AI & ML

Question Bank

Unit 1
Q.1 What is the definition of Artificial Intelligence?
a) The ability of machines to perform tasks that typically require human intelligence.
b) The study of how computers can perform tasks that require human-like intelligence.
c) The simulation of human intelligence processes by machines.
d) All of the above.

Answer: D

Q.2 Which of the following is NOT a goal of Artificial Intelligence?

a) Reasoning
b) Creativity
c) Emotion
d) Learning

Answer: C

Q.3 What is a primary task of Artificial Intelligence?

a) Recognizing patterns in data
b) Solving complex mathematical problems
c) Controlling robotic systems
d) All of the above

Answer: D
Q.4 Which of the following is NOT a component of computation in AI?
a) Algorithms
b) Data structures
c) Computer hardware
d) Psychology

Answer: D

Q.5 Which field studies mental processes such as perception, memory, and decision-making?
a) Computer Science
b) Cognitive Science
c) Robotics
d) Linguistics

Answer: B

Q.6 Which term describes the process by which an AI system interprets sensory inputs?
a) Perception
b) Understanding
c) Action
 d) Inference

Answer: A

Q.7 What is the key difference between Artificial Intelligence (AI) and Machine Learning
(ML)?
a) AI refers to the broader concept of machines being able to carry out tasks in an
"intelligent" manner, while ML is a subset of AI that focuses on algorithms that enable
computers to learn from data.
b) ML refers to the broader concept of machines being able to carry out tasks in an
"intelligent" manner, while AI is a subset of ML that focuses on algorithms that enable
computers to learn from data.
c) AI and ML are synonymous terms and can be used interchangeably.
d) There is no difference between AI and ML.

Answer: A

Q.8 Which of the following is NOT an application of Artificial Intelligence (AI) and
Machine Learning (ML) in the real world?
a) Virtual assistants like Siri and Alexa
b) Autonomous vehicles
c) Social media platforms
d) Weather forecasting

Answer: C

Q.9 What is a characteristic of Deep Learning algorithms?

a) They require labeled data for training
b) They rely on shallow neural networks
c) They can automatically discover patterns from data
d) They are computationally less expensive than traditional machine learning algorithms

Answer: C

Q.10 Which field focuses on designing algorithms for making decisions based on incomplete
and uncertain information?
a) Operations Research
b) Control Theory
c) Game Theory
d) Probabilistic Reasoning
Answer: D
Q.11 Which of the following statements best defines Artificial Intelligence (AI)?
a) AI is the study of how to program computers to perform tasks that require human
intelligence.
b) AI is the simulation of human-like emotions in machines.
c) AI refers to the ability of machines to think and feel like humans.
d) AI encompasses the development of systems that can perform tasks that typically require
human intelligence.
Answer: D

Q.12 What is one of the primary goals of Artificial Intelligence (AI)?

a) To eliminate the need for human intelligence.
b) To develop systems that surpass human intelligence.
c) To create machines with human emotions.
d) To enhance human capabilities and solve complex problems.

Answer: D

Q.13 Which task of AI involves learning patterns and relationships from data?
a) Reasoning
b) Perception
c) Learning
d) Understanding

Answer: C

Q.14 In the context of AI, what does computation primarily involve?

a) Performing mathematical calculations
b) Manipulating and processing information
c) Studying human cognition
d) Interacting with the physical world

Answer: B
Q.15 Which field studies how humans perceive, think, and learn?
a) Linguistics
b) Robotics
c) Cognitive Science
d) Computer Science

Answer: C

Q.16 Which term in AI refers to the ability to comprehend and interpret information?
a) Perception
b) Understanding
c) Action
d) Inference

Answer: B

Q.17 What distinguishes Deep Learning from other forms of Machine Learning?
a) Deep Learning involves the use of shallow neural networks.
b) Deep Learning requires manual feature extraction.
c) Deep Learning can automatically learn hierarchical representations from data.
d) Deep Learning is not a subset of Machine Learning.

Answer: C
 Q.18 Which of the following is NOT an application of Artificial Intelligence or Machine
Learning?
a) Fraud detection in banking systems
b) Autonomous driving
c) Recipe recommendations on cooking websites
d) Satellite image analysis for weather forecasting

Answer: C

Q.19 What type of neural networks are commonly used in Deep Learning for image
recognition tasks?
a) Recurrent Neural Networks (RNNs)
b) Convolutional Neural Networks (CNNs)
c) Long Short-Term Memory (LSTM) networks
d) Radial Basis Function (RBF) networks

Answer: B

Q.20 Which field focuses on designing algorithms inspired by the structure and function of the
human brain?
a) Neuroscience
b) Cybernetics
c) Cognitive Psychology
d) Computational Neuroscience

Answer: D

Q.21 What is meant by Artificial Intelligence?

