0% found this document useful (0 votes)
12 views37 pages

AI ModuleWise QstnBank

The document is a question bank for a Bachelor of Computer Application course on Artificial Intelligence, covering multiple-choice questions and descriptive questions related to AI concepts, applications, and algorithms. It includes topics such as machine learning, state space search, expert systems, and various search algorithms. The document serves as a study resource for students to prepare for exams in the subject.

Uploaded by

wbwwfmwx9y
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
12 views37 pages

AI ModuleWise QstnBank

The document is a question bank for a Bachelor of Computer Application course on Artificial Intelligence, covering multiple-choice questions and descriptive questions related to AI concepts, applications, and algorithms. It includes topics such as machine learning, state space search, expert systems, and various search algorithms. The document serves as a study resource for students to prepare for exams in the subject.

Uploaded by

wbwwfmwx9y
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 37

The Yenepoya Institute of Arts, Science, Commerce and Management

A Constituent Unit of Yenepoya (Deemed to be University)

III SEMESTER – BACHELOR OF COMPUTER APPLICATION


BCA304 - Introduction to Artificial Intelligence

Question Bank - MODUE 1

MCQ(1 Mark)
1. Who coined the term artificial intelligence in 1956
A) Marvin Minsky
B) Allen Newell
C) John McCarthy
D) Herbert A
2. The application/applications of Artificial Intelligence is/are
A) Expert Systems
B) Gaming
C) Vision Systems
D) All of the above
3. Artificial Intelligence is
A) Termed as Augmented Intelligence
B) Tool that uses computer to complete a task with little to no human intervention.
C) Teaching machines to learn, act and think like humans.
D) All of the above
4. What is the primary goal of machine learning in the context of artificial intelligence?
A) To create machines that can think like humans
B) To enable machines to learn from data and improve over time
C) To replace human workers in all jobs
D) To design hardware that is more efficient
5. Which term describes a type of AI that can perform tasks that are limited to a specific problem or
domain?
A) General AI
B) Quantum AI
C) Superintelligent AI
D) Narrow AI
6. What does NLP stand for in the context of AI?
A) Neural Learning Process
B) Networked Learning Protocol
C) Natural Language Processing
D) Non-Linear Programming
7. What is the primary purpose of computer vision in AI?
A) To enable machines to understand and process human language.
B) To allow machines to interpret and analyze visual information from the world.
C) To simulate human cognitive processes and reasoning.
D) To optimize decision-making processes based on historical data.
8.In which area of AI is Natural Language Processing (NLP) primarily used?
A) Analyzing and generating human language
B) Developing autonomous robots
C) Optimizing financial trading strategies
D) Enhancing computer vision capabilities
9.Which of the following is an example of narrow AI?
A) A general-purpose robot capable of performing a wide range of tasks.
B) An AI that can play chess at a grandmaster level.
C) A system designed to understand and generate human-like text across various topics.
D) An AI that can independently research and develop new theories in physics.
10.Among the given options, which search algorithm requires less memory?
A) Depth First Search
B) Breadth-First Search
C) Linear Search
D) Optimal Search
11. Which of the given language is not commonly used for AI?
A) LISP
B) PROLOG
C) Python
D) Perl
12. The component of an Expert system is .
A) Knowledge Base
B) Inference Engine
C) User Interface
D) All of the above
13. The available ways to solve a problem of state-space-search.
A) 1
B) 2
C) 3
D) 4
14. Which term describes the common-sense of the judgmental part of problem-solving?
A) Heuristic
B) Critical
C) Analytical
D) Values-based
15. In state-space, the set of actions for a given problem is expressed by the .
A) Intermediate States
B) Successor function that takes current action and returns next state
C) Initial States
D) None of the above
16. Ways to achieve AI in real-life are .
A) Machine Learning
B) Deep Learning
C) Both a & b
D) None of the above
17. Game Alpha Go is an example for .
A) Reinforcement Learning
B)Supervised Learning
C)Unsupervised Learning
D)All of the above
18. was originally called ‘the Imitation game’.
a) The Turning Test
b) LISP
c) The Halting Problem
d) None of the above
19. is the informed search method.
a) Memory Bound Heuristic Search
b) A * Search
c) Best First Search
d) All of the above
20. Define Artificial Intelligence?
Answer) It is a branch of computer science by which we can create intelligent machines
which can behave like a human, think like humans, and able to make decisions
(OR)
Answer) AI ,is the technology that enables computers and machines to simulate human
intelligence and problem solving capabilities.
21. Write any two goals of AI?
Answer)1.Replicate human intelligence.
2.Building a machine which can perform tasks that require human intelligence. Such as:
Playing chess, Plan some surgical operations etc.
22. What is supervised learning?

