- UNIT-I
.
Unit-1 MCQ's
Data Science and Machine Learning
a) Data Science
b) Artificial Intelligence
c) Statistics
d) All of the above
Answer: b) Artificial Intelligence
2. In Machine Learning, a model learns from:
a) Manually coded rules
b) Data
c) Expert systems
d) Random guessing
Answer: b) Data
3. The process of training a machine learning model involves:
a) Creating datasets
b) Learning patterns from data
c) Fine-tuning hyperparameters
d) All of the above
Answer: d) All of the above
4. Which of the following is not a type of Machine Learning?
a) Supervised Learning
b) Reinforcement Learning
c) Deep Learning
d) Software Engineering
Answer: d) Software Engineering
5. In Supervised Learning, the model learns from:
a) Labeled data
b) Unlabeled data
c) Reward-based learning
d) None of the above
Answer: a) Labeled data
6. Which Machine Learning technique is used when the dataset has no labels?
a) Supervised Learning
b) Unsupervised Learning
c) Reinforcement Learning
d) None of the above
Answer: b) Unsupervised Learning
7. Which of the following is an example of Reinforcement Learning?
a) Spam email classification
b) Self-driving cars learning from the environment
c) Clustering similar customer types
d) Predicting house prices
Answer: b) Self-driving cars learning from the environment
8. In Supervised Learning, which of the following is a Regression problem?
a) Predicting whether an email is spam or not
b) Predicting house prices
c) Identifying dog breeds from images
d) Detecting fraudulent transactions
Answer: b) Predicting house prices
9. In Unsupervised Learning, which of the following techniques is commonly used?
a) Classification
b) Clustering
c) Reinforcement
d) All of the above
Answer: b) Clustering
10. The performance of a Machine Learning model is evaluated using:
a) Accuracy
b) Precision and Recall
c) Mean Squared Error (MSE)
d) All of the above
Answer: d) All of the above
Machine Learning Algorithms MCQs
11. Which of the following algorithms is used for classification problems?
a) Linear Regression
b) Decision Trees
c) K-Means Clustering
d) Principal Component Analysis (PCA)
Answer: b) Decision Trees
12. Which algorithm is commonly used for recommendation systems?
a) K-Means
b) Neural Networks
c) Collaborative Filtering
d) Random Forest
Answer: c) Collaborative Filtering
13. Which Machine Learning algorithm is best suited for image recognition?
a) K-Nearest Neighbors (KNN)
b) Support Vector Machines (SVM)
c) Convolutional Neural Networks (CNN)
d) Decision Trees
Answer: c) Convolutional Neural Networks (CNN)
14. The "Curse of Dimensionality" occurs when:
a) There are too many features in the dataset
b) The dataset is too small
c) The model overfits the training data
d) The dataset is imbalanced
Answer: a) There are too many features in the dataset
15. Overfitting occurs when a Machine Learning model:
a) Performs well on training data but poorly on test data
b) Generalizes well to new data
c) Has too little training data
d) Has a high bias
Answer: a) Performs well on training data but poorly on test data
Neural Networks & Deep Learning MCQs
16. A neural network consists of:
a) Input layer, Hidden layers, and Output layer
b) Only input and output layers
c) Decision trees
d) Statistical equations
Answer: a) Input layer, Hidden layers, and Output layer
17. Which activation function is commonly used in deep learning models?
a) ReLU (Rectified Linear Unit)
