DATA SCIENCE & ARTIFICIAL INTELLIGENCE

01 Machine Learning-III
Learning-III

1. What is Machine Learning? (c) The model is too complex and memorizes
(a) A type of computer programming the training data
(b) A branch of artificial intelligence (d) The model is biased towards certain features
(c) A hardware system
7. Which of the following is an unsupervised
(d) A type of software development
learning algorithm used for clustering?
2. Which of the following is not a type of Machine (a) Support Vector Machines (SVM)
Learning? (b) K-Means
(a) Supervised Learning (c) Decision Trees
(b) Unsupervised Learning (d) Naive Bayes
(c) Reinforcement Learning 8. What is the purpose of the activation function in
(d) Deterministic Learning a neural network?
3. What is the primary goal of Supervised (a) To initialize weights
Learning? (b) To introduce non-linearity
(a) Minimize prediction error (c) To calculate the loss
(b) Discover hidden patterns in data (d) To reduce dimensionality
(c) Optimize decision-making processes 9. What is the main advantage of using dropout in
(d) Classify data into predefined categories neural networks?
(a) Reducing computation time
4. Which algorithm is commonly used for linear
(b) Preventing overfitting
regression in Machine Learning?
(c) Improving convergence speed
(a) Decision Trees (b) K-Means
(d) Enhancing interpretability
(c) Random Forest (d) Linear Regression
10. In a classification problem, what does precision
5. In Machine Learning, what is the purpose of a
measure?
confusion matrix?
(a) The proportion of true positives
(a) Evaluate model accuracy
(b) The proportion of true negatives
(b) Visualize decision boundaries
(c) The ratio of true positives to false positives
(c) Represent feature importance
(d) The ratio of true positives to true negatives
(d) Measure prediction error
11. What is the purpose of the learning rate in
6. What does “overfitting” mean in the context of
gradient descent optimization?
Machine Learning?
(a) Control the speed of convergence
(a) The model is too simple and doesn't capture (b) Determine the number of iterations
the underlying patterns (c) Set the threshold for model accuracy
(b) The model performs well on the training data (d) Define the number of layers in a neural
but poorly on new data network

© Copyright www.madeeasy.in
2 CPQ 2025 • Data Science & Artificial Intelligence

12. Which algorithm is suitable for both regression (b) It transforms the input data into a higher-
and classification problems? dimensional space
(a) Decision Trees (c) It controls the learning rate of the model
(b) K-Means (d) It measures the margin of the decision
(c) Naive Bayes boundary
(d) Support Vector Machines (SVM) 18. What is the primary use of the F1 score in
13. What is the curse of dimensionality in Machine classification tasks?
Learning? (a) Evaluating model accuracy
(a) High-dimensional datasets are (b) Balancing precision and recall
computationally cheaper to process (c) Measuring the area under the ROC curve
(d) Assessing feature importance
(b) Increased dimensionality improves model
accuracy 19. What does the term “bias” refer to in the context
(c) Increased dimensionality leads to sparsity of Machine Learning?
and overfitting (a) The model’s ability to generalize to new data
(d) It refers to the curse of low-dimensional data (b) The error introduced by approximating a
real-world problem
14. What is the purpose of cross-validation in
(c) The tendency of a model to perform well on
Machine Learning?
the training set but poorly on the test set
(a) Evaluating model performance on the (d) The ability of a model to capture complex
training set patterns in data
(b) Testing the model on unseen data
(c) Reducing overfitting by training on multiple 20. Which optimization algorithm is commonly used
subsets of the data for training deep neural networks?
(d) Preprocessing data before training a model (a) Random Search (b) Genetic Algorithm
(c) Adam (d) Naive Bayes
15. Which technique is commonly used for feature
21. What is the purpose of the “ReLU” activation
scaling in Machine Learning?
function in neural networks?
(a) Z-score normalization
(a) Introduce non-linearity
(b) One-hot encoding
(b) Ensure numerical stability
(c) Principal Component Analysis (PCA)
(c) Normalize input data
(d) Support Vector Machines (SVM)
(d) Define the learning rate
16. What is the key difference between bagging and
22. Which of the following is a hyperparameter in
boosting?
the k-nearest neighbors (KNN) algorithm?
(a) Bagging combines weak learners, while (a) Number of clusters
boosting focuses on strong learners (b) Learning rate
(b) Bagging reduces model complexity, while (c) Number of neighbors (k)
boosting increases it (d) Decision threshold
(c) Bagging aims to improve accuracy, while
boosting focuses on reducing bias 23. What is the Kullback-Leibler (KL) divergence
(d) Bagging and boosting are two terms for the used for in Machine Learning?
same technique (a) Clustering
(b) Feature selection
17. What is the purpose of the term “Kernel” in (c) Measuring the difference between
Support Vector Machines (SVM)? probability distributions
(a) It defines the number of support vectors (d) Dimensionality reduction