A) Learning without labeled data


B) Learning with labeled data
C) Learning by trial and error
D) None of the above
23. Which of the following is a supervised learning algorithm?
A) K-Means Clustering
B) Principal Component Analysis (PCA)
C) Decision Trees
D) Apriori Algorithm
24. What is the primary objective of unsupervised learning?
A) To predict future outcomes
B) To find hidden patterns in unlabeled data
C) To classify data points
D) To minimize error
25. Which machine learning algorithm is primarily used for classification tasks?
A) Linear Regression
B) Apriori
C) K-Means Clustering
D) K-Nearest Neighbors (KNN)
26. In Reinforcement Learning, what is the term used to describe the feedback the agent receives
after taking an action?
A) Reward
B) Policy
C) Environment
D) State
27. Which of the following best describes the primary goal of an agent in Reinforcement Learning?
A) To classify data accurately
B) To maximize cumulative rewards
C) To minimize prediction error
D) To label data points
30. In Reinforcement Learning, which of the following is NOT typically an element of the agent-
environment interface?
A) Reward
B) Action
C) Supervision
D) State
31. What does the agent do when it 'exploits' in Reinforcement Learning?
A) It tries new actions to gather information
B) It selects actions with minimal reward
C) It stops interacting with the environment
D) It takes actions that maximize known rewards based on past experience
32. What is a primary challenge in Reinforcement Learning, especially in complex environments?
A) Managing labeled data
B) Balancing exploration and exploitation
C) Reducing dimensionality of features
D) Ensuring a linear relationship between input and output
33. What type of learning is Reinforcement Learning primarily based on?
A) Supervised learning
B) Unsupervised learning
C) Semi-supervised learning
D) Trial and error learning
34. In Reinforcement Learning, which of the following describes the 'state'?
A) The set of all possible actions
B) The current situation of the environment observed by the agent
C) The agent’s current reward
D) The sequence of previous actions taken
35. Which of the following is an example of a Reinforcement Learning application?
A) Image classification
B) Clustering data points
C) Game playing, such as AlphaGo
D) Predicting stock prices
36. What is the primary goal of State Space Search in problem-solving?
A) To classify data into groups
B) To find a path from the initial state to the goal state
C) To minimize the number of states
D) To reduce the complexity of a problem
37. In State Space Search, what is a 'state'?
A) A possible condition or arrangement in the problem
B) The desired solution of the problem
C) The initial point where the search starts
D) The series of actions taken to solve the problem
38. Which of the following represents the starting point in State Space Search?
A) Goal State
B) Operator
C) Path
D) Initial State
39. What is a 'goal state' in State Space Search?
A) The beginning condition of the problem
B) A sequence of actions taken by the agent
C) The desired final state or solution of the problem
D) A state that has no further transitions
40. In the context of State Space Search, what are 'operators'?
A) Possible actions or moves that transition from one state to another
B) Solutions to the problem
C) Starting and ending points of the search
D) The states the agent aims to avoid
41. What is a 'path' in State Space Search?
A) The distance between the initial and goal state
B) The series of states from the initial state to the goal state
C) The operator that takes the agent to the goal
D) The first and last states in the problem
42. Which of the following best describes the purpose of 'operators' in State Space Search?
A) To define the initial and goal states
B) To minimize the path length
C) To reduce the number of states in the problem
D) To move from one state to another
43. What type of problems is State Space Search most commonly applied to?
A) Classification problems
B) Path-finding and puzzle-solving problems
C) Clustering problems
D) Regression problems
44. In State Space Search, what does the term 'solution path' mean?
A) The series of operators that minimize the search space
B) The shortest path between all possible states
C) The sequence of states and operators that lead from the initial state to the goal state
D) A set of disconnected states
45. Which of the following algorithms is NOT considered an Informed Search?
A) A* Search
B) Depth-First Search
C) Best-First Search
D) Hill Climbing
46. In a Sudoku puzzle, the domains of the variables are typically:
A) The set of possible moves
B) The numbers from 1 to 9
C) The rows and columns of the grid
D) The set of possible solutions

Descriptive Questions(10 Marks)


1. Explain various applications of Artificial Intelligence
2. Define Artificial Intelligence ? Explain the goals of AI
3. Explain the advantages and disadvantages of Artificial Intelligence
4. Discuss the different classifications of AI . (Ans : Narrow/Weak, General/strong and
super/conscious)
5. Define Machine Learning. Explain different types of Machine Learning.(Ans : Supervised,
Unsupervised and Reinforcement Learning)
6. Describe Turing test and its significance.
7. Write a short note on a) Knowledge Acquisition b) Expert systems c) Production
systems.
8. Write a short note on production systems and its components.
9. Explain state space search with a suitable example.
10. Explain the significance of means end of analysis and its procedure.
11. Explain any one expert system that you know in real life (E.g: ChatGPT, Alexa etc).
12. Justify why Artificial General Intelligence (AGI) is necessary with a suitable example.
Do you believe it's possible to develop it within this century?
13. Explain constraint satisfaction and its components with a suitable example.
14. What is the main difference between Uninformed Search and Informed Search
15. Explain the difference between Uninformed (Blind) Search and Informed Search
algorithms. How does the availability of information about the goal state affect the
performance and efficiency of each search type?
16. What is the role of the heuristic function in Hill Climbing and Best-First Search
algorithms? Explain how it helps these algorithms make more efficient decisions
compared to Uninformed Search methods.
17. What is a heuristic function, and why is it crucial in Informed Search algorithms like
A and Best-First Search? Provide examples of how heuristic functions improve search
efficiency
18. Explain how the Depth-First Search algorithm explores the search space. Under what
conditions would Depth-First Search be an ideal choice, and what limitations might it
face in reaching a solution?
19. How do variables, domains, and constraints work together in a Constraint Satisfaction
Problem? Use an example such as map-coloring to illustrate your answer.
20. Describe the process of solving a Constraint Satisfaction Problem. How does the
algorithm decide which values to assign to the variables

Descriptive Questions(15 Marks)

