Machine Learning MCQs
machine Learning mcqs
Basics & Supervised Learning (1-100)
1. What is the main goal of supervised learning?
- A. To group data without labels
- B. To discover hidden patterns
- C. To learn a mapping from inputs to known outputs
- D. To minimize entropy
Answer: C
Supervised learning uses labeled data to train models to predict outcomes based on input.
2. Which of the following is a supervised learning algorithm?
- A. K-means
- B. Linear Regression
- C. PCA
- D. DBSCAN
Answer: B
Linear Regression is a supervised learning algorithm that models the relationship between inputs and outputs.
3. What is the difference between classification and regression?
- A. Classification predicts continuous output, regression predicts categories
- B. Classification predicts categories, regression predicts continuous output
- C. Both predict categories
- D. Both predict continuous output
Answer: B
Classification predicts discrete labels, while regression predicts continuous values.
4. Which of these is NOT a supervised learning algorithm?
- A. Decision Trees
- B. Support Vector Machines
- C. K-Means Clustering
- D. Linear Regression
Answer: C
K-Means is an unsupervised learning algorithm used for clustering.
5. What is the purpose of training data in supervised learning?
- A. To test the model's performance
- B. To tune hyperparameters only
- C. To teach the model how to map inputs to outputs
- D. To evaluate the final model
Answer: C
Training data is used to train the model to learn the relationship between inputs and outputs.
6. What does overfitting mean?
- A. Model performs poorly on training data
- B. Model performs well on unseen data
- C. Model memorizes training data and performs poorly on test data
- D. Model is too simple
Answer: C
Overfitting happens when a model learns noise and details from training data, hurting its ability to generalize.
7. What is cross-validation used for?
- A. Increasing training data size
- B. Assessing model’s generalization on unseen data
- C. Visualizing data distribution
- D. Reducing dimensionality
Answer: B
Cross-validation helps estimate how well a model generalizes by testing it on multiple subsets of data.
8. Which loss function is commonly used for regression tasks?
- A. Mean Squared Error
- B. Cross Entropy
- C. Hinge Loss
- D. KL Divergence
Answer: A
Mean Squared Error measures the average squared difference between estimated values and actual values.
9. What technique is used to prevent overfitting?
- A. Regularization
- B. Increasing model complexity
- C. Ignoring validation data
- D. Using small training sets
Answer: A
Regularization adds a penalty to the loss function to discourage complex models.
10. Which activation function is commonly used in neural networks?
- A. ReLU
- B. Sigmoid
- C. Tanh
- D. All of the above
Answer: D
ReLU, Sigmoid, and Tanh are all common activation functions used in neural networks.
11. What is the main goal of supervised learning?
- A. To group data without labels
- B. To discover hidden patterns
- C. To learn a mapping from inputs to known outputs
- D. To minimize entropy
Answer: C
Supervised learning uses labeled data to train models to predict outcomes based on input.
12. Which of the following is a supervised learning algorithm?
- A. K-means
- B. Linear Regression
- C. PCA
- D. DBSCAN
Answer: B
Linear Regression is a supervised learning algorithm that models the relationship between inputs and outputs.
13. What is the difference between classification and regression?
- A. Classification predicts continuous output, regression predicts categories
- B. Classification predicts categories, regression predicts continuous output
- C. Both predict categories
- D. Both predict continuous output
Answer: B
Classification predicts discrete labels, while regression predicts continuous values.
14. Which of these is NOT a supervised learning algorithm?
- A. Decision Trees
- B. Support Vector Machines
- C. K-Means Clustering
- D. Linear Regression
Answer: C
K-Means is an unsupervised learning algorithm used for clustering.
15. What is the purpose of training data in supervised learning?
- A. To test the model's performance
- B. To tune hyperparameters only
- C. To teach the model how to map inputs to outputs
- D. To evaluate the final model
Answer: C
Training data is used to train the model to learn the relationship between inputs and outputs.
16. What does overfitting mean?
- A. Model performs poorly on training data
- B. Model performs well on unseen data
- C. Model memorizes training data and performs poorly on test data
- D. Model is too simple
Answer: C
Overfitting happens when a model learns noise and details from training data, hurting its ability to generalize.
