decision trees

Question 1

why do people use decision trees

Answer 1

simpliscity and interperability
extendability

Question 2

what is the predictive model structure of a decision learning tree

Answer 2

root or internal node: a feature
leaf node : target value
branch : represents a decision rule

Question 3

what does a root or internal node represent in a decision learning tree

Answer 3

root or internal node: a feature

Question 4

what does a leaf node represent in a decision learning tree

Answer 4

leaf node : target value

Question 5

what does a branch represent in a decision learning tree

Answer 5

branch : represents a decision rule

Question 6

what are the types of decision trees

Answer 6

classification : target variable takes categorical data
regression : target variable takes continuous data

Question 7

what are the difference between classification and regression trees

Answer 7

classification : target variable takes categorical data
regression : target variable takes continuous data

Question 8

what are classification trees

Answer 8

DTs where target variable takes categorical data

Question 9

what are regression trees

Answer 9

DTs where target variable takes continuous data

Question 10

whgat does information gain for DTs do

Answer 10

measures reduction in entropy after splitting a dataset based on a feature (not features)

Question 11

what is the equation for information gain

Answer 11

I(Y,X) = H(Y) - H(Y|X)

Question 12

what does each part of the equation stand for in information gain
I(Y,X) = H(Y) - H(Y|X)

Answer 12

Y - random variable representing target
X - random variable representing feature of input sample
H(Y) - entropy of y
H(X|Y) - conditional entropy of y given x

Question 13

what does information gain do

Answer 13

quantifies improvement of classifying labels after using a feature to split the dataset
feature that maxmises information gain is chosen for the split