1. Explain an Expert systems and its components with a neat diagram.


2. Differentiate different informed search and uninformed search with suitable examples.
3. Explain the characteristics and Algorithm Breadth First Search with example.
4. Explain the characteristics and Algorithm for DFS with example.
5. Explain the different Search Techniques in Branch and Bound
6. Discuss the heuristic function. Explain how the heuristic function helps during search
procedure. Explain with a suitable example.
7. Explain Best First Search Algorithm with an example.
8. Demonstrate AO* algorithm with a suitable example.
9. Discuss A* algorithm with suitable example.
10. Compare the A* and AO* algorithms in their effectiveness for problem solving in AI
systems.
11. Describe Breadth-First Search and Depth-First Search algorithms. How do these
Uninformed Search techniques operate without knowledge of the goal state, and what
are their potential advantages and disadvantages?
12. Explain the importance of hill climbing algorithm and detail on its procedure. List down any
two disadvantages.
13. Provide an example of a real-world problem that could be solved using Informed Search
techniques. Describe the role of heuristics in this example and discuss how they improve the
search process.
14. Given the following graph, explain how breadth first search and depth first search traversals
are performed. Draw the traversal paths of BFS and DFS separately (showing the step-by-
step process) and write down their output, if A is the source vertex.
15. Given the following graph, explain how breadth first search and depth first search traversals
are performed. Draw the traversal paths of BFS and DFS separately (showing the step-by-
step process) and write down their output, if A is the source vertex.

16. Find the shortest path in the following using Best First Search algorithm, if S is the source
vertex and G is the destination vertex. Observe that the heuristic values are provided in the
table.
17. Find all possible paths between the source (S) and destination (G) vertices using A*algorithm
for the following graph. Calculate the cost of each path and identify the shortest path among
them.

18. Find all possible paths between the source (S) and destination (G) vertices using A*algorithm
for the following graph. Calculate the cost of each path and identify the shortest path among
them.
19. Identify the following graph and suggest the best algorithm that can solve it using
heuristics.Detail on the solution procedure with necessary steps.
Question Bank - MODULE 2

MCQ (1 Mark)
1. Inference engines work based on the principle of
A) Backward Chaining
B) Forward Chaining
C) Both A and B
D) None of the above
2. The proposition symbols in AI are
A) true and false
B) true, false, and null
C) true
D) false
3. What can be viewed as a single literal of disjunction?
A) Multiple clause
B) Combine clause
C) Unit clause
D) None of the above
4. Knowledge in AI can be represented as
A) First order Logic
B) Production Rules
C) Both A and B
D) None of the above
5. are the actions which occurs in the world.
A) Events
B) Objects
C) Facts
D) Performance
6 . The components retrieves data or information from the environment.
A) Perception
B) Learning
C) Execution
D) Planning
7.Frames consists of and .
A) Syntax and Semantics
B) Attributes and Values
C) Facts and Objects
D) Knowledge and Reasoning

9.Semantic networks, represent our knowledge in the form of networks.


