Question

Which best describes the concept of information gain in decision tree construction?

A) The decision associated with a non-leaf node maximizes the entropy value.

B) The decision associated with a non-leaf node minimizes the entropy value.

C) As we descend the trees the subset of training data that satisfies the decisions associated with non-leaf nodes becomes more homogenous.

D) None of the above.

Answer 1

Answer)
The following is the statement which is used to best describe the concept of the information gain in the decision tree construction:
C) As we descend the trees the subset of training data that satisfies the decisions associated with non-leaf nodes becomes more homogeneous.

The information gain is the one that depends on the decrease in entropy when there is a data set split based on attributes. The highest information gain finding will be found in the decision tree for the most homogeneous branches. Thus when we descend the training data subset the decisions with the non-leaf nodes are more homogeneous.

**Please Hit Like if you appreciate my answer. For further doubts on the answer please drop a comment, I'll be happy to help. Thanks for posting.**