the procedure for
producing a proof by
resolution of RE with respect
to a FSA.
12 Define regular languages Understand The learner will Find the CO 2
with different operations straight forward procedure
in detail? that unifies two literals.
13 Explain the process of raw Remember The learner can Find the CO 2
text extraction from a concepts of determing the
document? conflicts and how they can be
unified with the text
extraction procedure.
14 Explain the process of text Understand The learner can Find and CO 2
normalization in detail Understand the how the
with suitable examples? text processing is represented
in two different ways.
15 Explain the process of Understand The learner can Recall CO 2
extracting terms form about the process extraction
tokens? and Understand how they
are represented in natural
Language processing.
16 Define the terms words Remember The learner will Find how CO 2
and patterns with suitable word patterns can be used to
example? solve search problems.
17 Explain the process of text Remember The learner will Find how CO 2
categorization in NLP? categorization can be used to
process the text.
18 Explain how NLTK Understand The learner get to Recall CO 2
supports in doing different how NLTP is used to perform
kinds of text processing in text processing and
NLP? Understand how it
supports NLP tasks.

Page 13
19 Explain the need of Understand The learner will Recall the CO 2
regular expressions in detail information about the
detail with suitable result of regular expressions
examples? and Understand how the
order in which the rules will
be applied is decided.
20 Explain various Understand The learner will Recall to CO 2
applications of regular know that which approach is
expressions in NLP used to solve a problem and
Understand the differences
between various applications.
PART-C SHORT ANSWER QUESTIONS
1 What is a Corpus and Remember The learner will Find various CO 2
explain about Brown standard sentences used in
University Corpus? representing the corpus.
2 Explain different types of Understand The learner Recall and CO 2
datasets with example? Understand how the
real-world sentences are
represented as datasets.
3 What is the basic need of Understand The learner is able to Recall CO 2
NLP? about the NLP need and
Understand how it is
appealing in suggesting a
powerful way of
communication with humans.
4 What is a regular Remember The learner will Find how CO 2
expression and explain the RE is used to represent
some common uses of patterms and understand the
regular expression? need of search.
5 What is tokenization in Remember The learner will Recall the CO 2
NLP? knowledge about the
tokenization in a class to
which the words belongs to
the text representations.
6 What is lemmatization in Remember The learner will Recall the CO 2
NLP? lemmatization procedure and
how it gains efficiency from
its applications.
7 What is the difference Remember The learner will Find to CO 2
between lemmatization know that individual purpose
and stemming? of NLP languages and the
process of text processing.

Page 14
8 Explain the need of using Undesrtand The learner can Recall the CO 2
Edit Distance? need of edit distance and
Understand the process of
finding the difference among
the words.
9 What is the difference Remember The learner can Find to CO 2
between tokenization and know how two tehniques can
stemming? be unified and Understand
the preprocessing process.
10 How regular expressions Understand The learner can Recall CO 2
can be used to specify unification with the negation
strings while extracting of the other and
from a document? Understand how it can be
used to determine the
contradictory among two
strings.
11 What is CORPUS and list Remember The learner will Recall how CO 2
out some existing corpus the NLP techniques can be
in the field of NLP? applied and Understand
how they are useful while
answering the questions.
12 Explain the basic need of Understand The learner will Recall how CO 2
regular expressions while the knowledge is specified
performing certain search and Understand to know
activities? what is to be done to the
search activities.
13 Explain with suitable Understand The learner can Recall how CO 2
examples how regular the RE is needed and
expressions can be used to Understand to use the
extract some range of characters extraction.
characters with examples
?
14 Explain the symbols * and Understand The learner will Recall how CO 2
+ in regular expressions RE can be used and
with suitable examples? Understand to produce
flexible structures for NLP
systems.
15 Explain the terms Understand The learner can Recall that CO 2
disjunction, grouping, and it is a language paradigm and
precedence? Understand the logical
assertions are viewed as
regular expressions.