A. Artificial intelligence is defined as a field aiming to make humans more intelligent.
B. Artificial intelligence is defined as a field aiming to improve security.
C. Artificial intelligence is defined as a field aiming to mine the data.
D. Artificial intelligence is defined as a field aiming to develop intelligent machines.
Answer D

Q.22 Which is considered the branch of Artificial Intelligence?

A. Cyber Forensics
B. Machine Learning
C. Network Design
D. Full-stack Developer
Answer B
Q.23 ____ is the common language for AI.
A. Lisp
B. Python
C. PHP
D. Java
Answer B
Q.24 ____ is a component of AI.
A. Training
 B. Designing
C. Learning
D. Puzzling
Answer C

Q.25 To build an AI system Which area can contribute

A. Sociology
B. Philosophy
C. Biology
D. All of the above
Answer D

Q.26 To build an intelligent system Which area can not contribute

A. Computer Science
B. Neuron science
C. Maths
D. Geology
Answer D

Q.27 Which is the application of AI

A. Vision Systems
B. Gaming
C. Expert Systems
D. All of the above
Answer D

Q.28 Goal of AI is
A. replication of human intelligence
B. solving problems that require knowledge
C. provide advise to user
D. all of above
Answer D

Q.29 Which is an advantage of AI

A. Accuracy is high
B. error is low
C. reliability is more
D. all of above
Answer D

Q.30 Which is disadvantage of AI

A. Accuracy is high
B. error is low
C. no feeling & emotions
D. all of above
Answer c
Q.31 Which of the following is an expansion of Artificial Intelligence application
A. Game Playing
B. Planning and Scheduling
C. Diagnosis
D. All of the mentioned
ANSWER D

Q.32 The ability to understand some thing is

A. Understanding
B. Perception
C. Psychology
D. Cognitive science
Answer B
Q.33 Accuracy is defined as
A. the percentage of correct predictions for the test data
B. the percentage of correct predictions for the training data
C. both (a) and (b)
D. none
Answer A

Q.34 Precision is defined as

A. the fraction of relevant examples (true positives) among some of the examples which
were predicted to belong in a certain class.

B. the fraction of relevant examples (true positives) among all of the examples which were
predicted to belong in a certain class.
C. both (a) and (b)
D. none
Answer B
Q.35 Which of the following is not an application of AI?
A. Intelligent Robots
B. Handwriting Recognition
C. Speech Recognition
D. Content mining
Answer D

Unit 2
Q.1 Which of the following is true about Hill Climbing?
A) It always guarantees finding the global
optimum.
B) It can get stuck in local optima.
C) It is a complete search algorithm.
D) It performs backtracking.

Answer: B
Q.2 The main drawback of Breadth-First Search (BFS) is:
A) It may get stuck in infinite loops.
B) It requires more memory.
C) It doesn't guarantee completeness.
D) It is not optimal

Answer: B

Q.3 What is the purpose of an activation function in a neural network?

a) To normalize input data
b) To introduce non-linearity
c) To reduce overfitting
d) To increase computational efficiency

Answer: B

Q.4 Imagine you are an employee at a Cola factory. Your goal is to present a comprehensive
analysis of our customer base to the company's CEO, highlighting the diverse groups of
individuals who purchase our product. To achieve this, you employ a machine learning
approach to categorize customers into 10 distinct groups based on various characteristics.
What type of machine learning method would be most suitable for this task?
a) Supervised Learning
b) Unsupervised Learning
c) Reinforcement Learning
d) Semi-Supervised Learning

Answer: B

Q.5 If a neural network has an input layer with 10 neurons, a hidden layer with 20 neurons,
and an output layer with 5 neurons, how many parameters (weights and biases) are there in
total, assuming each neuron is connected to every neuron in the subsequent layer?
a) 320
b) 322
c) 324
d) 325

Answer: D

Q.6 If the activation function used in a neural network is the Rectified Linear Unit (ReLU),
and the input to a neuron is -5, what will be the output of the neuron?
a) -5
b) 5
c) 0
d) 2

Answer: C

Q.7 What is the function of greedy best-first search?

a) f(n)
!= h(n)
b) f(n) < h(n)
c) f(n) = h(n)
d) f(n) > h(n)

Answer: C

Q.8 In Best First Search, if the heuristic function values of the nodes in the fringe are [10, 12,
14, 9], which node will be expanded next?
a) 10
b) 12
c) 14
d) 9

Answer: D

Q.9 Heuristic search algorithms are designed to:

A) Always find the optimal solution.
B) Minimize the number of nodes expanded.
C) Use domain-specific knowledge to guide search.
D) Randomly explore the search space.

Answer: C

Q.10 Which of the following is an advantage of Depth-First Search (DFS)?