A)Graphical/ Graphs
B) Logical
C)Frames
D)None of these
9.One among the following is not a type of knowledge. Identify it.
A) Declarative
B) Procedural
C) Heuristic
D) Principle
10.Which process makes two different Logical expressions look identical
A)Unification
B)Lifting
C)Inference Process
D)None of the above
11.In propositional Logic, which among the following statements is false?
A)The statements can represent existential quantifiers.
B) Each sentence is a declarative sentence.
C)Universal quantifiers are not allowed.
D)Negation is not allowed
12.First order logic Statements contains .
A)Predicate and Preposition
B)Subject and an Object
C)Predicate and Subject
D)None of the above
13.Identify the declarative knowledge from the following:
A)Blood is red
B)If it is cold, wear a jacket.
C)Follow the steps given for preparing tea: Boil the water with tea and sugar.
Filter andserve.
D)Google map providing the route map and providing additional knowledge
on thedistance, time taken and traffic congestions.
14. State whether the following condition is true or false?
“The word Orange contains the letter ‘a’.
A)True
B)False
15.Consider, A – It is hot
B – It is humid
C – It is raining.
In First Order Logic, how do you represent the following statement: “ If it is humid, then it is
hot”
a. A →B
B) B→A
C) C→A
D) A→D
16.What is the symbol of the word “IMPLIES”
A) =
B) ^
C) →
D) →
17.One among the following is not a representation in First Order Logic: .
b. ∃
c. ∀
d. →
e. **
18. Which among the following is not true for a Horn clause?
A) “The clause should contain atleast one positive literal”
B) “It is a clause with atmost one positive literal”
C) “Horn clause can be without a positive literal”
D) “It can be a clause with no negative literal”
19.AI Knowledge cycle starts with the following step:
A) Perception
B) Knowledge representation
C) Reasoning
D) Final Execution
20.In First Order Logic Statement - "Every human is mortal" can be written as:
E) (Human(x)→Mortal(x))
B) ∀x (Human(x)→Mortal(x))
C) ∃ x (Human(x)→Mortal(x))
D) ¬ (Human(x)→Mortal(x))
21. _______________ is a data-driven approach in reasoning.
A) Forward chaining
B) Backward chaining
C) Monotonic reasoning
D) Fuzzy reasoning
22.One among the following is not a property of backward chaining:
F) Data-driven
G) Goal-oriented
H) Reduces unnecessary exploration
I) Logical and direct
23.One among following is an issue in knowledge representation:
J) providing declarative knowledge
K) adding procedural knowledge
L) managing incomplete knowledge
M) All of the above
24.One among the following is not an issue in knowledge representation:
A) Identifying basic attributes
B) Identifying complex relationships among attributes
C) Choosing the granularity of representation
D) Representing set of objects
25.The knowledge representation techniques define
N) how information is entered into the knowledge base and enable learning
O) how to collect data
P) how to redefine information retrieval
Q) how to reason out during question answering
26.Without proper knowledge representation, expert systems will _
R) succeed
B) fail
C) survive
D) be fast in response
27.Which of the following is NOT a commonly used approach in knowledge representation?
S) Semantic Networks
T) Frames
U) Logic Programming
V) Virtual Reality
28. What is the characteristic of a "frame" in knowledge representation?
A) It is a method for representing procedural knowledge.
B) It is a data structure used to represent stereotypical situations.
C) It is a form of deductive logic.
D) It is a method for representing uncertain information.
29. Which of the following is NOT a feature of "propositional logic"?
A) The ability to express true/false statements.
B) The ability to express relationships between objects.
C) The ability to combine statements using logical connectives (AND, OR, NOT).
D) The ability to deal with quantifiers like "all" and "some".
30. What is a semantic network?
A) Type of neural network used in machine learning
B) Graph structure representing relationships between concepts
C) Database of words and their meanings
D) Mathematical model for sentence parsing
31. In a semantic network, what does a node typically represent?
A) Word in a sentence
B) Set of relationships
C) Concept or object
D) Rule for semantic interpretation
32. Which of the following is the correct negation of the proposition: "If it rains, then I will go
outside"?
A) It rains and I will not go outside.
B) It does not rain and I will go outside.
C) It rains and I will not go outside.
D) I will go outside, but it does not rain.
33. The logical operation that corresponds to the phrase "p implies q" is:
A) Conjunction
B) Disjunction
C) Negation
D) Implication
34. What is the main advantage of using a knowledge graph for knowledge representation?
A) It is simple and requires little storage.
B) It supports complex queries and relationships between entities.
C) It only works for structured data.
D) It has no advantage over traditional databases.
35. Which of the following is an example of a knowledge representation formalism that uses "if-then"
rules?
A) Semantic Network
B) Frame-based System
C) Production Rules
D) Bayesian Network
36. What is a "production rule" in Artificial Intelligence?
A) Set of mathematical formulas to perform calculations
B) Conditional statement used to guide behaviour in a production system
C) Specific language syntax
D) Algorithm to sort data
37. Which of the following is a primary advantage of using production rules in AI systems?
A)They help reduce the computational cost significantly.
B)They provide a modular way to represent and update knowledge.
C)They allow for faster data processing than neural networks.
D) They automate system testing.
38. In forward chaining, production rules are applied based on:
A) The goal to be achieved
B) Available data and facts
C) Neural network weights
D) Random choice
39. In forward reasoning, the process begins with:
A) Hypothesis
B) Conclusion
C) Set of initial facts
D) An unknown outcome
40. Backward reasoning is also known as:
A) Data-driven reasoning
B) Goal-driven reasoning
C) Forward-chaining reasoning
D) Procedural reasoning
41. A doctor suspects a specific disease based on a patient’s symptoms and works backward by
eliminating other possible diseases. This is an example of:
A) Forward reasoning
B) Backward reasoning
C) Inductive reasoning
Descriptive Questions (5 Marks)

1. Define Knowledge Representation. Mention the basic kinds of knowledge required by the
AI systems.
2. Explain any FOUR types of Knowledge representation in AI.
3. Explain Knowledge Representation Cycle in AI with a neat diagram.
4. Explain Components of First Order Logic.
5. Explain the following:
a) Resolution b) Unification c) Model Checking
6. Find the truth values of the Predicates.
a) Let p(x) denotes the statement “the word x contains the letter ‘a’ “. What are the
truth values of p(Orange), p(Lemon), p(true) and p(false)
b) Let q(x,y) denotes “x is the capital of y”. What are the truth values of
q(Delhi,India),q(New York ,New York) and q(Srilanka,Colombo)
c) If p(x) denotes (x<4) then what are the truth values of p(0),p(6) and p(10).
7. Describe Horn clauses and classify their types with suitable examples.
8. Using Semantic Network notation, illustrate the relationships in the given statements
a. Tom is a cat
b. Cat is a mammal
c. Mammal is an animal
d. Tom is coloured in white.
e. Cat likes fish
f. Cat sat on the mat
g. Tom is owned by John
9. Explain any THREE advantages of forward chaining.
10. Explain forward reasoning with examples.
11. Explain Backward reasoning with examples.
12. Explain why knowledge representation is difficult in medical diagnosis.

Descriptive Questions (15 Marks)

1. Differentiate declarative and procedural representation of knowledge. Describe


them indetail with appropriate examples.
2. Discuss different types of knowledge representation with suitable examples.
3. Explain the forward and backward reasoning with suitable steps and examples.
Compareand contrast them.

4. Explain the significance of First-order-logic and provide necessary representation


for thefollowing cases:
a. The sky is blue.
b. If it is raining, the ground is wet.
c. Every human is mortal.
d. Some numbers are prime.
e. It is neither sunny nor warm.
5. Explain semantic networks in detail. Draw a semantic network indicating the
relationshipof Tom (cat )and Jerry (mouse) in the cartoon.
6. Explain Frames with a suitable example. Compare it with other knowledge
representations.
7. Explain any FIVE issues in knowledge representation in detail.
8. Explain Prepositional Logic with necessary examples.
9. Explain First order Logic with examples
10. Explain the difference between forward reasoning and backward reasoning.