www.madeeasy.in © Copyright
 Machine Learning 3

24. What is the purpose of regularization in Machine (a) Naive Bayes

Learning? (b) K-Means
(a) Improve model interpretability (c) Decision Trees
(b) Reduce model complexity and prevent (d) Recurrent Neural Networks (RNN)
overfitting
31. In the context of clustering, what does the term
(c) Increase the number of features in the model
“centroid” refer to?
(d) Boost the accuracy of the training set
(a) A representative point of a cluster
25. Which evaluation metric is suitable for (b) The center of the dataset
imbalanced classification problems? (c) A measure of feature importance
(a) Accuracy (d) The output of a neural network layer
(b) Precision
32. What is the purpose of the term “dropout” in
(c) F1 score
neural networks?
(d) Mean Squared Error (MSE)
(a) Eliminate outliers from the training set
26. What is the role of the term “momentum” in (b) Regularize the model by randomly dropping
gradient descent optimization? neurons during training
(a) Control the learning rate (c) Adjust learning rates dynamically
(b) Accelerate convergence by introducing (d) Enhance the interpretability of the model
inertia
33. Which technique is commonly used for handling
(c) Regularize the model
missing data in Machine Learning?
(d) Define the loss function
(a) Mean imputation (b) Mode imputation
27. What is the purpose of a ROC curve in Machine (c) Median imputation (d) All of the above
Learning?
34. What is the purpose of a hyperparameter in the
(a) Visualizing the trade-off between sensitivity
context of Machine Learning?
and specificity
(a) It is learned from the training data
(b) Evaluating model accuracy
(b) It is a parameter automatically adjusted by
(c) Determining the optimal learning rate
the model during training
(d) Selecting features for the model
(c) It is set before training and influences the
28. Which algorithm is commonly used for image learning process
classification tasks? (d) It is a measure of model performance
(a) K-Means
35. Which of the following is an ensemble learning
(b) Decision Trees
method?
(c) Convolutional Neural Networks (CNN)
(a) Linear Regression
(d) Principal Component Analysis (PCA)
(b) Support Vector Machines (SVM)
29. What is the primary objective of reinforcement (c) Random Forest
learning? (d) K-Nearest Neighbors (KNN)
(a) Minimize prediction error
36. What is the objective of the Expectation-
(b) Discover hidden patterns in data
Maximization (EM) algorithm?
(c) Optimize decision-making processes
(a) Minimize prediction error
(d) Maximize cumulative reward over time
(b) Optimize clustering assignments in
30. Which algorithm is commonly used for natural unsupervised learning
language processing tasks, such as text (c) Improve model interpretability
classification? (d) Adjust learning rates dynamically