17. What is cross-validation used for?
- A. Increasing training data size
- B. Assessing model’s generalization on unseen data
- C. Visualizing data distribution
- D. Reducing dimensionality
Answer: B
Cross-validation helps estimate how well a model generalizes by testing it on multiple subsets of data.
18. Which loss function is commonly used for regression tasks?
- A. Mean Squared Error
- B. Cross Entropy
- C. Hinge Loss
- D. KL Divergence
Answer: A
Mean Squared Error measures the average squared difference between estimated values and actual values.
19. What technique is used to prevent overfitting?
- A. Regularization
- B. Increasing model complexity
- C. Ignoring validation data
- D. Using small training sets
Answer: A
Regularization adds a penalty to the loss function to discourage complex models.
20. Which activation function is commonly used in neural networks?
- A. ReLU
- B. Sigmoid
- C. Tanh
- D. All of the above
Answer: D
ReLU, Sigmoid, and Tanh are all common activation functions used in neural networks.
21. What is the main goal of supervised learning?
- A. To group data without labels
- B. To discover hidden patterns
- C. To learn a mapping from inputs to known outputs
- D. To minimize entropy
Answer: C
Supervised learning uses labeled data to train models to predict outcomes based on input.
22. Which of the following is a supervised learning algorithm?
- A. K-means
- B. Linear Regression
- C. PCA
- D. DBSCAN
Answer: B
Linear Regression is a supervised learning algorithm that models the relationship between inputs and outputs.
23. What is the difference between classification and regression?
- A. Classification predicts continuous output, regression predicts categories
- B. Classification predicts categories, regression predicts continuous output
- C. Both predict categories
- D. Both predict continuous output
Answer: B
Classification predicts discrete labels, while regression predicts continuous values.
24. Which of these is NOT a supervised learning algorithm?
- A. Decision Trees
- B. Support Vector Machines
- C. K-Means Clustering
- D. Linear Regression
Answer: C
K-Means is an unsupervised learning algorithm used for clustering.
25. What is the purpose of training data in supervised learning?
- A. To test the model's performance
- B. To tune hyperparameters only
- C. To teach the model how to map inputs to outputs
- D. To evaluate the final model
Answer: C
Training data is used to train the model to learn the relationship between inputs and outputs.
26. What does overfitting mean?
- A. Model performs poorly on training data
- B. Model performs well on unseen data
- C. Model memorizes training data and performs poorly on test data
- D. Model is too simple
Answer: C
Overfitting happens when a model learns noise and details from training data, hurting its ability to generalize.
27. What is cross-validation used for?
- A. Increasing training data size
- B. Assessing model’s generalization on unseen data
- C. Visualizing data distribution
- D. Reducing dimensionality
Answer: B
Cross-validation helps estimate how well a model generalizes by testing it on multiple subsets of data.
28. Which loss function is commonly used for regression tasks?
- A. Mean Squared Error
- B. Cross Entropy
- C. Hinge Loss
- D. KL Divergence
Answer: A
Mean Squared Error measures the average squared difference between estimated values and actual values.
29. What technique is used to prevent overfitting?
- A. Regularization
- B. Increasing model complexity
- C. Ignoring validation data
- D. Using small training sets
Answer: A
Regularization adds a penalty to the loss function to discourage complex models.
30. Which activation function is commonly used in neural networks?
- A. ReLU
- B. Sigmoid
- C. Tanh
- D. All of the above
Answer: D
ReLU, Sigmoid, and Tanh are all common activation functions used in neural networks.
31. What is the main goal of supervised learning?
- A. To group data without labels
- B. To discover hidden patterns
- C. To learn a mapping from inputs to known outputs
- D. To minimize entropy
Answer: C
Supervised learning uses labeled data to train models to predict outcomes based on input.
32. Which of the following is a supervised learning algorithm?
- A. K-means
- B. Linear Regression
- C. PCA
- D. DBSCAN
Answer: B
Linear Regression is a supervised learning algorithm that models the relationship between inputs and outputs.
33. What is the difference between classification and regression?
- A. Classification predicts continuous output, regression predicts categories
- B. Classification predicts categories, regression predicts continuous output
- C. Both predict categories
- D. Both predict continuous output
Answer: B
Classification predicts discrete labels, while regression predicts continuous values.