Page 15
16 Explain the terms Understand The learner can Recall how CO 2
utterance, disfluencies, statements can be
and lemma? represented using
lemmatization and
Understand the differences
between tokenization and
lemmatization.
17 Explain Herdan’s Law or Undesrtand The learner will Recall that CO 2
Heap’s Law? logic programming needs
only specification of rules and
Understand that a law is
built direcly into the
language.
18 What is a dataset and Understand The learner get to Recall CO 2
explain different how datasets is applied to
properties of dataset? solve a particular problem
and Understand the
forward text processing
process.
19 Explain Text Apply The learner get to Recall CO 2
Normalization with an how tokenization is applied
example? to solve a particular problem
and Understand and
Apply the text
preprocessing process.
20 Explain the need of Understand The learner will Find the CO 3
normalizing word formats? knowledge when to employ
this strategy and
Understand how to choose
among the applicable rules.

Page 16
 MODULE III
N-GRAM LANGUAGE MODELS
PART A-PROBLEM SOLVING AND CRITICAL THINKING QUESTIONS
1 Explain in detail the text Apply The learner will be able to CO 3
parsing techniques can be Recall the process of parsing
used for noun phrase the techniques, Understand
detection, verb phrase which is possible in search
detection, subject and Apply to perform
detection, and object detection process.
detection in NLP.
2 In a corpus of N Apply The learner can Recall, CO 3
documents, one randomly Understand the concepts
chosen document contains related to TF and Apply to
a total of T terms and the solve k times
term “hello” appears K
times. What is the correct
value for the product of
TF (term frequency) and
IDF (inverse-document-
frequency), if the term
“hello” appears in
approximately one-third of
the total documents?
3 How can n-gram language Understand The leraner will Recall how CO 3
models be used to improve the text applications is
the accuracy of NLP carried out and Understand
applications, such as how it is useful during NLP
speech recognition or problem solving.
machine translation?
4 How do different n-gram Apply The learner will Recall the CO 3
sizes impact the step by step procedure of
performance of language implementing algorithms,
models, and what are Understand how to
some best practices for Applyimplement different
choosing an appropriate n-grams.
n-gram size?
5 How do n-gram language Apply The learner will Find, CO 3
models differ from other Understand and Applythe
NLP techniques, such as concepts of various NLP
deep learning algorithms, techniques
and what are some
situations where one
approach may be more
effective than the other?

Page 17
6 How can n-gram language Apply The learner will Find, CO 3
models be used to improve Understand and Apply the
the performance of concepts of chatbots
chatbots or virtual applications.
assistants, and what are
some potential
applications of this
technology in other areas
of industry or society?
7 How can we use n-gram Apply The learner will Recall the CO 3
language models to step by step process of
improve the performance solving language issues,
of machine translation Understand the procedure
systems, and what are to Apply n-gram language
some potential challenges models.
that may arise when
working with multiple
languages?
8 How can we use n-gram Apply The learner will Recall how CO 3
language models to detect this technique is used to find
and correct spelling or the language models by
grammar errors in text correcting spellings,
data, and what are some Understand the limit and
potential applications of Apply until the defined goal
this technology in fields is reached.
such as education or
publishing?
9 How can we use n-gram Apply The learner will Recall how CO 3
language models to n-gram language models,
identify patterns or trends Understand the search
in large datasets, and process and Apply during
what are some potential various real-time
applications of this applications.
technology in fields such
as marketing or finance?
10 How can we use n-gram Apply The learner will able and CO 3
language models to Recall how to implement
enhance the user n-gram language models,
experience in applications Understand how the
such as search engines or algorithm performs the
chatbots, and what are search engines and Apply to
some best practices for improve human
doing so effectively? enhancement.

Page 18
 PART-B LONG ANSWER QUESTIONS
1 What is N-grams and how Apply The Learner will Recall the CO 3
it is useful in techniques that are able to
understanding the overcome the problem of
probability of the next understanding next word,
word to appear with an Understand which word
example? processes are so suitable to
Apply, in NLP related
problems.
2 How can we compute the Apply The learner can Recall how CO 3
probabilites of the entire to generate a probability,
sequence of words using Understand how to Apply
chain rule of prabability? text processing.
3 How do we estimate these Apply The learner will Recall and CO 3
bigram or n-gram Understand the information
probabilities. Explain about the n-grams, Apply
with an example? how that uses probabilities.
4 How different types of Apply The learner will Recall how CO 3
evaluation types can be to the procedure is used to help
used to evaluate the the generator decide and
performance of the Understand which direction
language models. Explain to move in the evaluation,
in detail? Apply language models.
5 Explain the concept of Apply The learner will Recall the CO 3
generalization and zeros in steps involved in defining a
terms of n-gram models solution of a NLP problem,
with suitable examples? Understand the program
model design, Apply in
various problem solving in
the best design approaches.
6 What is smoothing and Remember The learner will Find the CO 3
explain different types of smoothing techniques that
smoothing using represents the complete
mathematical notations mathematical solutions.
and examples?
7 What is kneser-key Remember The learner canFind the CO 3
smoothing and explain different types of a
how it is the best n-gram smoothing technique.
smoothing method in
detail ?

