DT and RF

Front

What is the maximum possible height of a decision tree when using information gain as the splitting criterion?

Back

number of attributes

Front

True/False? We can use the same attribute multiple times as a test attribute in the same decision tree.

Back

True

Front

We can represent a decision tree as a set of rules. True/False?

Back

True

Front

Gain Ratio tends to favor attributes that can create a high branching factor, even if their predictive power is not inherently superior. True/False?

Back

False

Front

Which of the following has entropy of 0? Options: 2 True, 2 False; 4 True, 0 False; 3 True, 1 False

Back

4 True, 0 False

Front

How does Gain Ratio address the bias of Information Gain?

Back

By normalizing Information Gain with respect to the intrinsic information of the attribute.

Front

In a random forest, how is the randomness introduced? A. By randomly selecting a subset of data points for each tree. B. By using a random combination of decision tree algorithms. C. By randomly selecting a subset of features for each split. Both A and C Both A and B

Back

Both A and C

Front

What is the primary advantage of ensemble learning methods such as Random Forest?

Back

Higher resilience to overfitting.

Front

What is "bagging" in the context of ensemble learning?

Back

Training models on randomly sampled subsets of data with replacement.

Front

If a decision tree is allowed to grow without any constraints, what is a common outcome?

Back

The tree will overfit the training data.