34. Which of these is NOT a supervised learning algorithm?
- A. Decision Trees
- B. Support Vector Machines
- C. K-Means Clustering
- D. Linear Regression
Answer: C
K-Means is an unsupervised learning algorithm used for clustering.
35. What is the purpose of training data in supervised learning?
- A. To test the model's performance
- B. To tune hyperparameters only
- C. To teach the model how to map inputs to outputs
- D. To evaluate the final model
Answer: C
Training data is used to train the model to learn the relationship between inputs and outputs.
36. What does overfitting mean?
- A. Model performs poorly on training data
- B. Model performs well on unseen data
- C. Model memorizes training data and performs poorly on test data
- D. Model is too simple
Answer: C
Overfitting happens when a model learns noise and details from training data, hurting its ability to generalize.
37. What is cross-validation used for?
- A. Increasing training data size
- B. Assessing model’s generalization on unseen data
- C. Visualizing data distribution
- D. Reducing dimensionality
Answer: B
Cross-validation helps estimate how well a model generalizes by testing it on multiple subsets of data.
38. Which loss function is commonly used for regression tasks?
- A. Mean Squared Error
- B. Cross Entropy
- C. Hinge Loss
- D. KL Divergence
Answer: A
Mean Squared Error measures the average squared difference between estimated values and actual values.
39. What technique is used to prevent overfitting?
- A. Regularization
- B. Increasing model complexity
- C. Ignoring validation data
- D. Using small training sets
Answer: A
Regularization adds a penalty to the loss function to discourage complex models.
40. Which activation function is commonly used in neural networks?
- A. ReLU
- B. Sigmoid
- C. Tanh
- D. All of the above
Answer: D
ReLU, Sigmoid, and Tanh are all common activation functions used in neural networks.
41. What is the main goal of supervised learning?
- A. To group data without labels
- B. To discover hidden patterns
- C. To learn a mapping from inputs to known outputs
- D. To minimize entropy
Answer: C
Supervised learning uses labeled data to train models to predict outcomes based on input.
42. Which of the following is a supervised learning algorithm?
- A. K-means
- B. Linear Regression
- C. PCA
- D. DBSCAN
Answer: B
Linear Regression is a supervised learning algorithm that models the relationship between inputs and outputs.
43. What is the difference between classification and regression?
- A. Classification predicts continuous output, regression predicts categories
- B. Classification predicts categories, regression predicts continuous output
- C. Both predict categories
- D. Both predict continuous output
Answer: B
Classification predicts discrete labels, while regression predicts continuous values.
44. Which of these is NOT a supervised learning algorithm?
- A. Decision Trees
- B. Support Vector Machines
- C. K-Means Clustering
- D. Linear Regression
Answer: C
K-Means is an unsupervised learning algorithm used for clustering.
45. What is the purpose of training data in supervised learning?
- A. To test the model's performance
- B. To tune hyperparameters only
- C. To teach the model how to map inputs to outputs
- D. To evaluate the final model
Answer: C
Training data is used to train the model to learn the relationship between inputs and outputs.
46. What does overfitting mean?
- A. Model performs poorly on training data
- B. Model performs well on unseen data
- C. Model memorizes training data and performs poorly on test data
- D. Model is too simple
Answer: C
Overfitting happens when a model learns noise and details from training data, hurting its ability to generalize.
47. What is cross-validation used for?
- A. Increasing training data size
- B. Assessing model’s generalization on unseen data
- C. Visualizing data distribution
- D. Reducing dimensionality
Answer: B
Cross-validation helps estimate how well a model generalizes by testing it on multiple subsets of data.
48. Which loss function is commonly used for regression tasks?
- A. Mean Squared Error
- B. Cross Entropy
- C. Hinge Loss
- D. KL Divergence
Answer: A
Mean Squared Error measures the average squared difference between estimated values and actual values.
49. What technique is used to prevent overfitting?
- A. Regularization
- B. Increasing model complexity
- C. Ignoring validation data
- D. Using small training sets
Answer: A
Regularization adds a penalty to the loss function to discourage complex models.
50. Which activation function is commonly used in neural networks?
- A. ReLU
- B. Sigmoid
- C. Tanh
- D. All of the above
Answer: D
ReLU, Sigmoid, and Tanh are all common activation functions used in neural networks.