A) It always finds the shortest path to the
goal.
B) It requires less memory.
C) It guarantees optimality.
D) It expands the shallowest node first.

Answer: B
Q.11 Which of the following statements about the learning rate in neural network training is
true?
a) A higher learning rate always leads to faster convergence
b) A lower learning rate can help prevent overshooting the minimum of
the loss function
c) The learning rate is independent of the number of epochs during
training
d) The learning rate does not affect the stability of the training process

Answer: B

Q.12 You wanted to predict the next Cooking show's winner. You have the data which include
the judge's comment's about the contestant's food who are participating and the ratings of their
previous cooked food which were given by the judges. Now your task is to predict the ratings
of the same chef's based on the new comments and dishes and decide who can win the title.
What type of machine learning method would be suitable for this task?
a) Supervised Learning
b) Unsupervised Learning
 c) Reinforcement Learning
d) Semi-Supervised Learning

Answer: A

Q.13 Suppose you have a neural network with three layers: an input layer with 50 neurons, a
hidden layer with 100 neurons, and an output layer with 10 neurons. If each neuron is
connected to every neuron in the subsequent layer and the weights are initialized randomly
between -1 and 1, how many weights are there in total in the network?
a) 5000
b) 6500
c) 6000
d) 5500

Answer: C

Q.14 Consider a neural network with a sigmoid activation function. If the input to a neuron is
x=0.8, what will be the output of the neuron?
a) 0.69
b) 0.48
c) 0.88
d) 0.23

Answer: A

Q.15 What is the function of uniform-cost search?

a) f(n) != g(n)
b) f(n) < g(n)
c) f(n) = g(n)
d) f(n) > g(n)

Answer: C
Q.16 In Best First Search, if the heuristic function
values of the nodes in the fringe are [12, 8, 15, 10], which node will be
expanded next?
a) 12
b) 8
c) 15
d) 10

Answer: B

Q.17 A production system consist of

A. a set of rules
B. knowledge that contain whatever information is appropriate for the particular task
C. a control strategy
D. all of above
Answer D

Q.18 A state space is

A. that contain all the possible configurations of the irrelevant objects.
B. that contain some possible configurations of the relevant objects.
C. that contain all the possible configurations of the relevant objects.
D. none
Answer C

Q.19 in uninformed search

A. without knowing the details we have to find the solution.
B. knowing the details we have to find the solution.
C. without knowing the details we can not find the solution.
D. none
Answer A

Q.20 Heuristic function h(n) gives

A. an estimation on the cost of getting from node ‘n’ to the goal state.
B. an estimation on the cost of getting from goal state to the node ‘n’ .
C. both (a) and (b)
D. none
Answer A

Q.21 in admissible heuristic search

A. the Heuristic function overestimates the cost of reaching the goal.
B. the Heuristic function some time overestimates the cost of reaching the goal.
C. the Heuristic function never overestimates the cost of reaching the goal.
D. none
Answer C

Q.22 which is not correct for hill climbing search

A. it is a local search algorithm.
B. greedy algorithm
C. backtracking is possible
D. backtracking is not possible
Answer C

Q.23 which is limitation of hill climbing search

A. local maxima
B. plateau
C. ridge
D. all of above
Answer D

Q.24 Breadth first search algorithm is

A. one of the complex search
 B. one of the simplest search
C. may be simplest or may be complex search
D. none
Answer B

Q.25 If there is a solution …………is guaranteed to find it

A. DFS
B. BFS
C. DFS and BFS
D. hill climbing
Answer B

Q.26 BFS algorithm cannot be effectively used unless the search space is ……..
A. quite large
B. large
C. quite small
D. none
Answer C
Q.27 Blind Search is used for which of the mentioned situations?
A. Advanced Game Theory
B. Real-life Simulation
C. Small Search Space
D. None of the above
Answer C
Q.28 ____ is the informed search method.
A. Memory Bound Heuristic Search
B. A * Search
C. Best First Search
D. All of the above
Answer D
Q.29 A ……... is a representation of problem elements at a given moment.
A. State Space
B. State
C. Problem
D. Search
Answer B

Q.30 ……… specifies the order in which the rules will be compared to the database.
A. One or more knowledge
B. A rule applier
C. A control strategy
D. A set of rules
Answer C

Q.31 ……… is the computational system that implements the control strategy and applies the
rules.
A. One or more knowledge
B. A rule applier
C. A control strategy
D. A set of rules
Answer B

Q.32 Why do we need biological neural networks?

A. to solve tasks like machine vision & natural language processing
B. to apply heuristic search methods to find solutions of problem
C. to make smart human interactive & user friendly system
D. all of the mentioned
Answer D

Q.33 What is plasticity in neural networks?

