Chp 6 Geron

Question 1

Leaf node

Answer 1

Has no child nodes

Question 2

Node samples

Answer 2

Counts how many training instances it applies to

Question 3

Node value

Answer 3

Tells you how many training instances of each class this node applies to

Question 4

Node Gini

Answer 4

Measures its impurity
Pure=Gini=0

Question 5

Binary trees

Answer 5

Nonleaf nodes, always 2 children

Question 6

Id3

Answer 6

Produce decision tree nodes with more than 2 children

Question 7

P

Answer 7

Set of problems that can be solved in polynomial time

Question 8

NP

Answer 8

Set of problems whose solutions can be verified in polynomial time

Question 9

NP-hard

Answer 9

A problem to which any NP problem can be reduced in polynomial time

Question 10

NP-complete

Answer 10

Both NP and NP-hard

Question 11

Information gain

Answer 11

Reduction of entropy

Question 12

Nonparametric model

Answer 12

Number of parameters is not determined prior to training, model structure is free to stick closely to the data —overfitting

Question 13

Parametric model

Answer 13

Pre-determined number of parameters, degree of freedom is limited, reducing risk of overfitting, but increasing risk of under fitting

Question 14

Regularization

Answer 14

Restricting decisions tree’s freedom during training
Helps to avoid overfitting

Question 15

Min_samples_spilt

Answer 15

Minimum umber of samples a node effect it can be spilt

Question 16

Min_samples_leaf

Answer 16

Minimum number of samples a leaf node must have

Question 17

Min_weight_fraction_leaf

Answer 17

Minimum number of samples a leaf node must have but expressed as a fraction of the total number of weighted instances

Question 18

Max_leaf_nodes

Answer 18

Max number of leaf nodes

Question 19

max_features

Answer 19

Max number of features that are evaluated for splitting at each node