© Copyright www.madeeasy.in
4 CPQ 2025 • Data Science & Artificial Intelligence

37. Which of the following is an example of a semi- (c) A node with no children
supervised learning technique? (d) A node with more than two children
(a) Decision Trees
44. Which algorithm is commonly used for
(b) K-Means
dimensionality reduction?
(c) Support Vector Machines (SVM)
(a) Decision Trees
(d) Self-training
(b) Principal Component Analysis (PCA)
38. What is the primary purpose of the term “batch (c) Naive Bayes
size” in training neural networks? (d) Random Forest
(a) Control the number of epochs
45. What is the purpose of the term “dropout” in
(b) Define the number of layers in the network
neural networks?
(c) Determine the amount of data processed in
(a) Regularize the model by randomly dropping
each iteration
neurons during training
(d) Set the learning rate
(b) Increase the model's capacity
39. In which scenario would you use the Kullback- (c) Reduce training time
Leibler (KL) divergence? (d) Enhance model interpretability
(a) Image classification
46. Which optimization algorithm is commonly used
(b) Text generation
for training deep neural networks?
(c) Reinforcement learning
(a) Stochastic Gradient Descent (SGD)
(d) Clustering
(b) Genetic Algorithm
40. What is the purpose of the term “bagging” in (c) AdaBoost
ensemble learning? (d) Random Search
(a) Combine predictions from multiple models
47. What is the purpose of regularization in Machine
to reduce variance
Learning?
(b) Increase model complexity to capture
(a) Improve model interpretability
intricate patterns
(b) Reduce model complexity and prevent
(c) Introduce non-linearity in the decision
overfitting
boundary
(c) Increase the number of features in the model
(d) Speed up the training process
(d) Boost the accuracy of the training set
41. Which of the following is a distance metric
48. Which evaluation metric is suitable for
commonly used in k-nearest neighbors (KNN)?
imbalanced classification problems?
(a) Hamming distance (b) Cosine similarity
(a) Accuracy
(c) Jaccard similarity (d) All of the above
(b) Precision
42. What does the term “precision” measure in a (c) F1 score
classification task? (d) Mean Squared Error (MSE)
(a) The proportion of true positives
49. What is the role of the term “momentum” in
(b) The proportion of true negatives
gradient descent optimization?
(c) The ratio of true positives to false positives
(a) Control the learning rate
(d) The ratio of true positives to true negatives
(b) Accelerate convergence by introducing
43. In a decision tree, what is a leaf node? inertia
(a) The root of the tree (c) Regularize the model
(b) A node with only one child (d) Define the loss function

www.madeeasy.in © Copyright
 Machine Learning 5

50. What is the purpose of a ROC curve in Machine (a) Mean imputation (b) Mode imputation
Learning? (c) Median imputation (d) All of the above
(a) Visualizing the trade-off between sensitivity
57. What is the purpose of a hyperparameter in the
and specificity
context of Machine Learning?
(b) Evaluating model accuracy
(a) It is learned from the training data
(c) Determining the optimal learning rate
(b) It is a parameter automatically adjusted by
(d) Selecting features for the model
the model during training
51. Which algorithm is commonly used for image (c) It is set before training and influences the
classification tasks? learning process
(a) K-Means (d) It is a measure of model performance
(b) Decision Trees
58. Which of the following is an ensemble learning
(c) Convolutional Neural Networks (CNN) method?
(d) Principal Component Analysis (PCA) (a) Linear Regression
52. What is the primary objective of reinforcement (b) Support Vector Machines (SVM)
learning? (c) Random Forest
(a) Minimize prediction error (d) K-Nearest Neighbors (KNN)
(b) Discover hidden patterns in data 59. What is the objective of the Expectation-
(c) Optimize decision-making processes Maximization (EM) algorithm?
(d) Maximize cumulative reward over time (a) Minimize prediction error
53. Which algorithm is commonly used for natural (b) Optimize clustering assignments in
language processing tasks, such as text unsupervised learning
classification? (c) Improve model interpretability
(a) Naive Bayes (d) Adjust learning rates dynamically
(b) K-Means 60. Which of the following is an example of a semi-
(c) Decision Trees supervised learning technique?
(d) Recurrent Neural Networks (RNN) (a) Decision Trees
(b) K-Means
54. In the context of clustering, what does the term
(c) Support Vector Machines (SVM)
“centroid” refer to?
(d) Self-training
(a) A representative point of a cluster
(b) The center of the dataset 61. What is the primary purpose of the term “batch
(c) A measure of feature importance size” in training neural networks?
(d) The output of a neural network layer (a) Control the number of epochs
(b) Define the number of layers in the network
55. What is the purpose of the term “dropout” in
(c) Determine the amount of data processed in
neural networks?
each iteration
(a) Eliminate outliers from the training set
(d) Set the learning rate
(b) Regularize the model by randomly dropping
neurons during training 62. In which scenario would you use the Kullback-
(c) Adjust learning rates dynamically Leibler (KL) divergence?
(d) Enhance the interpretability of the model (a) Image classification
(b) Text generation
56. Which technique is commonly used for handling
(c) Reinforcement learning
missing data in Machine Learning?
(d) Clustering