A. input pattern keeps on changing
B. input pattern has become static
C. output pattern keeps on changing
D. output is static
Answer A

Q.34 The fundamental unit of ANN is

A. brain
B. nucleus
C. neuron
D. axon
Answer c

Q.35 What are dendrites?

A. fibers of nerves
B. nuclear projections
C. other name for nucleus
D. none of the mentioned
Answer A

Q.36 Function of dendrites is?

A. Receptors
B. Transmitter
C. both receptor & transmitter
D. none of the mentioned
Answer A
Q.37 The neuron consists
A. cell body
B. dendrites
C. axon
D. all of the Above
Answer D
Q.38 Which search strategy is also called as blind search?
a) Uninformed search
b) Informed search
c) Simple reflex search
d) All of the mentioned
Answer A

Q.39 Which search is implemented with an empty first-in-first-out queue?

a) Depth-first search
b) Breadth-first search
c) Bidirectional search
d) None of the mentioned
Answer B

Unit 3

Q.1 The output of training process in machine learning is________________

A. machine learning algorithm
B. machine learning model
C. null
D. accuracy
Answer B

Q.2 Which of the following techniques would perform better for reducing dimensions of a data
set?
A. removing columns which have high variance in data
B. removing columns which have too many missing value
C. removing columns with dissimilar data trends
D. None of the above
Answer B
Q.3 Which of the following is a type of neural network?
A. Decision tree
B. Random forest
C. Convolutional neural network
D. Linear regression
Answer c
Q.4 What is the purpose of cross-validation in machine learning?
A. To evaluate the performance of a model on a held-out test set
B. To evaluate the performance of a model on different subsets of the data
C. To compare the performance of different models
D. To tune the hyperparameters of a model
Answer B

Q.5 What is the purpose of the confusion matrix in machine learning?

A. To visualize the distribution of the data in a dataset
 B. To compare the performance of different models
C. To evaluate the performance of a classification model
D. To evaluate the performance of a regression model
Answer c

Q.6 What is the purpose of regularization in machine learning?

A. To prevent overfitting
B. To increase the accuracy of the model
C. To reduce the variance of the model
D. To reduce the bias of the model
Answer A
Q.7 Which of the following is an example of a deep learning architecture?
A. K-nearest neighbors (KNN)
B. Random forest
C. Convolutional neural network (CNN)
D. Gradient boosting machine (GBM)
Answer c

Q.8 Which of the following is a common activation function used in deep learning?
A. Sigmoid
B. Linear
C. Exponential
D. Quadratic
Answer A
Q.9 The goal of ML is to enable the machine to learn
A. from past experiences
B. to think without any human intervention
C. from present and past experiences
D. All of the above
Answer A
Q.10 Machine learning was first introduced by Arthur Samuel in
A. 1959
B. 1965
C. 1975
D. 1955
Answer A

Q.11 Machine learning system may be

A. Supervised
B. Unsupervised
C. semisupervised
D. all of above
Answer D

Q.12 What is a primary difference between traditional programming and machine learning?
 a) In traditional programming, rules are explicitly defined by programmers, while
in machine learning, rules are learned from data.
b) Traditional programming requires less computational power compared to
machine learning.
c) Machine learning always requires labeled data, whereas traditional
programming does not.
d) Traditional programming cannot handle large datasets efficiently.
Answer: a) In traditional programming, rules are explicitly defined by
programmers, while in machine learning, rules are learned from data.
Q.13 Which of the following is a key element of the machine learning process?

A. Debugging
B. Code compilation
C. Data Collection
D. Unit testing
Answer: c) Data Collection
Q.14 Which step of the machine learning process involves transforming raw data into a
format suitable for training a model?

A. Data Collection
B. Data Preparation
C. Model Training
D. Model Evaluation
Answer: b) Data Preparation
Q.15 What is the purpose of model selection in machine learning?

A. To choose the most complex model available

B. To minimize the computational resources required for training
C. To select the best algorithm and its hyperparameters for the given problem
D. To avoid using deep learning models
Answer: c) To select the best algorithm and its hyperparameters for the
given problem

Q.15 Which type of learning involves learning from a dataset with labeled examples?

A. Supervised Learning
B. Unsupervised Learning
C. Reinforcement Learning
D. Semi-supervised Learning
Answer: a) Supervised Learning
Q.16 In which type of problem does regression aim to predict continuous-valued outputs?

A. Image classification
B. Sentiment analysis
C. Stock price prediction
D. Email spam detection
 Answer: c) Stock price prediction
Q.17 Which of the following is an example of a classification problem?

A. Predicting the temperature for the next day

B. Identifying whether an email is spam or not
C. Predicting the sales volume of a product
D. Estimating the time to deliver a package
Answer: b) Identifying whether an email is spam or not
Q.18 What is the primary goal of model evaluation in machine learning?