Page 19
8 Explain about huge Apply The learner will Find the CO 3
language models and difference between the
stupid backoff using language models.
Google Web corpus
example?
9 Explain how stupid Remember The learner get to Find the CO 3
backoff algorithm can how the approach is best in
make the language model the probability distribution.
a true probability
distribution?
10 Explain how perplexity is Apply The learner get to Recall CO 3
used to evaluate the that it is a best evaluation
performance of the n-gram process and Understand
models relation to and Apply the performance
entropy? of the models.
11 What is morphology Remember The learner will Recall that CO 3
explain how it is used to it is a morphological unit and
define the minimal bearing the others properties that
unit in a language? comes in language processing.
12 Explain inflectional Apply The learner will able to CO 3
morphology using some Recall that it is a special
examples of english nouns kind of morphology and
and verbs? Understand and Apply the
difference between POS
tagging.
13 Explain derivational Apply The learner can able to CO 3
morphology in terms of Recall the process of
normalization process deviational morphology and
with suitable examples? Understand and Apply the
different directions in the
process of normalization.
14 Describe finite-state Remember The learner will able to Find CO 3
morphological parsing how how FSM process used to
it is used to proceed with parse the text.
the problem of parsing
morphology?
15 Explain the complete Understand The learner will Recall how CO 3
procedure to build a FSL performs the text
finite-state lexicon with processing in a better way,
suitable example? Understand how it is better
than the other language
models.

Page 20
16 What are finite-state Understand The learner get to Recall CO 3
transducers how it is used that it is the process of
to represent the establishing the
morphotactic structure of morphological parsing and
a lexicon with an Understand the FST.
example?
17 Explain the task of Understand The learner get to Recall CO 3
morphological parsing that it is the process of
with respect to FST using establishing the
some examples? morphological parsing and
Understand the FST.
18 How can we combine FST Apply The learner to Recall that it CO 3
lexicon and rules for is a series of actions that are
language parsing and responsible for a particular
generating with suitable outcome Understand and
examples? Apply the credit
assignments in text
processing.
19 Explain the problem of Understand The learner get to Recall CO 3
segmentation in terms of that it is a process of
word and sentence segmenting and
tokenization? Understand how it is useful
in text preprocessing
20 Explain the process of Understand The learner get to Recall CO 3
detecting and correcting that it is a spelling procedure
spelling errors in modern used to handle mistakes and
word processors and Understand the procedure.
search engines?
PART-C SHORT ANSWER QUESTIONS
1 Why probability is Remember The Learner will Find the CO 3
essential in speech techniques that are able to
recognition? solve speech recognition.
2 Why probability is Remember The learner will Find how CO 3
essential in machine the algorithm can generate a
translation? possible solution in the
machine translation.
3 Explain language models Remember The learner will Recall the CO 3
in terms of n-grams? information about the
planning process,
Understand how that uses
various language models,
Apply n-grams and
Analyze the algorithms
performance.

Page 21
4 How N-grams allows to Apply The learner get to Recall CO 3
predict the next word for how the procedure is used to
a given sequence in a help to predict the sentence
sentence? sequence Understand and
Apply the procedure
followed in n-grams.
5 Explain bigram with a Apply The learner get to Recall CO 3
suitable example? the steps involved in defining
a solution Understand and
Apply a NLP problem.
6 What is Markov Remember The learner will get to Find CO 3
assumption in n-grams? the the states represents the
complete potential solutions.
7 What is the purpose od Understand The learner can Recall CO 3
using Maximum likelihood about different states of a
estimation in NLP? likehood process and
Understand the NLP
estimation.
8 What is relative frequency Remember The learner will Find the CO 3
and how it is used in difference between the
n-gram probability n-grams efficiency.
estimation?
9 What is extrinsic Understand The learner get to Recall CO 3
evaluation and how it is the how the approach is best
used to evaluate the in evaluation process and
language models? Understand the concept of
NLP agents.
10 What is the other variant Understand The learner get to Recall CO 3
used to evaluate the that it is a state that is
language models? better than all its language
models and Understand the
procedure to solve other
variants of language
processing.
11 What is a branching factor Understand The learner can Recall that CO 3
and how do you think in it is a flat area of the
terms of perplexity? branching factor and
Understand the whole set
of perplexity.
12 What is the problem of Understand The learner will Recall how CO 3
sparsity describes about? to learn that it is a special
kind of sparsity and
Understand the describe
the concept.

