Skip to main content

MCQ

2025​

1. Decision tree uses the inductive learning machine learning approach.​

a) False
b) True

Answer: b) True

Support: Decision trees learn general rules from training examples, which is the principle of inductive learning.

2. Machine learning is a subset of?.​

a) Deep learning
b) IoT
c) Computer Science
d) Artificial Intelligence

Answer: d) Artificial Intelligence

Support: Machine Learning is a branch of Artificial Intelligence that enables systems to learn from data.

3. In what type of learning, labelled data is used?​

a) Supervised learning
b) Unsupervised learning
c) Active learning
d) Reinforcement learning

Answer: a) Supervised learning

Support: Supervised learning requires input-output (labeled) training data.

4. Among the following option identify the one which is not a type of learning​

a) Semi unsupervised learning
b) Supervised learning
c) Unsupervised learning
d) Reinforcement learning

Answer: a) Semi unsupervised learning

Support: The standard term is "Semi-supervised learning", not "Semi unsupervised learning".

5. Which type of machine learning algorithm falls under the category of "unsupervised learning"?​

a) Linear Regression
b) K-means Clustering
c) Decision Trees
d) Random Forest

Answer: b) K-means Clustering

Support: K-Means groups unlabeled data into clusters and is an unsupervised algorithm.

6. Given the entropy for a split, E(after)=0.39 and entropy before split, E(before)=1. What is the Information Gain?​

a) 1
b) 0.39
c) 0.61
d) 2.56

Answer: c) 0.61

Support: Information Gain = E(before) − E(after) = 1 − 0.39 = 0.61.

7. Which of the following is a lazy learning algorithm?​

a) SVM
b) KNN
c) Random Forest
d) Decision tree

Answer: b) KNN

Support: KNN stores training data and performs computation only during prediction.

8. How are the new states generated in genetic algorithm?​

a) Composition
b) Mutation
c) Cross-over
d) Both Mutation & Cross-over

Answer: d) Both Mutation & Cross-over

Support: New offspring are generated using crossover and mutation operations.

9. Which machine learning algorithm is used for dimensionality reduction?​

a) Principal Component Analysis (PCA)
b) Naïve Bayes
c) Linear Regression
d) Decision tree

Answer: a) Principal Component Analysis (PCA)

Support: PCA reduces the number of features while preserving maximum variance.

10. Logistic regression is also called​

a) Login regression
b) Linear regression
c) Log regression
d) Logit regression

Answer: d) Logit regression

Support: Logistic Regression uses the logit (log-odds) function.

11. Which layer has feedback weights in competitive neural networks?​

a) Input layer
b) Second layer
c) Both input and second layer
d) None of the mentioned

Answer: b) Second layer

Support: Competitive neural networks generally use feedback connections in the output/competitive layer.

12. What do you mean by MDP in reinforcement learning?​

a) Markov discount procedure
b) Markov discount process
c) Markov deciding procedure
d) Markov decision process

Answer: d) Markov Decision Process

Support: MDP is the mathematical framework used to model reinforcement learning problems.

13. How many types of machine learning are there?​

a) 2
b) 3
c) 6
d) 5

Answer: b) 3

Support: The three main types are Supervised, Unsupervised, and Reinforcement Learning.

14. What is the objective of the backpropagation algorithm?​

a) to develop learning algorithm for multilayer feedforward neural network
b) to develop learning algorithm for single layer feedforward neural network
c) to develop learning algorithm for multilayer feedforward neural network, so that network can be trained to capture the mapping implicitly
d) none of the mentioned

Answer: c) to develop learning algorithm for multilayer feedforward neural network, so that network can be trained to capture the mapping implicitly

Support: Backpropagation trains multilayer neural networks by adjusting weights to minimize error.

15. Reinforcement learning is a __​

a) Prediction-based learning technique
b) Feedback-based learning technique
c) History results-based learning technique
d) none of these

Answer: b) Feedback-based learning technique

Support: The agent learns using rewards and penalties received from the environment.

2024​

1. Which of the following indicates cases where the classifier predicted 1 when it was actually 0?​

a) True positive
b) False positive
c) True negative
d) False negative

Answer: b) False positive

Support: The model predicts positive but the actual class is negative.

2. Which of the following is not a supervised machine learning algorithm?​

a) K-means
b) Naïve Bayes
c) SVM
d) Decision tree

Answer: a) K-means

Support: K-Means is an unsupervised clustering algorithm.

3. Which algorithm is best suited for a binary classification problem?​

a) K-nearest Neighbours
b) Logistic Regression
c) SVM
d) Linear Regression

Answer: b) Logistic Regression

Support: Logistic Regression is specifically designed for binary classification.

4. What is the key difference between supervised and unsupervised learning?​

Answer: a) Supervised learning uses labeled data, while unsupervised learning works with unlabeled data.

Support: Supervised learning uses labeled data, while unsupervised learning works with unlabeled data.

5. A single iteration over the entire training set is called as _.​

a) Cycle
b) Pass
c) Epoch
d) Clock

Answer: c) Epoch

Support: One complete pass through the training dataset is called an epoch.

6. The learner is trying to predict the cost of papaya based on its size. The variable "cost" is _.​

a) independent variable
b) target variable
c) ranked variable
d) categorical variable

Answer: b) Target variable

Support: Cost is the output being predicted.

7. What is the goal of gradient descent?​

a) Reduce complexity
b) Reduce overfitting
c) Maximize cost function
d) Minimize cost function

Answer: d) Minimize cost function

Support: Gradient descent iteratively reduces the error/loss function.

8. Which of the following statements is not true about SVM?​

a) It is memory efficient
b) It can address a large number of predictor variables
c) It is versatile
d) It doesn't require feature scaling

Answer: d) It doesn't require feature scaling

Support: SVM generally performs better when features are scaled.

9. The output of a neuron may vary from _.​

a) -∞ to +∞
b) -1 to 1
c) 0 to ∞
d) 1 to ∞

Answer: b) -1 to 1

Support: With bipolar activation functions, neuron outputs lie between -1 and 1.

10. Identify the parametric machine learning algorithm.​

a) CNN
b) KNN
c) Naïve Bayes
d) SVM

Answer: c) Naïve Bayes

Support: Naïve Bayes learns a fixed set of parameters from the training data.

11. What are general limitations of back propagation rule?​

a) local minima problem
b) slow convergence
c) scaling
d) all of the mentioned

Answer: d) all of the mentioned

Support: Backpropagation can suffer from local minima, slow learning, and scaling issues.

12. Which of the following statements is not true about the Decision tree?​

Answer: None of the options are clearly false.

Support: All four statements correctly describe decision trees. This appears to be a question-setting error.

13. Logistic regression is also named as _.​

a) Login regression
b) Linear regression
c) Log regression
d) Logit Regression

Answer: d) Logit Regression

Support: Logistic regression is based on the logit function.

14. Can a cancer detection problem (from an image dataset) be solved by logistic regression?​

a) Sometimes
b) no
c) Yes
d) Depends on the dataset

Answer: d) Depends on the dataset

Support: Logistic regression may work for some extracted features, but performance depends on the dataset and complexity.

15. Reinforcement Learning involves -​

a) an agent
b) An environment
c) Reward
d) All of the above

Answer: d) All of the above

Support: Reinforcement learning consists of an agent interacting with an environment and receiving rewards.