A. To make the model as complex as possible

B. To ensure the model performs well on the training data
C. To assess how well the model generalizes to unseen data
D. To increase the bias of the model
Answer: c) To assess how well the model generalizes to unseen data
Q.19 Which step in the machine learning process involves fine-tuning the model to improve
its performance?

A. Data Collection
B. Model Selection
C. Model Evaluation
D. Evaluation and Optimization
Answer: d) Evaluation and Optimization
Q.20 What type of learning involves learning from feedback received as a result of actions
taken in an environment?

A. Supervised Learning
B. Unsupervised Learning
C. Reinforcement Learning
D. Semi-supervised Learning
Answer: c) Reinforcement Learning
Q.21 Which of the following statements best describes traditional programming?

A. Traditional programming requires labeled data for training.

B. Traditional programming involves learning rules from data.
C. Traditional programming relies on statistical techniques.
D. Traditional programming involves writing explicit instructions to perform a task.
Answer: d) Traditional programming involves writing explicit instructions to perform a
task.
Q.22 In machine learning, what does the term "representation" refer to?
A. It refers to how the model is trained.
B. It refers to the format of the input data.
C. It refers to the process of feature engineering.
D. It refers to the selection of the machine learning algorithm.
Answer: b) It refers to the format of the input data.
Q.23 Which of the following is a key component of data preparation in machine learning?
 A. Model Training
B. Model Selection
C. Feature Engineering
D. Model Evaluation
Answer: c) Feature Engineering
Q.24 Which type of learning does not require labeled data for training?

A. Supervised Learning
B. Unsupervised Learning
C. Reinforcement Learning
D. Semi-supervised Learning
Answer: b) Unsupervised Learning
Q.25 What is the primary objective of reinforcement learning?

A. Minimizing prediction errors

B. Maximizing the likelihood of the observed data
C. Learning from labeled examples
D. Maximizing cumulative rewards by taking actions in an environment
Answer: d) Maximizing cumulative rewards by taking actions in an environment
Q.26 Which of the following problems would be best suited for regression?

A. Customer segmentation
B. Sentiment analysis
C. Predicting the price of a house
D. Image classification
Answer: c) Predicting the price of a house
Q.27 In a classification problem, what is the output typically?

A. A continuous value
B. A category or class label
C. A probability score
D. A decision boundary
Answer: b) A category or class label
Q.28 What is the primary goal of model evaluation in machine learning?

A. Maximizing model complexity

B. Minimizing training time
C. Ensuring the model performs well on the training data
D. Assessing how well the model generalizes to unseen data
Answer: d) Assessing how well the model generalizes to unseen data
Q.29 What technique is used in model optimization to improve performance on unseen data?

A. Cross-validation
B. Overfitting
C. Feature selection
D. Hyperparameter tuning
Answer: d) Hyperparameter tuning
Q.30 Which type of learning involves learning from both labeled and unlabeled data?

A. Supervised Learning
B. Unsupervised Learning
C. Reinforcement Learning
D. Semi-supervised Learning
Answer: d) Semi-supervised Learning

Unit 4

Q.1 Following are the descriptive models________________

A. classification
B. clustering
C. association rule
D. Both (a) and (b)
Answer D

Q.2 You are given seismic data and you want to predict next earthquake , this is an example
of__________________
A. supervised learning
B. unsupervised learning
C. reinforcement learning
D. dimensionality reduction
Answer A
Q.3 which is a type of supervised Machine learning algorithm
A. Regression
B. Classification
C. Both (a) and (b)
D. None
Answer C
Q.4 Disadvantage of supervised learning
A. Have limited scope
B. Collecting labeled dataset is expensive and time consuming
C. Wrong prediction
D. All of above
Answer D
Q.5 Which of following is regression algorithm
A. Decision tree
B. Random Forest
C. Linear regression
D. All of above
Answer D

Q.6 Which of following is classification algorithm

A. Decision tree
B. Random Forest
 C. Logistic regression
D. All of above
Answer D
Q.7 Reason for Under fitting
A. The model has a high bias
B. The size of the training dataset used is not enough
C. The model is too simple
D. All of above
Answer D
Q.8 supervised learning
A. takes direct feedback.
B. do not take any feedback
C. takes indirect feedback
D. none
Answer A

Q. 9 A ________ problem is when the output variable is a category

A. clustering
B. reinforcement learning
C. semi supervised
D. classification
ANSWER D
Q.10 A ________ problem is when the output variable is a real value.
A. Regression
B. reinforcement learning
C. semi supervised
D. classification
ANSWER A
Q.11 Random Forest has _________ as base learning models
A. multiple decision trees
B. bagging
C. Entropy
D. none of these
ANSWER A
Q.12 What is the primary goal of linear regression?