Page 22
13 What is smoothing and Understand
The learner can able to CO 3
explain different types of Recall it is a process of
smoothing? moving to some earlier node
and Understand the
process of going in a different
direction.
14 Describe Kneser-Key Remember The learner will Find to CO 3
Smoothing with an know how smoothing is used
example? to smoothing technique.
15 What is entropy and how Remember The learner will Find how CO 3
it helps in evaluating n-grams algorithm performs
n-gram models? the search process in a better
way that evaluates n-gram
models.
16 How could you define Understand The learner get to Recall CO 3
entropy rate in n-gram that it is a goal to discover
models? some problemof n-grams that
satisfies a given set of
entrophy rate and
Understand the n-gram
probability.
17 Describe cross-entropy in Remember The learner will Find the CO 3
terms of probability process of establishing the
distribution? preconditions of the
probability distribution and
Understand the concept of
cross-entropy.
18 Explain about inflectional Understand The learner is able to Recall CO 3
morphology and how it is that it is a series of actions
useful in understanding that are responsible for a
English? particular outcome and
Understand the english
language.
19 Explain what is Understand The learner get to Recall CO 3
nominalization in terms of that it is a normalization
derivational morphology? procedure and Understand
how the morphology is used
20 Explain finite-state Remember The learner will Recall that CO 3
morphological parsing in it is a parsing procedure
simple words? which handling simple words
and the problem of FSM.
MODULE IV
WORD SENSE DISAMBIGUATION
PART A- PROBLEM SOLVING AND CRITICAL THINKING QUESTIONS

Page 23
1 How can we address the Apply The learner get to Recall CO 4
problem of ambiguity in how NLP and Understand
natural language and Apply how it allows
processing (NLP), and text processing.
what techniques can be
used to disambiguate
words with multiple
meanings?
2 How can we improve the Apply The Learner get to Recall CO 4
accuracy of word sense how word sense
disambiguation (WSD) disambiguation (WSD)
algorithms, particularly in algorithms is implemented
cases where there are few and Understand and
contextual clues available? Apply the knowledge
representation.
3 How can we address the Apply The learner get to Recall CO 4
problem of polysemy, how multiple meaning
where a single word may techniques and Understand
have multiple related the concept of inferencing
meanings, and what and Apply disambiguate
techniques can be used to techniques.
disambiguate these
meanings effectively?
4 How can we optimize the Apply the learner will get to Recall CO 4
performance of WSD the WSD is done in case of a
algorithms for different language processing and
languages or dialects, and Understand the concept of
what are some best best practices along with its
practices for doing so Application.
effectively?
5 How can we use WSD to Apply The learner will get to CO 6
improve the accuracy of Recall different ways of
information retrieval applying WSD Understand
systems, particularly when and Apply how to solve the
searching for specific search problems.
topics or keywords within
large datasets?
6 How can we use WSD to Apply The learner get to Recall, CO 6
improve the performance Understand the
of sentiment analysis implementation of WSD in
algorithms, particularly in medical domain and Apply
cases where the sentiment sentiment analysis.
of a word may vary
depending on its context?

Page 24
7 How can we use WSD to Apply The learner get to Recall CO 6
improve the performance and Understand and
of named entity Apply the process of NER
recognition (NER) systems.
systems, particularly in
cases where a named
entity may have multiple
possible meanings or
referents?
8 How can we use WSD to Apply The learner will able to CO 4
identify and correct errors Recall, Understand how
in machine translation machine learning systems
systems, particularly in and Apply the translation
cases where a word may systems
be mistranslated due to
its ambiguous meaning?