© Copyright www.madeeasy.in
6 CPQ 2025 • Data Science & Artificial Intelligence

63. What is the purpose of the term “bagging” in 69. Which optimization algorithm is commonly used
ensemble learning? for training deep neural networks?
(a) Combine predictions from multiple models (a) Stochastic Gradient Descent (SGD)
to reduce variance (b) Genetic Algorithm
(b) Increase model complexity to capture (c) AdaBoost
intricate patterns (d) Random Search
(c) Introduce non-linearity in the decision 70. What is the primary goal of unsupervised
boundary learning?
(d) Speed up the training process (a) Minimize prediction error
(b) Discover hidden patterns in data
64. Which of the following is a distance metric
(c) Optimize decision-making processes
commonly used in k-nearest neighbors (KNN)? (d) Classify data into predefined categories
(a) Hamming distance
(b) Cosine similarity 71. Which technique is used to prevent the vanishing
gradient problem in deep neural networks?
(c) Jaccard similarity
(a) Batch normalization
(d) All of the above
(b) Dropout
65. What does the term “precision” measure in a (c) Gradient clipping
classification task? (d) L1 regularization
(a) The proportion of true positives 72. What is the primary difference between
(b) The proportion of true negatives regression and classification?
(c) The ratio of true positives to false positives (a) Regression deals with discrete outcomes,
(d) The ratio of true positives to true negatives while classification deals with continuous
outcomes.
66. In a decision tree, what is a leaf node?
(b) Regression predicts labels, while
(a) The root of the tree
classification predicts numerical values.
(b) A node with only one child
(c) Regression predicts numerical values, while
(c) A node with no children classification predicts labels.
(d) A node with more than two children (d) There is no difference; the terms are
67. Which algorithm is commonly used for interchangeable.
dimensionality reduction? 73. What is the purpose of the term “batch
(a) Decision Trees normalization” in neural networks?
(b) Principal Component Analysis (PCA) (a) Normalize the input data
(c) Naive Bayes (b) Speed up the training process
(d) Random Forest (c) Stabilize and accelerate the training of deep
neural networks
68. What is the purpose of the term “dropout” in (d) Regularize the model by randomly dropping
neural networks? neurons during training
(a) Regularize the model by randomly dropping
74. Which algorithm is suitable for outlier detection?
neurons during training
(a) K-Means
(b) Increase the model's capacity
(b) K-Nearest Neighbors (KNN)
(c) Reduce training time
(c) Support Vector Machines (SVM)
(d) Enhance model interpretability (d) Naive Bayes

www.madeeasy.in © Copyright
 Machine Learning 7

75. What is the purpose of the term “hyperparameter (c) Standardize numerical features
tuning”? (d) Normalize data distribution
(a) Adjusting the model parameters during
81. What is the primary purpose of the term “cross-
training
validation” in machine learning?
(b) Finding the optimal values for parameters
(a) Train a model on multiple subsets of the
set before training
data to improve generalization
(c) Increasing the number of features in the
(b) Select the optimal hyperparameters for a
model
model
(d) Reducing the learning rate dynamically
(c) Evaluate model performance on the training
76. In reinforcement learning, what is the role of the set
“reward” signal? (d) Test the model on unseen data
(a) Define the loss function
82. What is the significance of the term “bag of
(b) Measure the model’s accuracy
words” in natural language processing?
(c) Guide the learning process by providing
(a) It represents the collection of all unique
feedback
words in a document
(d) Control the exploration-exploitation trade-off
(b) It defines the order of words in a document
77. Which algorithm is commonly used for (c) It calculates the term frequency of words in
recommendation systems? a document
(a) K-Means (d) It measures the semantic similarity between
(b) Decision Trees documents
(c) Collaborative Filtering
83. Which ensemble learning method builds multiple
(d) Support Vector Machines (SVM)
models sequentially and corrects errors made
78. What is the purpose of the term “cross-entropy” by the previous ones?
in classification problems? (a) Bagging (b) Boosting
(a) It measures the difference between (c) Stacking (d) Random Forest
predicted and actual values
84. What is the purpose of the term “mean squared
(b) It defines the learning rate of the model
error” in regression problems?
(c) It calculates the margin of the decision
(a) Measure the accuracy of a classification
boundary
model
(d) It determines the number of clusters in the
(b) Evaluate the performance of a clustering
data
algorithm
79. Which of the following is a drawback of the (c) Quantify the difference between predicted
k-nearest neighbors (KNN) algorithm? and actual values
(a) Sensitivity to outliers (d) Determine the optimal number of clusters
(b) Inability to handle categorical data in K-Means
(c) Requires extensive hyperparameter tuning
85. Which algorithm is suitable for solving a binary
(d) Limited to linear relationships in data
classification problem with a large number of
80. What is the purpose of the term “one-hot features?
encoding” in machine learning? (a) Support Vector Machines (SVM)
(a) Encode ordinal variables (b) Naive Bayes
(b) Represent categorical variables as binary (c) Decision Trees
vectors (d) K-Means