51. What is the main goal of supervised learning?
- A. To group data without labels
- B. To discover hidden patterns
- C. To learn a mapping from inputs to known outputs
- D. To minimize entropy
Answer: C
Supervised learning uses labeled data to train models to predict outcomes based on input.
52. Which of the following is a supervised learning algorithm?
- A. K-means
- B. Linear Regression
- C. PCA
- D. DBSCAN
Answer: B
Linear Regression is a supervised learning algorithm that models the relationship between inputs and outputs.
53. What is the difference between classification and regression?
- A. Classification predicts continuous output, regression predicts categories
- B. Classification predicts categories, regression predicts continuous output
- C. Both predict categories
- D. Both predict continuous output
Answer: B
Classification predicts discrete labels, while regression predicts continuous values.
54. Which of these is NOT a supervised learning algorithm?
- A. Decision Trees
- B. Support Vector Machines
- C. K-Means Clustering
- D. Linear Regression
Answer: C
K-Means is an unsupervised learning algorithm used for clustering.
55. What is the purpose of training data in supervised learning?
- A. To test the model's performance
- B. To tune hyperparameters only
- C. To teach the model how to map inputs to outputs
- D. To evaluate the final model
Answer: C
Training data is used to train the model to learn the relationship between inputs and outputs.
56. What does overfitting mean?
- A. Model performs poorly on training data
- B. Model performs well on unseen data
- C. Model memorizes training data and performs poorly on test data
- D. Model is too simple
Answer: C
Overfitting happens when a model learns noise and details from training data, hurting its ability to generalize.
57. What is cross-validation used for?
- A. Increasing training data size
- B. Assessing model’s generalization on unseen data
- C. Visualizing data distribution
- D. Reducing dimensionality
Answer: B
Cross-validation helps estimate how well a model generalizes by testing it on multiple subsets of data.
58. Which loss function is commonly used for regression tasks?
- A. Mean Squared Error
- B. Cross Entropy
- C. Hinge Loss
- D. KL Divergence
Answer: A
Mean Squared Error measures the average squared difference between estimated values and actual values.
59. What technique is used to prevent overfitting?
- A. Regularization
- B. Increasing model complexity
- C. Ignoring validation data
- D. Using small training sets
Answer: A
Regularization adds a penalty to the loss function to discourage complex models.
60. Which activation function is commonly used in neural networks?
- A. ReLU
- B. Sigmoid
- C. Tanh
- D. All of the above
Answer: D
ReLU, Sigmoid, and Tanh are all common activation functions used in neural networks.
61. What is the main goal of supervised learning?
- A. To group data without labels
- B. To discover hidden patterns
- C. To learn a mapping from inputs to known outputs
- D. To minimize entropy
Answer: C
Supervised learning uses labeled data to train models to predict outcomes based on input.
62. Which of the following is a supervised learning algorithm?
- A. K-means
- B. Linear Regression
- C. PCA
- D. DBSCAN
Answer: B
Linear Regression is a supervised learning algorithm that models the relationship between inputs and outputs.
63. What is the difference between classification and regression?
- A. Classification predicts continuous output, regression predicts categories
- B. Classification predicts categories, regression predicts continuous output
- C. Both predict categories
- D. Both predict continuous output
Answer: B
Classification predicts discrete labels, while regression predicts continuous values.
64. Which of these is NOT a supervised learning algorithm?
- A. Decision Trees
- B. Support Vector Machines
- C. K-Means Clustering
- D. Linear Regression
Answer: C
K-Means is an unsupervised learning algorithm used for clustering.
65. What is the purpose of training data in supervised learning?
- A. To test the model's performance
- B. To tune hyperparameters only
- C. To teach the model how to map inputs to outputs
- D. To evaluate the final model
Answer: C
Training data is used to train the model to learn the relationship between inputs and outputs.
66. What does overfitting mean?
- A. Model performs poorly on training data
- B. Model performs well on unseen data
- C. Model memorizes training data and performs poorly on test data
- D. Model is too simple
Answer: C
Overfitting happens when a model learns noise and details from training data, hurting its ability to generalize.