A) Maximize the variance of the target variable

B) Minimize the sum of squared errors between predicted and actual values
C) Maximize the number of features in the dataset
D) Minimize the number of observations in the dataset
Answer: B
Q.13 How is the relationship between the independent and dependent variables represented in
linear regression?

A. Exponentially
B. Linearly
 C. Logarithmically
D. Quadratically
Answer: B
Q.14 What is the process of determining the coefficients in linear regression called?

A.Gradient descent
B.Feature extraction
C.Model selection
D.Model training
Answer: D
Q.15 Which of the following is NOT a common application of linear regression?

A. Stock price prediction

B. Sales forecasting
C. Image classification
D. House price prediction
Answer: C
Q.16 What are the performance parameters used to evaluate the performance of a linear
regression model?

A.Accuracy and precision

B.Mean squared error and R-squared
C.F1 score and recall
D.True positive rate and false positive rate
Answer: B
Q.17 Which of the following statements about linear regression implementation is true?

A. It can only handle categorical variables

B. It requires the data to be normally distributed
C. It works best for nonlinear relationships between variables
D. It involves finding the line that best fits the data points
Answer: D
Q.18 In linear regression, what does the term "intercept" refer to?

A. The slope of the regression line

B. The variability of the dependent variable
C. The value of the dependent variable when all independent variables are zero
D. The number of observations in the dataset
Answer: C
Q.19 Which of the following techniques is commonly used to find the optimal coefficients in
linear regression?

A. Principal Component Analysis (PCA)

B. Support Vector Machines (SVM)
C. Ordinary Least Squares (OLS)
 D. K-Means clustering
Answer: C
Q.20 What is the purpose of the cost function in linear regression?

A.To maximize the accuracy of predictions

B.To minimize the error between predicted and actual values
C.To maximize the number of features in the dataset
D.To minimize the complexity of the model
Answer: B
Q.21 Which of the following statements about the coefficients in linear regression is true?

A. They represent the intercept of the regression line only.

B. They are determined based on the mean values of the independent and dependent
variables.
C. They quantify the strength and direction of the relationship between independent
and dependent variables.
D. They are constant values and do not change during model training.
Answer: C
Q.22 In linear regression, what does multicollinearity refer to?
A. The presence of multiple dependent variables
B. The presence of multiple independent variables that are highly correlated
C. The presence of outliers in the dataset
D. The presence of non-linear relationships between variables
Answer: B
Q.23 What does the R-squared value indicate in linear regression?

A. The percentage of variance in the dependent variable explained by the independent

variables
B. The total number of data points in the dataset
C. The significance of the coefficients in the regression model
D. The mean squared error between predicted and actual values
Answer: A
Q.24 Which of the following is NOT a common assumption of linear regression?

A. Linearity: The relationship between independent and dependent variables is linear.

B. Homoscedasticity: The variance of the residuals is constant across all levels of
the independent variables.
C. Independence: Observations are independent of each other.
D. Normality: The residuals follow a normal distribution.
Answer: B
Q.25 What is the purpose of feature scaling in linear regression?

A. To increase the complexity of the model

B. To decrease the interpretability of the coefficients
 C. To standardize the scale of independent variables to facilitate model training
D. To introduce non-linearity into the relationship between variables
Answer: C
Q.26 Which of the following techniques can help address overfitting in linear regression?
A. Adding more features to the model
B. Removing outliers from the dataset
C. Regularization techniques such as Ridge or Lasso regression
D. Increasing the number of iterations during model training
Answer: C

Q.27 What is a decision tree classifier primarily used for?

A. Clustering
B. Classification
C. Regression
D. Dimensionality reduction
Answer: B
Q.28 What is the main difference between classification and regression trees?

A. Classification trees predict categorical outcomes, while regression trees

predict continuous outcomes.
B. Classification trees use the Gini index, while regression trees use information gain.
C. Classification trees have fewer decision nodes than regression trees.
D. Classification trees are binary trees, while regression trees can have multiple
splits. Answer: A
Q.29 Which metric is commonly used to measure the quality of a split in a decision tree based
on impurity reduction?

A.Accuracy
B.Gini index
C.Mean squared error
D.F1 score
Answer: B
Q.30 How does the Gini index measure impurity in a decision tree node?

A. It calculates the mean squared error between predicted and actual values.
B. It measures the probability of incorrect classification.
C. It quantifies the variability of the target variable.
D. It evaluates the homogeneity of class labels in a node.
Answer: D
Q.31 What does the Information Gain measure in a decision tree?