© Copyright www.madeeasy.in
8 CPQ 2025 • Data Science & Artificial Intelligence

86. What does the term “cross-entropy loss” measure 91. What is the main advantage of using L1
in a neural network? regularization in linear regression?
(a) The difference between predicted and (a) Encourages sparsity in the feature space
actual values (b) Reduces the risk of overfitting
(b) The distance between data points in a high- (c) Improves the interpretability of the model
dimensional space (d) Increases the number of relevant features
(c) The complexity of the model
(d) The dissimilarity between probability 92. What is the purpose of regularization in Machine
distributions Learning?
(a) Improve model interpretability
87. In which scenario would you use Principal
(b) Reduce model complexity and prevent
Component Analysis (PCA)?
overfitting
(a) Feature selection in a high-dimensional
dataset (c) Increase the number of features in the model
(b) Clustering data points into groups (d) Boost the accuracy of the training set
(c) Building decision trees for classification 93. Which evaluation metric is suitable for
(d) Generating synthetic data for training imbalanced classification problems?
88. What is the primary advantage of using a (a) Accuracy
learning rate schedule in training a neural network? (b) Precision
(a) Speed up the convergence of the model (c) F1 score
(b) Stabilize the training process and avoid (d) Mean Squared Error (MSE)
overshooting the minimum
94. What is the role of the term “momentum” in
(c) Increase the model's capacity to capture
complex patterns gradient descent optimization?
(d) Simplify the optimization algorithm (a) Control the learning rate
(b) Accelerate convergence by introducing
89. Which technique is used for reducing the inertia
dimensionality of data while preserving its
(c) Regularize the model
variance?
(d) Define the loss function
(a) Principal Component Analysis (PCA)
(b) Singular Value Decomposition (SVD) 95. What is the purpose of a ROC curve in Machine
(c) t-Distributed Stochastic Neighbor Embedding Learning?
(t-SNE) (a) Visualizing the trade-off between sensitivity
(d) Random Forest and specificity
90. What is the primary purpose of the term “grid (b) Evaluating model accuracy
search” in hyperparameter tuning? (c) Determining the optimal learning rate
(a) Randomly selecting hyperparameter values (d) Selecting features for the model
from a grid
96. Which algorithm is commonly used for image
(b) Systematically searching through a
classification tasks?
predefined hyperparameter space
(a) K-Means
(c) Dynamically adjusting hyperparameters
(b) Decision Trees
during training
(c) Convolutional Neural Networks (CNN)
(d) Assessing the impact of different hyper-
parameter values on model performance (d) Principal Component Analysis (PCA)

www.madeeasy.in © Copyright
 Machine Learning 9

97. What is the primary objective of reinforcement 103. Which of the following is an ensemble learning
learning? method?
(a) Minimize prediction error (a) Linear Regression
(b) Discover hidden patterns in data (b) Support Vector Machines (SVM)
(c) Optimize decision-making processes (c) Random Forest
(d) Maximize cumulative reward over time (d) K-Nearest Neighbors (KNN)

98. Which algorithm is commonly used for natural 104. What is the objective of the Expectation-
language processing tasks, such as text Maximization (EM) algorithm?
classification? (a) Minimize prediction error
(a) Naive Bayes (b) Optimize clustering assignments in
(b) K-Means unsupervised learning
(c) Decision Trees (c) Improve model interpretability
(d) Recurrent Neural Networks (RNN) (d) Adjust learning rates dynamically