9 How can we evaluate the Apply The learner will Recall how CO 4
performance of WSD knowledge representation is
algorithms, and what so important, Understand
metrics should we use to how they Apply while
measure their accuracy solving some specific critical
and effectiveness? thinking problems.
10 How can we use WSD to Apply The learner will Recall, CO 5
enhance the accessibility Understand why these
of digital content for techniques are combinley
individuals with Apply to handle WSD
disabilities, such as those problem solving process and
who are visually impaired the individual performances.
or have dyslexia?
PART-B LONG ANSWER QUESTIONS
1 Create an artificial Apply The Learner get to Recall CO 4
training and test set using how this kind of reasoning is
pseudowords. Evaluate used to reason effectively and
one of the supervised Understand how it is even
algorithms on it? used when a complete,
consistent, and Apply
constant model of the world
is not available.
2 In light of the discussion Understand The Learner will get to CO 4
on closely related and Recall how the pseudocodes
‘co-activated’ senses, and Understand the
discuss to what extent different ambiguties.
pseudowords model
ambiguity well?

Page 25
3 How would one design a Understand The learner will get to CO 4
Bayes classifier that uses Recall the representation is
only one feature and an extended to allow some kind
information-theoretic of numeric measure of
method that uses many certainity and Understand
features? the feature extraction.
4 Discuss the validity of the Understand The learner get to Recall CO 4
“one sense per discourse” how variant different
constraint for different ambiguities and
types of ambiguity (types Understand how it support
of usages, homonyms one sense per disclosure.
etc.)?
5 Construct examples where Apply The Learner to Recall about CO 4
the constraint is expected different kinds of language
to do well and examples models Understand and
where it is expected to do Apply the various
poorly? applications.
6 Evaluate the one sense per Apply The Learner get to Recall CO 4
discourse constraint on a the use of constraint corpus
corpus. Find sections or and Understand and
articles with multiple uses Apply the concept of
of an ambiguous word, different senses.
and work out how often
they have different senses?
7 Pick one of the Understand The learner will able to CO 4
disambiguation algorithms Recall and Understand a
and apply it to sentence simple kind of minimalist
boundary identification? reasoning suggested by CWS.
8 Explain several important Understand The Learner will Recall CO 4
methodological some morphological
preliminaries in the preliminaries and
context of word sense Understand the word sence
disambiguation? disambiguation.
9 Explain the process to Apply The learner will Recall CO 4
apply flip-flop algorithm varieties of flip flop
to find indicators for algorithms, Understand
ambiguation? how that can be derived from
a set of ambiguties, and
Apply in solving
ambiguities.
10 Describes disambiguation Remember The learner will Find how CO 4
methods that rely on the disambiguities can be
definition of senses in resolved.
dictionaries and thesauri?

Page 26
11 Explain dictionary-based Remember The learner will Find about CO 4
disambiguation algorithm the direction based
in detail? disambiguation.
12 Explain in detail the Remember The learner will Find how CO 4
simple thesaurus-based semantic categorization is
algorithm to exploit the used in NLP.
semantic categorization?
13 Describe Yarowsky’s Remember The learner get to Recall CO 4
algorithms for adapting a how these theorems are used
semantic categorization of as Yarowsky’s algorithms and
words and for the semantic application.
thesaurus-based
disambiguation?
14 Describe dictionary-based Understand The learner will get to CO 4
algorithm by making use Recall the practical way of
of word correspondences compromising
in a bilingual dictionary? dictionary-based algorithms
and Understand different
bilingual factors.
15 Explain one sense per Understand The learner will get to CO 4
constraint in detail with Recall that the use of one
an example? sense per constraint and
Understand how the system
is used.
16 Explain Disambiguation Understand The learner get to Recall CO 4
based on “one sense per the mechanism behind the
collocation” and “one process of one sense per
sense per discourse.”? collocation and Understand
how it is useful in
disambiguation.
17 Describe context-group Understand The learner will get to CO 4
discrimination? Recall the mechanisms for
computing and Understand
the influence of any
context-group discrimination.
18 Explain the wide range of Understand The learner will Recall the CO 4
different approaches to alternative technique and
word sense Understand the sets of
disambiguation? word sense disambiguation.