Question Bank - MODUE 3

MCQ (1 Mark)

1. Which of the following is a common application of NLP?

• A) Image recognition
• B) Speech recognition
• C) Video processing
• D) Game development

2. Which technique is used for determining the sentiment of a piece of text?

• A) Named Entity Recognition


• B) Part-of-Speech Tagging
• C) Sentiment Analysis
• D) Machine Translation

3. What is the purpose of tokenization in NLP?

• A) To convert text into numerical data


• B) To break text into smaller pieces, like words or phrases
• C) To translate text into another language
• D) To identify the grammatical structure of sentence

4. What does 'lemmatization' aim to achieve in NLP?


• A) To remove stop words
• B) To find the root form of a word
• C) To classify text
• D) To identify named entities

5. Which of the following tasks is NOT typically associated with NLP?

• A) Language translation
• B) Image classification
• C) Chatbots
• D) Text summarization

6. In which application would you find 'language modeling' being used?

• A) Weather forecasting
• B) Speech recognition
• C) Stock market prediction
• D) Image editing

7. What is the first step in the NLP pipeline?

• A) Tokenization
• B) Data Collection
• C) Feature Extraction
• D) Text Preprocessing

8. Which of the following processes involves converting text into a structured format?

A) Parsing

B) Tokenization

C) Stop Word Removal

D) Lemmatization

9. During which step are common words like 'the', 'is', and 'and' removed from the text?

• A) Tokenization
• B) Stop Word Removal
• C) Stemming
• D) Part-of-Speech Tagging

10. What does stemming aim to achieve?

• A) Translate text
• B) Normalize words to their root forms
• C) Analyze sentiment
• D) Classify text

11. Which of the following properties is true about a context-free grammar (CFG)?

• Every context-free grammar is ambiguous.


• B) A context-free grammar can generate only regular languages.
• C) A context-free grammar can generate non-regular languages.
• D) A context-free grammar cannot generate palindromes.

12. Which of the following is not a characteristic of context-free grammars (CFGs)?

• A) They can generate all regular languages.


• B) They are used to describe the syntax of programming languages.
• C) They can describe nested structures like parentheses.
• D) They can generate languages that are not regular.

13. Which of the following is NOT a common preprocessing step in NLP?

• A) Tokenization
• B) Lemmatization
• C) Part-of-speech tagging
• D) Data augmentation

14. What is the purpose of stopword removal in NLP?

• To improve the computational efficiency of text processing


• B) To reduce the size of the vocabulary
• C) To remove commonly used words that do not contribute much to meaning
• D) All of the above

15. Which of the following tasks does tokenization involve?

• Splitting text into sentences


• B) Splitting text into words
• C) Part-of-speech tagging
• D) Sentiment analysis

16. What does the term "lemmatization" refer to in text processing?

• Removing non-alphanumeric characters


• B) Reducing words to their root forms based on the dictionary
• C) Splitting text into individual words
• D) Removing stop words

17. What is the primary task of a syntactic parser in NLP?

• To identify entities and their relationships


• B) To determine the grammatical structure of a sentence
• C) To classify the sentiment of a sentence
• D) To generate a response to a query

18. What is the main difference between "stemming" and "lemmatization"?

• Stemming removes prefixes, while lemmatization removes suffixes


• B) Stemming works based on a dictionary, while lemmatization uses heuristic
rules
• C) Stemming reduces words to their base form, while lemmatization reduces them
to their dictionary form
• D) Stemming is used in machine translation, while lemmatization is used in
sentiment analysis

19. Which of the following is NOT a part of a typical NLP pipeline?

• Text tokenization
• B) Word embedding
• C) Model evaluation
• D) Data mining

20. What is a "context-free grammar" (CFG) used for in NLP?

• A) To describe the syntax of natural language


• B) To generate word embeddings
• C) To identify named entities
• D) To classify text

21. What does a "language model" in NLP do?

• Translates text from one language to another


• B) Predicts the probability of the next word in a sentence
• C) Identifies named entities in a text
• D) Summarizes long documents

22. Which of the following methods is used to break down a text into sentences?

• Lemmatization
• B) Tokenization
• C) Sentence segmentation
• D) Stopword removal

23. Which of the following is a technique used to handle misspelled words in NLP?

• A) Tokenization
• B) Spelling correction
• C) Lemmatization
• D) Named entity recognition

24. Which technique is used to assign words their base or dictionary form?

• Lemmatization
• B) Tokenization
• C) Stopword removal
• D) Word2Vec

25. Which of the following is a type of token that is generally removed in NLP
preprocessing?

• A) Nouns
• B) Verbs
• C) Stopwords
• D) Named entities

26. In the context of NLP, what is "lemmatization"?

• Reducing a word to its root form using a dictionary lookup


• B) Removing stopwords
• C) Converting a word to lowercase
• D) Breaking text into sentences

27. Which of the following steps in NLP is used to normalize text?

• Tokenization
• B) Lemmatization
• C) Part-of-speech tagging
• D) Stopword removal

28. In NLP, what is the primary goal of tokenization?

• A) To remove unwanted symbols from text


• B) To split text into smaller units like words or subwords
• C) To identify named entities
• D) To assign part-of-speech tags

29. Which of the following NLP steps involves identifying the part of speech (POS) for
each word in a sentence?

• Lemmatization
• B) Part-of-speech tagging
• C) Named entity recognition
• D) Tokenization

30. What is "stemming" in the context of text preprocessing?

• Reducing words to their dictionary form


• B) Removing stopwords from a text
• C) Breaking text into words or phrases
• D) Cutting off prefixes or suffixes from words to find their root form

31. Which of the following is NOT typically part of the text preprocessing pipeline in
NLP?

• Tokenization
• B) Lemmatization
• C) Text normalization
• D) Support vector machine training

32. What does a dependency parser do?

• Identifies the parts of speech of words in a sentence


• B) Determines the relationships between words in a sentence
• C) Splits a sentence into individual words
• D) Assigns a root word to each sentence.
33. In a dependency parse tree, what does each edge represent?