67. What is cross-validation used for?
- A. Increasing training data size
- B. Assessing model’s generalization on unseen data
- C. Visualizing data distribution
- D. Reducing dimensionality
Answer: B
Cross-validation helps estimate how well a model generalizes by testing it on multiple subsets of data.
68. Which loss function is commonly used for regression tasks?
- A. Mean Squared Error
- B. Cross Entropy
- C. Hinge Loss
- D. KL Divergence
Answer: A
Mean Squared Error measures the average squared difference between estimated values and actual values.
69. What technique is used to prevent overfitting?
- A. Regularization
- B. Increasing model complexity
- C. Ignoring validation data
- D. Using small training sets
Answer: A
Regularization adds a penalty to the loss function to discourage complex models.
70. Which activation function is commonly used in neural networks?
- A. ReLU
- B. Sigmoid
- C. Tanh
- D. All of the above
Answer: D
ReLU, Sigmoid, and Tanh are all common activation functions used in neural networks.
71. What is the main goal of supervised learning?
- A. To group data without labels
- B. To discover hidden patterns
- C. To learn a mapping from inputs to known outputs
- D. To minimize entropy
Answer: C
Supervised learning uses labeled data to train models to predict outcomes based on input.
72. Which of the following is a supervised learning algorithm?
- A. K-means
- B. Linear Regression
- C. PCA
- D. DBSCAN
Answer: B
Linear Regression is a supervised learning algorithm that models the relationship between inputs and outputs.
73. What is the difference between classification and regression?
- A. Classification predicts continuous output, regression predicts categories
- B. Classification predicts categories, regression predicts continuous output
- C. Both predict categories
- D. Both predict continuous output
Answer: B
Classification predicts discrete labels, while regression predicts continuous values.
74. Which of these is NOT a supervised learning algorithm?
- A. Decision Trees
- B. Support Vector Machines
- C. K-Means Clustering
- D. Linear Regression
Answer: C
K-Means is an unsupervised learning algorithm used for clustering.
75. What is the purpose of training data in supervised learning?
- A. To test the model's performance
- B. To tune hyperparameters only
- C. To teach the model how to map inputs to outputs
- D. To evaluate the final model
Answer: C
Training data is used to train the model to learn the relationship between inputs and outputs.
76. What does overfitting mean?
- A. Model performs poorly on training data
- B. Model performs well on unseen data
- C. Model memorizes training data and performs poorly on test data
- D. Model is too simple
Answer: C
Overfitting happens when a model learns noise and details from training data, hurting its ability to generalize.
77. What is cross-validation used for?
- A. Increasing training data size
- B. Assessing model’s generalization on unseen data
- C. Visualizing data distribution
- D. Reducing dimensionality
Answer: B
Cross-validation helps estimate how well a model generalizes by testing it on multiple subsets of data.
78. Which loss function is commonly used for regression tasks?
- A. Mean Squared Error
- B. Cross Entropy
- C. Hinge Loss
- D. KL Divergence
Answer: A
Mean Squared Error measures the average squared difference between estimated values and actual values.
79. What technique is used to prevent overfitting?
- A. Regularization
- B. Increasing model complexity
- C. Ignoring validation data
- D. Using small training sets
Answer: A
Regularization adds a penalty to the loss function to discourage complex models.
80. Which activation function is commonly used in neural networks?
- A. ReLU
- B. Sigmoid
- C. Tanh
- D. All of the above
Answer: D
ReLU, Sigmoid, and Tanh are all common activation functions used in neural networks.
81. What is the main goal of supervised learning?
- A. To group data without labels
- B. To discover hidden patterns
- C. To learn a mapping from inputs to known outputs
- D. To minimize entropy
Answer: C
Supervised learning uses labeled data to train models to predict outcomes based on input.
82. Which of the following is a supervised learning algorithm?
- A. K-means
- B. Linear Regression
- C. PCA
- D. DBSCAN
Answer: B
Linear Regression is a supervised learning algorithm that models the relationship between inputs and outputs.
83. What is the difference between classification and regression?
- A. Classification predicts continuous output, regression predicts categories
- B. Classification predicts categories, regression predicts continuous output
- C. Both predict categories
- D. Both predict continuous output
Answer: B
Classification predicts discrete labels, while regression predicts continuous values.