Page 27
19 In light of the discussion Understand The learner will Recall how CO 4
on closely related and the motivation for extent
‘co-activated’ senses, pseudowords and also
discuss to what extent Understand the need to
pseudowords model represent the ambiguity as
ambiguity well? well.
20 Discuss differences and Understand The learner will Recall how CO 4
other similarities between such similarities and
the information-theoretic Understand the algorithm
disambiguation and implementation.
Yarowsky’s algorithm?
PART C - SHORT ANSWER QUESTIONS
1 Explain supervised Understand The Learner will Recall how CO 4
disambiguation based on the techniques that are
labeled training set? unable to overcome the
combinational explosive and
Understand how the
disambiguation is handled.
2 Explain dictionary-based Remember The learner will Find how CO 4
disambiguation based on the algorithm can generate a
lexical resources? possible solution in the
dictionary-based
disambiguation.
3 Explain unsupervised Remember The learner will Find the CO 4
disambiguation based on information about the
unlabeled training set planning process that uses
corpora? unsupervised disambiguation
technique which uses
different types of corpora.
4 What is the difference Remember The learner will Find how CO 4
between supervised and the procedure is used to help
unsupervised learning? the generator decide which
direction to move in the
search space.
5 Explain how do you Understand The learner get to Recall CO 4
differentiate classification the steps involved in defining
and clustering tasks in a solution of a AI problem
terms of algorithms? and Understand how such
systems are designed.
6 How can we test the Understand The learner get to Recall CO 4
performance of how the axioms in propability
disambiguation algorithms are representing and
on a natural ambiguous Understand its application.
word?

Page 28
7 Explain the term Understand The learner get to Recall CO 4
supervised disambiguation about how to use bayes rules
with respect to the and Understand its
available corpus? applications.
8 Explain how Bayes Understand The learner get to Recall CO 4
decision rule minimizes about the various semantics
the probability of errors? of the network and
Understand such networks.
9 Explain Naive Bayes Remember The learner can get to CO 4
assumption in detail? Recall the inference achieved
in this network.
10 Explain Decision rule for Remember The learner can get to CO 4
naive bayes? Recall about the
implementation of the model.
11 Explain the term Remember The learner can get to CO 4
dictionary-based Recall about the kinds of
disambiguation? uncertainty.
12 Explain the term Understand The learner can get to CO 4
Thesaurus-based Recall the implementation
disambiguation? of the model at a particular
applications and
Understand how it is useful
in speech recognition.
13 Explain the Understand The learner can Recall how CO 4
dictionary-based to handle inferences and
algorithm which refers to Understand how the
the language of network inferences are
application? handled.
14 Explain the term Remember The learner will Find various CO 4
unsupervised components of PS.
disambiguation?
15 What is a word sence and Understand The learner can get to CO 4
explain different Recall about the learning
approaches to word sense process and Understand its
disambiguation? What are categories.
its types?
16 Explain the Remember The learner get to Recall the CO 4
disambiguation based on basics of the program and
sense definitions? Understand its application.
17 Describe how Understand The learner get to Recall CO 4
thesaurus-based about the generalization
disambiguation exploits technique and Understand
the semantic the generalization concept.
categorization?

Page 29
18 Describe the Understand The learner will Recall and CO 4
implementation of Understand the
Flip-Flop algorithm? information about STRIPS.
19 Explain how upper and Understand The learner can Recall and CO 4
lower bounds are used to Understand to know about
measure the performance the definition?
of one’s algoithm?
20 Explain the term Understand
The learner can Recall and CO 4
pseudowords in word Understand the need of
sence disambiguation? nonlinear plan.
MODULE V
MARKOV MODEL AND POS TAGGING
PART A-PROBLEM SOLVING AND CRITICAL THINKING QUESTIONS)
1 How can we improve the Apply The learner will able to CO 5
accuracy of Markov Recall, Understand the
models in natural concepts of effective
language processing communication and Apply
(NLP) tasks such as in acquiring domain specific
part-of-speech tagging or knowledge
text generation?
2 How can we optimize the Apply The leaner will Find the CO 3
performance of Markov problem related to certain
models for different POS issues that impacts the
languages or dialects, and Understand the
what are some best performance and Apply in
practices for doing so some best practices.
effectively?
3 How can we address the Apply The learner will Find CO 3
problem of overfitting Understand how the rules
when building Markov are choosen and Apply to
models, and what solve some domain specific
techniques can be used to tasks.
mitigate this issue?
4 How can we use Markov Apply The learner will Recall how CO 4
models to identify the earliest techniques
patterns or trends in large developed for solving
datasets, and what are compound goals,
some potential Understand how they may
applications of this interact with the use of goal
technology in fields such stack and Apply in solving
as marketing or finance? blocks world problem.