• The syntactic category of a word


• B) The grammatical relation between two words
• C) The part-of-speech of a word
• D) The word’s frequency in the document

34. What is the main goal of parsing in natural language processing?

• a) To convert text into numerical form


• b) To identify the grammatical structure of a sentence
• c) To translate sentences between languages
• d) To extract named entities from the text

35. Which of the following is a common method for syntactic parsing in NLP?

• a) Rule-based parsing
• b) Neural network-based parsing
• c) Statistical parsing
• d) All of the above

36. Which algorithm is commonly used for context-free grammar (CFG) parsing?

• a) Dijkstra's algorithm
• b) Earley’s algorithm
• c) K-means algorithm
• d) Naive Bayes classifier

37. Which of the following is NOT a type of parsing strategy?

• a) Top-down parsing
• b) Bottom-up parsing
• c) Left-to-right parsing
• d) Forward parsing

38. What is the advantage of using dependency parsing over constituency parsing?

• a) It can handle ambiguous sentences better.


• b) It does not require a pre-defined grammar.
• c) It focuses on syntactic relations between words rather than hierarchical structure.
• d) It can be more easily interpreted by machines.

39. n dependency parsing, what does a 'head' refer to?

• a) The root word in a sentence


• b) A word that governs another word
• c) The subject of the sentence
• d) A word that is dependent on another word

40. What does a "parse tree" represent in the context of parsing?


• a) The phonetic structure of the sentence
• b) The semantic meaning of the sentence
• c) The syntactic structure of the sentence
• d) The lexicon used in the sentence

41. Which parser would be most suitable for handling ambiguous grammar in NLP?

• a) CYK parser
• b) LL parser
• c) Recursive descent parser
• d) Earley parser

Descriptive Questions (5 Marks)

1. What is parsing in the context of NLP, and why is it important?


2. Explain the difference between syntactic parsing and semantic parsing.
3. Describe the data collection phase in NLP?
4. What is lexical analysis, and what role does it play in the broader context of Natural
Language Processing?
5. Explain how NLP is applied in real-world applications.
6. Explain the concept of Natural Language Processing (NLP) and its importance in bridging
the gap between human communication and computer understanding.
7. What are some real-world applications of NLP in healthcare?
8. Discuss the role of NLP in chatbots and virtual assistants.
9. Explain the concept of syntactic parsing in NLP. Compare and contrast constituency parsing
and dependency parsing
10. What is a context-free grammar (CFG)? Explain its components and how it is used to
describe the syntax of a language.
11. What is the role of semantic analysis in NLP?
12. What is the role of lexical analysis in NLP?
13. What is lemmatization in NLP, and how does it differ from stemming?
14. What is the role of syntactic analysis in NLP?
15. Describe the concept of a parse tree in NLP. How does a parse tree represent the syntactic
structure of a sentence
16. Describe the process of Tokenization.
17. Define Lemmatization and Compare it with Stemming
18. Differentiate Between Syntax and Semantics in NLP.
19. Explain the Concept of a Parse Tree in Syntactic Analysis.
20. What is Constituency Parsing?

Descriptive Questions (15 Marks)

1. Describe the complete NLP pipeline and explain the significance of each step in
transforming raw text into meaningful insights.

2. Examine the importance of parsing in NLP. What role does it play in understanding the
syntactic structure of sentences?

3. Discuss the key concepts of Natural Language Processing (NLP)?


4. Describe the various applications of Natural Language Processing (NLP) in the real
world.
5. Explain the concept of context-free grammar (CFG)?

6.Discuss how lexical analysis contributes to the overall NLP process.

7. How does dependency parsing differ from constituency parsing in terms of syntactic
analysis?

8. Discuss the Basic Steps in the NLP Pipeline and Explain the Role of Each Step in
Preprocessing Text Data.

9. Explain the Process of Tokenization in NLP and Discuss Its Importance in Preprocessing Text
Data.

10. What Are the Main Parsing Techniques Used in NLP? Discuss the Difference Between Top-
Down and Bottom-Up Parsing Approaches

QUESTION BANK FOR MODULE 4

SECTION A – MULTIPLE CHOICE QUESTIONS:

1. What is an expert system primarily designed to emulate?

o A) Human emotions
o B) Basic math calculations
o C) Decision-making abilities of a human expert
o D) Internet browsing
o Answer: C
2. In which field was MYCIN developed for use?

o A) Finance
o B) Medicine
o C) Engineering
o D) Education
o Answer: B
3. What component of an expert system stores domain-specific information?

o A) Inference Engine
o B) User Interface
o C) Knowledge Base
o D) Database
o Answer: C
4. What type of reasoning does the inference engine use to apply rules?
o A) Deductive and Inductive
o B) Forward and Backward Chaining
o C) Lateral and Vertical Thinking
o D) Probabilistic and Deterministic Reasoning
o Answer: B
5. Which of the following is NOT a component of an expert system?

