15 MCQs ML (DT Classification)

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 6

Multiple-choice questions (MCQs) related to Decision Tree

Classification:

Name:
Usama Javed
Abdul Basit
Muhammad Haseeb
Noor Muhammad

**Questions:**
1: What is the main advantage of using a decision tree for
classification?
A) Handles non-linear relationships well
B) Requires minimal training time
C) Works best with large datasets
D) Assumes normal distribution of data
**Answer:** A) Handles non-linear relationships well

2: In the context of decision trees, what is the purpose of pruning?


A) Enhance model complexity
B) Reduce overfitting
C) Increase training time
D) Improve accuracy
**Answer:** B) Reduce overfitting

3: How does the Gini index differ from entropy in decision tree
classification?
A) Gini index is less sensitive to outliers
B) Entropy is used for regression tasks
C) Gini index measures classification error
D) Entropy handles categorical features better
**Answer:** C) Gini index measures classification error

4: What is the significance of the information gain criterion in decision tree


splitting?
A) Maximizes the number of nodes
B) Minimizes the number of leaf nodes
C) Chooses the attribute that best separates the data
D) Balances the tree depth
**Answer:** C) Chooses the attribute that best separates the data

5: How does the decision tree algorithm handle missing values?


A) Removes instances with missing values
B) Assigns the mean of the feature to missing values
C) Ignores missing values during training
D) Creates a separate branch for missing values
**Answer:** D) Creates a separate branch for missing values

6: What is the purpose of feature scaling in decision tree classification?

2
A) Improves interpretability of the tree
B) Accelerates training time
C) Ensures all features contribute equally
D) Prevents overfitting
**Answer:** C) Ensures all features contribute equally

7: In decision tree pruning, what is pre-pruning?


A) Trimming nodes after the tree is fully grown
B) Trimming nodes during tree growth
C) Removing branches with the fewest instances
D) Adding nodes to increase accuracy
**Answer:** B) Trimming nodes during tree growth

8: What is the main challenge of using decision trees for imbalanced


datasets?

A) Prone to overfitting
B) Biased towards the majority class
C) Requires more computational resources
D) Ignores minority class
**Answer:** B) Biased towards the majority class

9: How does increasing the maximum depth of a decision tree affect its
performance?

3
A) Reduces overfitting
B) Increases model complexity
C) Improves generalization
D) Minimizes information gain
**Answer:** B) Increases model complexity

10: Which algorithm is an ensemble method combining multiple decision


trees?

A) ID3
B) Random Forest
C) C4.5
D) CART
**Answer:** B) Random Forest

11: What is the purpose of feature importance in decision tree models?

A) Identifying irrelevant features


B) Ranking features based on their contribution
C) Adjusting the learning rate
D) Controlling model complexity
**Answer:** B) Ranking features based on their contribution

12: How does reducing the minimum samples required to split a node
impact the decision tree?

4
A) Increases the depth of the tree
B) Decreases model complexity
C) Prunes the tree aggressively
D) Improves interpretability
**Answer:** A) Increases the depth of the tree

13: What is the primary limitation of decision trees in handling continuous


data?

A) Sensitivity to outliers
B) Overfitting
C) Inability to handle numerical values
D) Lack of interpretability
**Answer:** C) Inability to handle numerical values

14: In decision tree classification, what is a split point?

A) A threshold used to divide continuous features


B) The number of branches at each node
C) The depth of the tree
D) The proportion of samples in each leaf node
**Answer:** A) A threshold used to divide continuous features

15: What is the purpose of cross-validation in evaluating decision tree


models?

5
A) Assessing model performance on new data
B) Adjusting hyperparameters
C) Increasing training time
D) Reducing model complexity
**Answer:** A) Assessing model performance on new data

You might also like