Page 30
5 How can we use Markov Apply The learner will Find how a CO 3
models to analyze and non linear plan is composed
model the behavior of of a linear sequence of
users in online platforms, complete subplans,
such as social media or Understand the use of the
e-commerce sites, and NLP models and Apply the
what insights can we gain model behaviours.
from this analysis?
6 How can we use Markov Apply The learner will Recall, CO 5
models to detect and Understand the possibilities
analyze the evolution of to Apply this procedure
language over time, and while playing chess and the
what insights can we gain learning performance.
from this analysis?
7 How can we use Markov Apply The learner will Recall the CO 4
models to analyze and concepts of version space,
model the behavior of Understand the working
users in online platforms, and Apply the algorithm.
such as social media or
e-commerce sites, and
what ethical
considerations should we
keep in mind when doing
so?
8 How can we use Markov Apply The learner will Recall the CO 4
models to generate new idea about the developing a
text data that is similar in program, Understand some
style or content to existing language grammer and
text data, and what are Apply the same in natural
the potential applications language understanding.
of this technology in fields
such as advertising or
creative writing?
9 How can we use Markov Apply The learner Recall, CO 5
models to detect and Understand how the
correct errors in machine rule-based systems Apply
translation systems, for critical thinking and the
particularly in cases where performance of RBS.
the errors may be caused
by subtle differences in
meaning between words or
phrases?

Page 31
10 How can we use Markov Apply The learner will Recall the CO 4
models to improve the some effective factors of
performance of named neural networks that are
entity recognition (NER) different from both the
systems, particularly in systems, Understand how
cases where a named they are useful during
entity may have multiple knowledge representation and
possible meanings or Apply during knowledge
referents? representation.
PART-B LONG ANSWER QUESTIONS
1 Explain the three Understand The learner can Recall the CO 5
fundamental questions for information aboutexpert
HMMs in detail? system shell and
Understand the effective
graph data visualization
2 What do you mean by a Understand The will Recall the CO 5
markov assumption and characteristics of scatter
explain in detail the plots and Understand the
markov model? problems in expert systems
3 What is Emission Understand The learner will Recall CO 5
Probability? Explain in about graphs and
detail in terms of Understand the process of
transition probabilities? knowledge acquisition
4 Discuss briefly about Understand The learner will Recall and CO 5
linear interpolation in Understand about verbal
terms of n-gram model? communnication in deatil.
5 Explain the general form Understand The learner can Recall the CO 5
of an HMM with its presenting skills and
types? Understand its applications
6 What is the process of Understand The learner can Recall CO 5
finding the probability of about the uses of bar graphs
an observation for a given and Understand about
state sequence? various diagrams in expert
system
7 Describe how dynamic Understand The learner Recalls the CO 5
programming or knowledge of verbal
memoization and explain communication and
how it avoids the Understand its types
complexity in
remembering partial
results?

Page 32
8 Differentiate forward and Understand The learner will able to CO 5
backward procedure that Recall the line grapth and
can be implemented in Understand the expert
terms of one shorter system building process
subpaths?
9 Explain the calculation Understand The learner get to Recall CO 5
procedure of forward and how to improve your
backward variables for the communication skills and
soft drink machine using Understand various roles
an example?
10 Describe the procedure og Understand The learner will Recall the CO 5
finding the best state basic idea about expert
sequence to explain the systems and Understand
observations in markov various issues that arises
models? during the development of
expert systems.
11 How can we can work out Apply The learner get to Find the CO 5
the probability that the steps steps for presenting
Markov process went from Understand the probability
state Si to s, at time 1? calculation and Apply while
solving similar problems.
12 Explain in detail various Understand The learner will Recall, CO 5
practical implementation Understandthe detail of
issues of HMMs? verbal communication
13 What is the ultimate goal Remember The learner get to Recall CO 5
of research on natural about persuasive
language processing. communication and the
Explain in detail? features of expert systems
14 What are other linguistic Understand The learner will Recall the 5 CO 5
phenomena that are not P’s of presentation and
modeled correctly by Understand the key
Markov chains? Which characteristics of expert
general property of system
language is common to
these phenomena?
15 Why is Time Invariance Understand The learner can Recall and CO 5
problematic for modeling Understand about he
language? effective speech
16 Compute the following Apply The learner can Recall CO 5
two probabilities: n P(AT Understand the steps for
NN BEZ IN AT NN( The creating the effective
bear is on the move.) = visualization and Apply to
P(AT NN BEZ IN AT vB( solve the problem
The bear is on the move.)?