b) Linear Function
c) Sigmoid
d) Both a & c
Answer: d) Both a & c
18. What is the purpose of the loss function in a Machine Learning model?
a) To optimize the dataset
b) To calculate the error in predictions
c) To visualize data distribution
d) To classify images
Answer: b) To calculate the error in predictions
19. Gradient Descent is used for:
a) Finding the minimum error of a model
b) Clustering data
c) Labeling the dataset
d) Increasing dataset size
Answer: a) Finding the minimum error of a model
20. What is Transfer Learning in Machine Learning?
a) Training a new model from scratch
b) Using a pre-trained model for a different task
c) Copying data from one dataset to another
d) Removing outliers from the dataset
Answer: b) Using a pre-trained model for a different task
Regression vs. Classification MCQs
21. Which of the following problems is best solved using Regression?
a) Predicting whether a tumor is malignant or benign
b) Predicting tomorrow’s temperature
c) Classifying emails as spam or not spam
d) Identifying dog breeds from images
Answer: b) Predicting tomorrow’s temperature
22. In Classification problems, the target variable is:
a) Continuous
b) Discrete
c) Both continuous and discrete
d) None of the above
Answer: b) Discrete
23. Which of the following is not a Classification algorithm?
a) Decision Tree
b) Logistic Regression
c) Random Forest
d) Linear Regression
Answer: d) Linear Regression
24. Which algorithm is used for Regression?
a) K-Nearest Neighbors (KNN)
b) Logistic Regression
c) Support Vector Machine (SVM)
d) Linear Regression
Answer: d) Linear Regression
25. In Regression problems, the output variable is:
a) Categorical
b) Continuous
c) Binary
d) Discrete
Answer: b) Continuous
26. Which of the following is a Classification problem?
a) Predicting the price of a house
b) Predicting the number of units sold
c) Identifying whether an email is spam or not spam
d) Predicting the temperature in Celsius
Answer: c) Identifying whether an email is spam or not spam
27. Which of the following algorithms can be used for both Classification and Regression?
a) Decision Tree
b) Logistic Regression
c) Support Vector Machine
d) Both a & c
Answer: d) Both a & c
28. Logistic Regression is used for:
a) Regression problems
b) Classification problems
c) Both a & b
d) None of the above
Answer: b) Classification problems
29. Which of the following methods can be used to evaluate Regression models?
a) Accuracy
b) Mean Squared Error (MSE)
c) Precision and Recall
d) Confusion Matrix
Answer: b) Mean Squared Error (MSE)
30. Which of the following is not a metric for Classification problems?
a) Accuracy
b) Precision
c) Root Mean Squared Error (RMSE)
d) Recall
Answer: c) Root Mean Squared Error (RMSE)
Regression MCQs
31. Linear Regression assumes a relationship between:
a) Independent and dependent variables
b) Only categorical variables
c) Non-linear data
d) Clusters of data
Answer: a) Independent and dependent variables
32. The equation of a simple Linear Regression model is:
a)
b)
c) Both a & b
d) None of the above
Answer: c) Both a & b
33. The main objective of a Regression model is to:
a) Predict categories
b) Estimate relationships between variables
c) Classify input data
d) Find clusters
Answer: b) Estimate relationships between variables
34. Which of the following is not a Regression algorithm?
a) Support Vector Regression (SVR)
b) Decision Tree Regression
c) K-Means Clustering
d) Polynomial Regression
Answer: c) K-Means Clustering
35. If a Regression model has high training accuracy but poor test accuracy, it is:
a) Underfitting
b) Overfitting
c) Well-generalized
d) None of the above
Answer: b) Overfitting
36. A high R² (coefficient of determination) value means:
a) Poor model performance
b) Good model performance
c) High bias
d) High variance
Answer: b) Good model performance
37. Which of the following is not an assumption of Linear Regression?
a) Linearity
b) Homoscedasticity
c) Multicollinearity
d) Normality
Answer: c) Multicollinearity
Classification MCQs
38. The Confusion Matrix is used to evaluate:
a) Regression models
b) Classification models
c) Clustering algorithms
d) Neural networks
Answer: b) Classification models
39. Precision is defined as:
a)
b)
c)
d)
Answer: a)
40. The F1-score is useful when:
a) There is an equal distribution of classes
b) The dataset is large
c) The dataset has an imbalanced class distribution
d) There are no outliers
Answer: c) The dataset has an imbalanced class distribution
41. Which of the following Classification algorithms is best for large datasets with many features?
a) Decision Trees
b) K-Nearest Neighbors (KNN)