A. The reduction in entropy after a split

B. The improvement in accuracy after a split
C. The difference in mean squared error before and after a split
 D. The increase in Gini index after a split
Answer: A

Unit 5

Q.1 The problem of finding hidden structure in unlabeled data is called______________

A. unsupervised learning
B. reinforcement learning
C. supervised learning
D. None
Answer A

Q.2 Which learning Requires Self-Assessment to identify patterns within data?

A. supervised learning
B. unsupervised learning
C. semi supervised learning
D. reinforced learning
Answer B
Q.3 Which of the following is an example of a clustering algorithm?
A. Decision tree
B. Random forest
C. K-means
D. Gradient descent
Answer c
Q.4 which is a type of unsupervised Machine learning algorithm
A. clustering
B. association
C. Both (a) and (b)
D. None
Answer C
Q.5 Disadvantage of unsupervised learning
A. Less accuracy
B. Time consuming
C. More the complexity
D. All of above
Answer D
Q. 6 In ________training model has only input parameter values.
A. supervised learning
B. Unsupervised learning
C. reinforcement learning
D. None of these
ANSWER B

Q. 7 In _________little bit labelled and rest large portion of it is unlabeled.

A. Clustering
B. Association
 C. semi supervised
D. Reinforcement
ANSWER C

Q. 8 In _________ model keeps on increasing its performance using a Reward Feedback to learn
the behavior or pattern
A. Clustering
B. reinforcement learning
C. semi supervised
D. reinforcement
ANSWER B

Q.9 ______is Less accurate.

A. Unsupervised learning
B. reinforcement learning
C. semi supervised
D. classification
ANSWER A
Q. 10 What is the primary goal of unsupervised learning?

Minimize error between predicted and actual outputs

A)
Discover patterns or structures in input data
B)
Classify data into predefined categories
C)
Predict future outcomes based on historical data
D)
Answer: B
Q.11 Which algorithm is commonly used for clustering in unsupervised learning?

A. Decision Trees
B. Support Vector Machines
C. K-Means
D. Random Forest
Answer: C) K-Means
Q.13 Which of the following tasks is NOT typically associated with unsupervised learning?

A. Anomaly detection
B. Dimensionality reduction
C. Classification
D. Clustering
Answer: C) Classification
Q.14 What is a common application of Principal Component Analysis (PCA) in
unsupervised learning?

A. Data labeling
B. Outlier detection
C. Feature extraction and dimensionality reduction
D. Classifying data points into categories
 Answer: C) Feature extraction and dimensionality reduction
Q.15 Which evaluation metric is often used to assess the quality of clustering algorithms?

A. F1 Score
B. Accuracy
C. Silhouette Score
D. Mean Absolute Error
Answer: C) Silhouette Score
Q.16 Which of the following is NOT a type of clustering algorithm?
A. Hierarchical Clustering
B. K-Means Clustering
C. DBSCAN
D. Decision Tree
Answer: D) Decision Tree
Q.16 Which unsupervised learning technique is suitable for detecting outliers in a dataset?

A. K-Means clustering
B. PCA
C. Isolation Forest
D. Linear Regression
Answer: C) Isolation Forest

Q.17 What does the "elbow method" typically help determine in K-Means clustering?
A. The optimal number of clusters
B. The best feature to use for clustering
C. The outliers in the dataset
D. The distance between data points
Answer: A) The optimal number of clusters
Q.18 Which clustering algorithm is known for its ability to handle clusters of arbitrary shapes
and sizes?

A. K-Means
B. DBSCAN
C. Agglomerative Hierarchical Clustering
D. Divisive Clustering
Answer: B) DBSCAN
Q.19 What does DBSCAN stand for?

A. Density-Based Spatial Clustering of Applications with Noise

B. Distribution-Based Spectral Clustering with Accuracy
C. Divisive Binary Search Clustering for Analyzing Networks
D. Density-Boosted Sequential Classification with Neural Networks
Answer: A) Density-Based Spatial Clustering of Applications with Noise
Q.20 Which of the following is a characteristic of DBSCAN?
A. It requires the number of clusters to be specified in advance.
 B. It assigns each data point to the nearest centroid.
C. It is sensitive to the order of data points.
D. It can identify outliers as noise.
Answer: D) It can identify outliers as noise.
Q.21 Which clustering algorithm is based on the concept of "density reachability" and
"density connectivity"?

A. K-Means
B. Agglomerative Hierarchical Clustering
C. DBSCAN
D. Distribution Model-Based Clustering
Answer: C) DBSCAN
Q.22 What does the acronym "DB" in DBSCAN refer to?

A. Density-Based
B. Distribution-Based
C. Divisive-Based
D. Distance-Based
Answer: A) Density-Based
Q.23 Which clustering algorithm forms clusters by merging or dividing them based on
their distance or similarity?

