XAI - Assignment II

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

• It can be applied on binary classification problems only

• It is a predictor that predicts the label associated with an instance by traveling from a root node of a tree to a leaf

• At each node, the successor child is chosen on the basis of a splitting of the input space

• The splitting is based on one of the features or on a predefined set of splitting rules

Consider the figure. If person A starts driving at 8:30 AM and there are no other vehicles on the road, and another person B starts driving at 10 AM and there is an accident on the road, what will be the commute time of A and B respectively?

• LONG, LONG

• LONG, SHORT

• SHORT, LONG

• SHORT, SHORT

What does the following figure represent?

• Decision tree for OR

• Decision tree for AND

• Decision tree for XOR

• Decision tree for XNOR

What is the main advantage of using the Random Forest algorithm?

• It always gives 100% accuracy

• It reduces overfitting and improves generalization

• It requires only one decision tree for better performance

• It works only for classification problems

How does Random Forest prevent overfitting?

• By using a single deep decision tree

• By selecting all features at each split

• By averaging predictions from multiple trees trained on different subsets of data

• By using only the most important features

How does Random Forest select features at each split in a tree?

• It considers all features every time

• It uses only the most correlated features

• It removes features with missing values

• It selects a random subset of features

What is a key characteristic of Random Forest?

• 
  It requires extensive data preprocessing

• It works only with numerical data

• It is a supervised learning algorithm

• 
  It cannot handle missing values

A rule-based classifier uses a set of IF-THEN rules for classification. Which of the following is the correct form of IF-THEN rule?

• IF conclusion THEN condition

• IF THEN condition conclusion

• IF condition THEN conclusion

• None of the above

Which of the following statements is TRUE regarding rule-based classifiers?

• Rules are always mutually exclusive.

• Decision boundaries are always non-linear.

• They use a set of "if-then" rules to make class decisions.

• They are primarily used for regression tasks.

In the context of rule-based classification, what does a "decision list" imply?

• A set of unordered rules with no priority.

• A list of rules where the order of rules matters, and the first matching rule determines the class.

• A list of rules where all matching rules contribute equally to the final decision.

• A system that only uses a single rule for classification.