84. Which of these is NOT a supervised learning algorithm?
- A. Decision Trees
- B. Support Vector Machines
- C. K-Means Clustering
- D. Linear Regression
Answer: C
K-Means is an unsupervised learning algorithm used for clustering.
85. What is the purpose of training data in supervised learning?
- A. To test the model's performance
- B. To tune hyperparameters only
- C. To teach the model how to map inputs to outputs
- D. To evaluate the final model
Answer: C
Training data is used to train the model to learn the relationship between inputs and outputs.
86. What does overfitting mean?
- A. Model performs poorly on training data
- B. Model performs well on unseen data
- C. Model memorizes training data and performs poorly on test data
- D. Model is too simple
Answer: C
Overfitting happens when a model learns noise and details from training data, hurting its ability to generalize.
87. What is cross-validation used for?
- A. Increasing training data size
- B. Assessing model’s generalization on unseen data
- C. Visualizing data distribution
- D. Reducing dimensionality
Answer: B
Cross-validation helps estimate how well a model generalizes by testing it on multiple subsets of data.
88. Which loss function is commonly used for regression tasks?
- A. Mean Squared Error
- B. Cross Entropy
- C. Hinge Loss
- D. KL Divergence
Answer: A
Mean Squared Error measures the average squared difference between estimated values and actual values.
89. What technique is used to prevent overfitting?
- A. Regularization
- B. Increasing model complexity
- C. Ignoring validation data
- D. Using small training sets
Answer: A
Regularization adds a penalty to the loss function to discourage complex models.
90. Which activation function is commonly used in neural networks?
- A. ReLU
- B. Sigmoid
- C. Tanh
- D. All of the above
Answer: D
ReLU, Sigmoid, and Tanh are all common activation functions used in neural networks.
91. What is the main goal of supervised learning?
- A. To group data without labels
- B. To discover hidden patterns
- C. To learn a mapping from inputs to known outputs
- D. To minimize entropy
Answer: C
Supervised learning uses labeled data to train models to predict outcomes based on input.
92. Which of the following is a supervised learning algorithm?
- A. K-means
- B. Linear Regression
- C. PCA
- D. DBSCAN
Answer: B
Linear Regression is a supervised learning algorithm that models the relationship between inputs and outputs.
93. What is the difference between classification and regression?
- A. Classification predicts continuous output, regression predicts categories
- B. Classification predicts categories, regression predicts continuous output
- C. Both predict categories
- D. Both predict continuous output
Answer: B
Classification predicts discrete labels, while regression predicts continuous values.
94. Which of these is NOT a supervised learning algorithm?
- A. Decision Trees
- B. Support Vector Machines
- C. K-Means Clustering
- D. Linear Regression
Answer: C
K-Means is an unsupervised learning algorithm used for clustering.
95. What is the purpose of training data in supervised learning?
- A. To test the model's performance
- B. To tune hyperparameters only
- C. To teach the model how to map inputs to outputs
- D. To evaluate the final model
Answer: C
Training data is used to train the model to learn the relationship between inputs and outputs.
96. What does overfitting mean?
- A. Model performs poorly on training data
- B. Model performs well on unseen data
- C. Model memorizes training data and performs poorly on test data
- D. Model is too simple
Answer: C
Overfitting happens when a model learns noise and details from training data, hurting its ability to generalize.
97. What is cross-validation used for?
- A. Increasing training data size
- B. Assessing model’s generalization on unseen data
- C. Visualizing data distribution
- D. Reducing dimensionality
Answer: B
Cross-validation helps estimate how well a model generalizes by testing it on multiple subsets of data.
98. Which loss function is commonly used for regression tasks?
- A. Mean Squared Error
- B. Cross Entropy
- C. Hinge Loss
- D. KL Divergence
Answer: A
Mean Squared Error measures the average squared difference between estimated values and actual values.
99. What technique is used to prevent overfitting?
- A. Regularization
- B. Increasing model complexity
- C. Ignoring validation data
- D. Using small training sets
Answer: A
Regularization adds a penalty to the loss function to discourage complex models.
100. Which activation function is commonly used in neural networks?
- A. ReLU
- B. Sigmoid
- C. Tanh
- D. All of the above
Answer: D
ReLU, Sigmoid, and Tanh are all common activation functions used in neural networks.