c) Naïve Bayes
d) Neural Networks
Answer: d) Neural Networks
42. K-Nearest Neighbors (KNN) works based on:
a) Distance between data points
b) Probabilistic classification
c) Decision trees
d) Gradient boosting
Answer: a) Distance between data points
43. Which algorithm is best for detecting fraudulent transactions?
a) Logistic Regression
b) Decision Trees
c) Random Forest
d) Both b & c
Answer: d) Both b & c
44. Which algorithm is used for text classification (e.g., spam detection)?
a) Naïve Bayes
b) K-Means
c) Decision Trees
d) Support Vector Machines
Answer: a) Naïve Bayes
45. Which Classification algorithm is based on Bayes’ Theorem?
a) KNN
b) Naïve Bayes
c) Decision Trees
d) Random Forest
Answer: b) Naïve Bayes
46. One-vs-All (OvA) is used in:
a) Multi-class classification
b) Regression
c) Binary classification
d) Clustering
Answer: a) Multi-class classification
47. The Receiver Operating Characteristic (ROC) curve is used to evaluate:
a) Regression models
b) Classification models
c) Clustering models
d) Anomaly detection
Answer: b) Classification models
48. Supervised learning requires:
a) Labeled data
b) Unlabeled data
c) No data
d) Only categorical variables
Answer: a) Labeled data
49. Which of the following is an example of Supervised Learning?
a) Clustering customers into groups
b) Predicting house prices based on historical data
c) Finding patterns in unstructured data
d) Identifying anomalies in network security
Answer: b) Predicting house prices based on historical data
50. In Unsupervised Learning, the algorithm learns from:
a) Labeled data
b) Unlabeled data
c) Both labeled and unlabeled data
d) No data
Answer: b) Unlabeled data
51. Which of the following is an example of Unsupervised Learning?
a) Predicting the stock market price
b) Identifying customer segments in a shopping website
c) Spam email classification
d) Diagnosing diseases based on symptoms
Answer: b) Identifying customer segments in a shopping website
52. A dataset that contains input features and corresponding correct outputs is used in:
a) Unsupervised Learning
b) Supervised Learning
c) Reinforcement Learning
d) None of the above
Answer: b) Supervised Learning
53. The primary goal of Unsupervised Learning is to:
a) Predict an output variable
b) Classify new data points
c) Discover hidden patterns and structures
d) Train models for reinforcement
Answer: c) Discover hidden patterns and structures
Algorithms in Supervised and Unsupervised Learning
54. Which of the following is not a Supervised Learning algorithm?
a) Decision Trees
b) Linear Regression
c) K-Means Clustering
d) Support Vector Machines
Answer: c) K-Means Clustering
55. Which of the following algorithms is used for Unsupervised Learning?
a) Random Forest
b) Logistic Regression
c) Hierarchical Clustering
d) Naïve Bayes
Answer: c) Hierarchical Clustering
56. K-Means Clustering is an example of:
a) Supervised Learning
b) Unsupervised Learning
c) Reinforcement Learning
d) Semi-Supervised Learning
Answer: b) Unsupervised Learning
57. Which of the following is not a clustering algorithm?
a) K-Means
b) DBSCAN
c) Decision Trees
d) Hierarchical Clustering
Answer: c) Decision Trees
58. In Supervised Learning, classification problems have:
a) Continuous target variables
b) Discrete target variables
c) No target variable
d) No input variables
Answer: b) Discrete target variables
59. Which of the following algorithms is used for classification in Supervised Learning?
a) Linear Regression
b) K-Means
c) Logistic Regression
d) Principal Component Analysis (PCA)
Answer: c) Logistic Regression
60. Which of the following algorithms is only used for Regression problems?
a) Support Vector Machines (SVM)
b) K-Means
c) Linear Regression
d) Naïve Bayes
Answer: c) Linear Regression
Applications of Supervised and Unsupervised Learning
61. Identifying fraudulent transactions is an example of:
a) Supervised Learning
b) Unsupervised Learning
c) Semi-Supervised Learning
d) Reinforcement Learning
Answer: a) Supervised Learning
62. Grouping customers based on shopping behavior is an example of:
a) Supervised Learning
b) Unsupervised Learning
c) Reinforcement Learning
d) None of the above
Answer: b) Unsupervised Learning
63. Face recognition in a security system is an application of:
a) Supervised Learning
b) Unsupervised Learning
c) Reinforcement Learning
d) None of the above
Answer: a) Supervised Learning
64. Market Basket Analysis, which finds items that are frequently bought together, is an example of:
a) Supervised Learning
b) Unsupervised Learning
c) Reinforcement Learning
d) None of the above
Answer: b) Unsupervised Learning
65. Supervised Learning models are used in:
a) Weather forecasting
b) Spam email filtering
c) Credit scoring
d) All of the above
Answer: d) All of the above
66. Which of the following is a real-world application of Unsupervised Learning?
a) Medical diagnosis
b) Self-driving cars
c) Customer segmentation
d) Stock price prediction
Answer: c) Customer segmentation
Performance and Evaluation
67. Supervised Learning models are evaluated using:
a) Accuracy
b) Precision and Recall
c) Mean Squared Error (MSE)
d) All of the above
Answer: d) All of the above
68. Which metric is used to evaluate clustering algorithms?
a) Mean Absolute Error
b) Silhouette Score
c) Confusion Matrix
d) F1 Score
Answer: b) Silhouette Score
69. Which problem is most likely to be solved using Unsupervised Learning?
a) Predicting student exam scores
b) Identifying spam emails
c) Finding groups of similar documents
d) Diagnosing diseases based on symptoms
Answer: c) Finding groups of similar documents
70. Semi-Supervised Learning is a combination of:
a) Supervised and Reinforcement Learning
b) Supervised and Unsupervised Learning
c) Clustering and Classification
d) Regression and Classification
Answer: b) Supervised and Unsupervised Learning
71. The main drawback of Supervised Learning is:
a) Requires labeled data
b) Cannot make predictions
c) Does not learn from experience
d) Only works with categorical data
Answer: a) Requires labeled data
72. In Supervised Learning, the Bias-Variance Tradeoff means:
a) Increasing bias improves variance
b) High bias and high variance improve model accuracy
c) Reducing bias increases variance and vice versa
d) There is no relationship between bias and variance
Answer: c) Reducing bias increases variance and vice versa
73. Unsupervised Learning models are difficult to evaluate because:
a) There is no labeled data
b) They require more computational power
c) They have fixed accuracy
d) They use supervised labels
Answer: a) There is no labeled data
No comments:
Post a Comment