o A) Knowledge Base
o B) Inference Engine
o C) Random Number Generator
o D) User Interface
o Answer: C
6. The primary function of the inference engine is to:

o A) Display output
o B) Apply rules from the knowledge base to make decisions
o C) Update the knowledge base
o D) Improve system design
o Answer: B
7. Which type of inference engine operates with certainty?

o A) Probabilistic
o B) Deterministic
o C) Adaptive
o D) Random
o Answer: B
8. What is the primary purpose of MYCIN?

o A) Computer configuration
o B) Medical diagnosis
o C) Financial planning
o D) Customer service automation
o Answer: B
9. In XCON, what process is used to match components and configurations?
o A) Probabilistic reasoning
o B) Forward chaining
o C) Machine learning
o D) Random sampling
o Answer: B
10. In which company was XCON developed?
o A) IBM
o B) Digital Equipment Corporation (DEC)
o C) Apple
o D) Microsoft
o Answer: B
11. A rule in a knowledge base is often written as:

o A) Statement-Conclusion
o B) Cause-Effect
o C) If-Then
o D) Input-Output
o Answer: C
12. What is the primary role of a user interface in an expert system?

o A) To process data
o B) To allow interaction between the user and the system
o C) To store data
o D) To analyze trends
o Answer: B
13. What does XCON’s inference engine ensure for the VAX system configuration?

o A) Accurate diagnosis
o B) Compatibility of components
o C) Cost reduction
o D) Easy maintenance
o Answer: B
14. Which type of chaining starts with a goal and works backward to find evidence?

o A) Forward chaining
o B) Side chaining
o C) Backward chaining
o D) Linear chaining
o Answer: C
15. Expert systems are widely used in which of the following fields?

o A) Medicine
o B) Finance
o C) Engineering
o D) All of the above
o Answer: D
16. What type of knowledge acquisition method includes observing experts in action?

o A) Protocol Analysis
o B) Interview
o C) Observation
o D) Case Study
o Answer: C
17. Genetic Algorithms are inspired by:

o A) Human reasoning
o B) Laws of physics
o C) Principles of natural selection
o D) Mathematical equations
o Answer: C
18. Which expert system was specifically developed for bacterial infection diagnosis?

o A) R1
o B) INTERNIST
o C) MYCIN
o D) CLIPS
o Answer: C
19. Neural networks are based on:

o A) Logical operators
o B) Human brain structure
o C) Simple calculations
o D) Physics laws
o Answer: B
20. Which component in an expert system provides explanations for its decisions?

o A) Knowledge Base
o B) Inference Engine
o C) Explanation Facility
o D) User Interface
o Answer: C
21. Which component of an expert system helps keep track of data during a session?
o A) User Interface
o B) Working Memory
o C) Knowledge Base
o D) Explanation Facility
o Answer: B
22. In expert systems, knowledge acquisition is important because it:
o A) Updates the system's software
o B) Gathers and organizes knowledge from experts
o C) Simplifies the user interface
o D) Enhances the display features
o Answer: B
23. Which type of inference engine would be suitable for applications needing high
accuracy and consistent outcomes?
o A) Probabilistic
o B) Deterministic
o C) Adaptive
o D) Heuristic
o Answer: B
24. What is a significant benefit of using expert systems in customer support?
o A) Providing varied answers to the same question
o B) Offering consistent and fast responses
o C) Reducing the need for training employees
o D) Automatically designing new products
o Answer: B
25. Which of the following is a reason why an expert system would be chosen over a human
expert?
o A) It can learn from its experiences
o B) It can provide expertise 24/7 without fatigue
o C) It can make ethical decisions
o D) It can build new knowledge independently
o Answer: B
26. What type of learning does a genetic algorithm typically represent?
o A) Supervised Learning
o B) Unsupervised Learning
o C) Evolutionary Learning
o D) Reinforcement Learning
o Answer: C
27. Which method is commonly used in expert systems to structure knowledge in a way
that machines can interpret?
o A) Flowcharts
o B) If-then rules
o C) Graphs
o D) Spreadsheets
o Answer: B
28. The term "domain-specific" in expert systems refers to:
o A) Focusing on one area of expertise, like medicine or finance
o B) Being able to apply knowledge to any subject
o C) Creating knowledge in multiple unrelated areas
o D) Avoiding the use of any specific terminology
o Answer: A
29. Which AI field benefits the most from using neural networks for complex pattern
recognition?
o A) Inventory management
o B) Image and speech recognition
o C) Financial planning
o D) Simple data sorting
o Answer: B
30. In a genetic algorithm, what is the main purpose of the mutation step?
o A) To enhance user interaction
o B) To introduce diversity within the population
o C) To train the user on system rules
o D) To simplify the solution
o Answer: B
31. In MYCIN, what system was introduced to handle uncertainty in diagnosis?
o A) Knowledge Graphs
o B) Fuzzy Logic
o C) Certainty Factors
o D) Bayesian Inference
o Answer: C
32. Which type of inference engine would be suitable for handling ambiguous information?
o A) Deterministic
o B) Probabilistic
o C) Heuristic
o D) Manual
o Answer: B
33. In expert systems, “knowledge acquisition” refers to:
o A) Gathering and structuring expert knowledge for the system
o B) Displaying output to the user
o C) Designing user-friendly interfaces
o D) Implementing inference rules
o Answer: A
34. How does a forward-chaining inference engine process information?
o A) It works backward from a conclusion
o B) It starts with given facts and applies rules to infer conclusions
o C) It combines random elements to find solutions
o D) It only uses deterministic rules
o Answer: B
35. The MYCIN system primarily relied on what type of rules?
o A) Stochastic rules
o B) Hard-coded algorithms
o C) If-then rules
o D) Statistical models
o Answer: C
36. In XCON, what happens if an additional cooling unit is required?
o A) The system will request an engineer to install it manually
o B) The system automatically adds it to the configuration
o C) The customer is notified for additional purchases
o D) The system cancels the configuration process
o Answer: B
37. Which feature makes MYCIN capable of explaining its reasoning?
o A) Backward chaining
o B) Explanation facility
o C) User interface
o D) Heuristic processing
o Answer: B
38. An expert system designed for financial fraud detection would likely use which
inference method?
o A) Deterministic
o B) Probabilistic
o C) Genetic Algorithm
o D) Neural Network
o Answer: B
39. What does the term "scalability" refer to in the context of expert systems?
o A) Ability to solve basic problems
o B) Extending expert system support across multiple locations or departments
o C) Limiting the system to one domain
o D) Reducing system size over time
o Answer: B
40. Which stage of knowledge acquisition involves organizing data into usable formats for
the system?
o A) Knowledge Structuring
o B) Knowledge Validation
o C) Knowledge Updating
o D) Data Analysis
o Answer: A
41. Which limitation prevented MYCIN from being clinically deployed in real healthcare
settings?
o A) Lack of accuracy
o B) Dependency on hardware
o C) Legal and ethical concerns
o D) Expensive maintenance
o Answer: C
42. In XCON, what would happen if a peripheral device was found incompatible with the
configuration?
o A) It would alert the engineer to replace or adjust the device.
o B) The system would remove the device automatically.
o C) The system would proceed without the device.
o D) It would attempt to modify the device settings.
o Answer: A
43. Certainty factors in MYCIN were used to:
o A) Guarantee accuracy of diagnosis
o B) Express the level of confidence in diagnoses
o C) Eliminate uncertain diagnoses
o D) Provide user interface feedback
o Answer: B