99. In the context of clustering, what does the term 105. Which of the following is an example of a semi-
“centroid” refer to? supervised learning technique?
(a) A representative point of a cluster (a) Decision Trees
(b) K-Means
(b) The center of the dataset
(c) Support Vector Machines (SVM)
(c) A measure of feature importance
(d) Self-training
(d) The output of a neural network layer
106. What is the primary purpose of the term “batch
100. What is the purpose of the term “dropout” in
size” in training neural networks?
neural networks?
(a) Control the number of epochs
(a) Eliminate outliers from the training set
(b) Define the number of layers in the network
(b) Regularize the model by randomly dropping
(c) Determine the amount of data processed in
neurons during training
each iteration
(c) Adjust learning rates dynamically
(d) Set the learning rate
(d) Enhance the interpretability of the model
107. In which scenario would you use the Kullback-
101. Which technique is commonly used for handling
Leibler (KL) divergence?
missing data in Machine Learning?
(a) Image classification
(a) Mean imputation
(b) Text generation
(b) Mode imputation
(c) Reinforcement learning
(c) Median imputation
(d) Clustering
(d) All of the above
108. What is the purpose of the term “bagging” in
102. What is the purpose of a hyperparameter in the ensemble learning?
context of Machine Learning? (a) Combine predictions from multiple models
(a) It is learned from the training data to reduce variance
(b) It is a parameter automatically adjusted by (b) Increase model complexity to capture
the model during training intricate patterns
(c) It is set before training and influences the (c) Introduce non-linearity in the decision
learning process boundary
(d) It is a measure of model performance (d) Speed up the training process

© Copyright www.madeeasy.in
10 CPQ 2025 • Data Science & Artificial Intelligence

109. Which of the following is a distance metric 116. Which technique is used to prevent the vanishing
commonly used in k-nearest neighbors (KNN)? gradient problem in deep neural networks?
(a) Hamming distance (a) Batch normalization
(b) Cosine similarity (b) Dropout
(c) Jaccard similarity (c) Gradient clipping
(d) All of the above (d) L1 regularization
110. What does the term “precision” measure in a 117. What is the purpose of the Viterbi algorithm in
classification task? machine learning?
(a) The proportion of true positives (a) Feature selection
(b) The proportion of true negatives (b) Sequence alignment
(c) The ratio of true positives to false positives (c) Hyperparameter tuning
(d) The ratio of true positives to true negatives (d) Clustering
111. In a decision tree, what is a leaf node? 118. Which technique is commonly used for handling
(a) The root of the tree imbalanced datasets in classification
(b) A node with only one child problems?
(c) A node with no children (a) Principal Component Analysis (PCA)
(d) A node with more than two children (b) SMOTE (Synthetic Minority Over-sampling
112. Which algorithm is commonly used for Technique)
dimensionality reduction? (c) L1 regularization
(a) Decision Trees (d) Expectation-Maximization (EM)
(b) Principal Component Analysis (PCA) 119. In reinforcement learning, what is the significance
(c) Naive Bayes of the discount factor (gamma)?
(d) Random Forest (a) Controls the exploration rate
113. What is the purpose of the term “dropout” in (b) Determines the learning rate
neural networks? (c) Adjusts the rate of future rewards
(a) Regularize the model by randomly dropping (d) Sets the probability of taking a random
neurons during training action
(b) Increase the model's capacity
120. What is the primary goal of ensemble methods
(c) Reduce training time
such as stacking?
(d) Enhance model interpretability
(a) Improve model interpretability
114. Which optimization algorithm is commonly used (b) Combine multiple models to enhance
for training deep neural networks? performance
(a) Stochastic Gradient Descent (SGD) (c) Reduce dimensionality
(b) Genetic Algorithm (d) Speed up the training process
(c) AdaBoost
121. What is the purpose of the term “batch
(d) Random Search
normalization” in deep neural networks?
115. What is the primary goal of unsupervised (a) Normalize the input data
learning? (b) Speed up the training process
(a) Minimize prediction error (c) Stabilize and accelerate the training of deep
(b) Discover hidden patterns in data neural networks
(c) Optimize decision-making processes (d) Regularize the model by randomly dropping
(d) Classify data into predefined categories neurons during training

www.madeeasy.in © Copyright
 Machine Learning 11

122. Which algorithm is commonly used for anomaly 123. What is the purpose of the term “word
detection in multivariate time series data? embeddings” in natural language processing
(a) K-Means (NLP)?
(b) Isolation Forest (a) Represent words as dense vectors in a
(c) Support Vector Machines (SVM) continuous space
(d) Naive Bayes (b) Tokenize sentences into individual words
(c) Remove stop words from text
(d) Convert words into one-hot encoded vectors

„„„„

© Copyright www.madeeasy.in