A. DBSCAN
B. Distribution Model-Based Clustering
C. Agglomerative Hierarchical Clustering
D. Divisive Clustering
Answer: C) Agglomerative Hierarchical Clustering
Q.23 In hierarchical clustering, what is the agglomerative approach?

A. Starting with each point as a separate cluster and then merging them iteratively.
B. Starting with one cluster containing all points and then dividing it into smaller clusters.
C. Assigning each point to the nearest centroid.
D. Dividing the dataset into equal-sized partitions.
Answer: A) Starting with each point as a separate cluster and then merging them iteratively.
Q.24 Which of the following clustering algorithms does not require the number of clusters to
be specified beforehand?

A. K-Means
B. Distribution Model-Based Clustering
C. Agglomerative Hierarchical Clustering
D. DBSCAN
Answer: D) DBSCAN

Q.25 Which clustering algorithm is based on the concept of iteratively partitioning the data
into clusters until convergence?
 A.
K-Means
B.
DBSCAN
C.
Agglomerative Hierarchical Clustering
D.
Distribution Model-Based Clustering
Answer: A) K-Means
Q.26 Which clustering algorithm is sensitive to the initial placement of centroids?
A. DBSCAN
B. Agglomerative Hierarchical Clustering
C. K-Means
D. Divisive Clustering
Answer: C) K-Means
Q.27 Which clustering algorithm is based on modeling the distribution of data points in the
feature space?

A.
K-Means
B.
DBSCAN
C.
Distribution Model-Based Clustering
D.
Agglomerative Hierarchical Clustering
Answer: C) Distribution Model-Based Clustering
Q.28 In hierarchical clustering, which approach starts with one cluster containing all points
and then divides it into smaller clusters?

A.
Agglomerative
B.
Divisive
C.
DBSCAN
D.
K-Means
Answer: B) Divisive
Q.29 What is the main disadvantage of agglomerative hierarchical clustering?
A. It requires a predefined number of clusters.
B. It is computationally expensive for large datasets.
C. It cannot handle clusters of arbitrary shapes.
D. It is sensitive to outliers.
Answer: B) It is computationally expensive for large datasets.
Q.30 Which clustering algorithm is based on the concept of partitioning the data into
spherical clusters?

A. DBSCAN
B. Agglomerative Hierarchical Clustering
C. K-Means
D. Distribution Model-Based Clustering
Answer: C) K-Means
Q.31 Which clustering algorithm forms clusters by continuously merging the nearest clusters
until a stopping criterion is met?

A. K-Means
B. DBSCAN
 C. Agglomerative Hierarchical Clustering
D. Distribution Model-Based Clustering
Answer: C) Agglomerative Hierarchical Clustering

Q.32 Which distance measure calculates the distance between two points as the sum of the
absolute differences of their coordinates?

A. Euclidean Distance
B. Manhattan Distance
C. Cosine Similarity
D. Minkowski Distance
Answer: B) Manhattan Distance
Q.33 Which distance measure is often used for text mining and document clustering, where
the magnitude of the vectors is not important?

A. Euclidean Distance
B. Manhattan Distance
C. Cosine Similarity
D. Hamming Distance
Answer: C) Cosine Similarity
Q34 What is the range of values for Euclidean distance?

A. [0, ∞)
B. (-∞, ∞)
C. [0, 1]
D. [0, n]
Answer: A) [0, ∞)
Q.35 Which distance measure is suitable for data with binary attributes?

A. Euclidean Distance
B. Manhattan Distance
C. Cosine Similarity
D. Hamming Distance
Answer: D) Hamming Distance
Q.36 Which distance measure is also known as the L2 norm?

A. Euclidean Distance
B. Manhattan Distance
C. Cosine Similarity
D. Hamming Distance
Answer: A) Euclidean Distance
Q.37 Which distance measure considers both the magnitude and orientation of vectors, often
used in recommendation systems and text mining?

A. Euclidean Distance
B. Manhattan Distance
C. Cosine Similarity
 D. Hamming Distance
Answer: C) Cosine Similarity
Q.38 Which distance measure is also known as the city block distance or taxicab metric?
A. Euclidean Distance
B. Manhattan Distance
C. Cosine Similarity
D. Minkowski Distance
Answer: B) Manhattan Distance
Q.39 Which distance measure generalizes both Euclidean and Manhattan distances based
on a parameter p?

A. Euclidean Distance
B. Manhattan Distance
C. Cosine Similarity
D. Minkowski Distance
Answer: D) Minkowski Distance
Q.40 What is the primary goal of the K-means clustering algorithm?

A. Minimize the within-cluster variance

B. Maximize the between-cluster variance
C. Minimize the number of clusters
D. Maximize the number of clusters
Answer: A) Minimize the within-cluster variance