44. Which of the following describes backward chaining?

o A) Working from known facts to reach a conclusion


o B) Trying every possible rule to find a solution
o C) Starting with a goal and working backward to find supporting evidence
o D) Using probability to reach a decision
o Answer: C

45, In an expert system, the “explanation facility” enhances which key aspect?

o A) Complexity
o B) Transparency
o C) Speed
o D) Automation
o Answer: B

46. Which knowledge acquisition method involves experts verbalizing their thought process as
they solve a problem?

o A) Document Analysis
o B) Observation
o C) Protocol Analysis
o D) Case Study
o Answer: C
SECTION B – FIVE MARKS QUESTIONS:

(Understanding Level: 1-3, Application Level: 4-6, Analysis Level: 7,8, Evaluation Level: 9,
10)
1. Explain the primary purpose of an expert system and provide examples of fields where
expert systems are commonly used.
2. Describe the components of an expert system and briefly explain the role each
component plays in the system’s functionality.
3. What is the difference between deterministic and probabilistic inference engines in
expert systems? Provide examples of scenarios where each might be used.

4. Using MYCIN as a case study, explain how a knowledge base and inference engine
work together to assist with medical diagnosis.
5. Apply the concept of backward chaining in expert systems to a real-world problem (e.g.
diagnosing issues in a vehicle).
6. Illustrate how genetic algorithms mimic the process of natural selection.

7. Analyze the role of the explanation facility in expert systems and discuss its importance
in building trust among users.
8. Identify whether chatGPT is an expert system. Justify your reasons.

9. Discuss the strengths and limitations of rule-based reasoning in expert systems. What
challenges might arise when applying rule-based systems to complex, real-world
problems?
9. Evaluate the effectiveness of learning automation in AI by comparing traditional expert
systems with modern AI models that can adapt and update from new data. Discuss the
benefits and limitations of each approach.
10. Critically assess the ethical and practical challenges of deploying expert systems in
high-stakes fields, such as healthcare. What considerations must be addressed to ensure
these systems are reliable and trustworthy?

SECTION C – FIFTEEN MARKS QUESTIONS:

1. Explain in detail the components of an expert system. Describe the function of each
component and provide examples of how they work together to make decisions.
(Bloom’s Level: Understanding)
2. Describe the process of knowledge acquisition in expert systems. Discuss different methods used to
gather expert knowledge and explain how this knowledge is structuredwithin the system. (Bloom’s
Level: Understanding)
4. Describe the process of a genetic algorithm in detail. Explain each step and provide a practical
example of how a genetic algorithm could be used to solve a problem. (Bloom’s Level: Application)
5. Analyze the concept of learning by induction in AI. Explain how inductive learning derives general
rules from specific examples and discuss different methods of inductivelearning. Provide examples
of real-world problems where learning by induction wouldbe particularly effective. (Bloom’s Level:
Analysis)
5. Evaluate the role of neural networks in AI and discuss their impact on the field of machine learning.
Explain the structure of a neural network and the challenges neural networks face. (Bloom’s Level:
Evaluation)
6. Critically evaluate the role of expert systems in high-stakes fields like healthcare, finance, and
engineering. Discuss the ethical and practical challenges involved in deploying these systems,
including issues of trust, accountability, and the potential impact on human decision-making. What
measures could be taken to address these concerns and ensure reliable and fair outcomes? (Bloom’s
Level: Evaluation)

**********

You might also like