Page 33
17 Explain in detail the Understand The learner get to Recall CO 5
algorithm for tagging with the keywords related to these
a Visible Markov Model terms and Understand the
Tagger? expert system shell
18 Explain how to estimate Understand The learner can able to CO 5
word generation Recall the correlation with
probabilities for words its types and Understand
that occur in the corpus. ? the need of XCON expert
systems
19 Describe the procedure of Understand The learner will Recall the CO 5
applying the HMMs to skills for data presentation,
POS tagging in detail? Understand the design
process during the
development of an POS
tagging
20 Explain clearly the effect Understand The learner will Recall and CO 5
of initialization on HMM Understand the different
training with parts of line graph
forward-backward
algorithm?
PART-C SHORT ANSWER QUESTIONS
1 Define the term Hidden Understand The Learner will Recall the CO 5
Markov Model? knowledge and Understand
the definition of hidden
markov model.
2 Describe how to Remember The learner will Find the CO 5
distinguish the visible different characteristics of
markov models from hidden markov model and
hidden markov models? Understand the various
features of ES.
3 What do you meant by Understand The learner will Recall the CO 5
markov assumption and purpose of developing and
explain in its properties? Understand using markov
assumption
4 How markov models are Understand The learners can Recall the CO 5
used to model the major areas of ES and
probability of a linear Understand how it is
sequence of events? implemented successfully.
5 Briefly explain HMM with Understand The learner will Recall the CO 5
an example? basic information about the
process of acquiring and
Understand the knowledge
while developing an NLP
systems

Page 34
6 What is the purpopse of Understand The learner will Recall the CO 5
using HMMs? information about the
importance of MODEL and
Understand the HMM
concepts.
7 Explain the general form Understand The learner will able to CO 5
of an HMM? Recall different strategies
that are being used in expert
systems and Understand
the various strategies
8 Differentiate the terms Understand The learner can Recall the CO 5
Arc-Emission HMM and information about the basic
State-Emission HMMs? information and
Understand about HMM
9 How can we find the Understand The Learner can Recall and CO 5
proabability of an Understand the idea about
observation? different applications where
the POS is implemented.
10 What do you mean by Understand The learner can Recall CO 5
dynamic programming about various tools available
memorization? in the ES environments and
Understand the helping
while developing dynamic
programming.
11 Explain the process of Understand The learner can Recall CO 5
finding the best state about the usage of various
sequence in HMM? tools and Understand how
they are used while
developing HMM.
12 Explain the problem of Understand The Learner can Recall CO 5
parameter estimation? different programming
languages and Understand
how they are used to develop
parameter estimation
applications.
13 Explain the iterative POS Understand The learner get to Recall CO 5
tagging? some of the tools and
Understand how they are
required to develop expert
system.

Page 35
 14 Explain the practical Understand The learner can Recall CO 5
issues of implementing different methodologies and
HMMs? Understand how they are
supported to implement
HMMs.
15 Explain briefly different Understand The learner Recall and CO 5
variant forms of HMMs? Understand the knowledge
representation types where
HMMs are used.
16 How can one decide the Understand The learner can Recall the CO 5
correct part of speech for drawbacks in certain areas
a word used in a context? where speech recognition has
been failed and Understand
the concept of domain expert.
17 Explain Markov Model Remember The learner will able to Find CO 5
Taggers in probabilistic about the application areas
model? where the Markov Model is
being used.
18 Explain the step-by-step Understand The learner can Recall the CO 5
process of implementing concepts related to ES and
Viterbi algorithm? Understand the using
viterbi algorithm.
19 how to estimate word Understand The leaner will Recall and CO 5
generation probabilities Understand the information
for words that occur in about word generation.
the corpus?
20 What is a tagger, explain Understand The learner will Recall CO 5
different types of taggers? about various tagger types
and Understand how they
are already succeeded after
tagger.

Course Coordinator: HOD,CSE(AI&ML)

Dr. M.Nagaraju, Assistant Professor

Page 36