Question 1

In a decision tree, the recursive process of partitions continues and only terminates when the Gini index reaches 0.5.

Accepted Answer

A) True 
 B)False

Question 2

Small changes in the training set, while using the CART algorithm, will result in drastically different trees.

Accepted Answer

A) True 
 B)False

Question 3

In a random forest model, as a guideline the user needs to select a number of the random features for each tree. If there are 196 predictor variables in the data, each tree will randomly select how many features to be included in the tree?

Accepted Answer

A)  5 
B)  18 
C)  196 
D)  14 
A)  5 
B)  18 
C)  196 
D)  14

Question 4

In R, to determine the number of splits in the default classification tree, the rpart function uses what to determine when to stop growing the tree?

Accepted Answer

A)  nsplit 
B)  complexity parameter 
C)  prune 
D)  predict 
A)  nsplit 
B)  complexity parameter 
C)  prune 
D)  predict

Question 5

Ensemble tree models combine multiple single-tree models to reduce the variation in prediction error. Of the strategies, which may lead to overfitting?

Accepted Answer

A)  boosting 
B)  random Forest 
C)  bagging 
D)  banking 
A)  boosting 
B)  random Forest 
C)  bagging 
D)  banking

Question 6

A subset with the highest degree of impurity is when a 50% and 50% split occur between classes.

Accepted Answer

A) True 
 B)False

Question 7

Robin wanted to know if the age partition chosen for her data was the best fit for her 20 case, 90% Class 1, 10% Class 0 partition. She completed the Gini impurity index with the results of (Age < 32) = 0.2312 and (Age $\ge$ 32) = 0.2934. What is the weighted combination and what did partition at Age 32 produce?

Accepted Answer

A)  Robin was able to reduce the Gini index from 0.2934 to 0.2641 confirming the best split for age. 
B)  Robin was able to reduce the Gini index from 0.2934 to 0.20 confirming the best split for age. 
C)  Robin was able to reduce the Gini index from 0.2934 to 0.2374 confirming the best split for age. 
D)  Robin realized with the 0.2641 weighted average, the age split was not the best split for the age range. 
A)  Robin was able to reduce the Gini index from 0.2934 to 0.2641 confirming the best split for age. 
B)  Robin was able to reduce the Gini index from 0.2934 to 0.20 confirming the best split for age. 
C)  Robin was able to reduce the Gini index from 0.2934 to 0.2374 confirming the best split for age. 
D)  Robin realized with the 0.2641 weighted average, the age split was not the best split for the age range.

Question 8

Using the following chart for age and income, determine the split points for income.

Accepted Answer

A)  {33000, 36000, 41000, 47000, 51500, 57500, 63000} 
B)  {33000, 41000, 47000, 51500, 63000} 
C)  {36000, 47000, 57500, 63000} 
D)  {36000, 41000, 47000, 51500, 57500} 
A)  {33000, 36000, 41000, 47000, 51500, 57500, 63000} 
B)  {33000, 41000, 47000, 51500, 63000} 
C)  {36000, 47000, 57500, 63000} 
D)  {36000, 41000, 47000, 51500, 57500}

Question 9

When using the CART algorithm, the Gini index is used in the classification tree, however in a regression tree, _____ is used to measure impurity.

Accepted Answer

A)  mean percentage error 
B)  mean squared error 
C)  mean absolute deviation 
D)  mean absolute percentage error 
A)  mean percentage error 
B)  mean squared error 
C)  mean absolute deviation 
D)  mean absolute percentage error

Question 10

If 73% of the cases belong to Class 0 and 27% belong to Class 1, what is the Gini index?

Accepted Answer

A)  0.39 
B)  0 
C)  0.54 
D)  0.15 
A)  0.39 
B)  0 
C)  0.54 
D)  0.15

Question 11

When generating a single regression tree visually, the prp function is used. Based on the following example code, what does setting type = 1 mean?
>prp(default_tree, type = 1, extra = 1, under = TRUE)

Accepted Answer

A)  type = 1 argument is the number of observations that fall into each node displayed. 
B)  type = 1 argument places the number of cases under each decision node in the diagram. 
C)  type = 1 argument allows for all nodes, except leaf nodes, to be labeled in the diagram. 
D)  type = 1 argument allows for the predicting variable to be displayed in root node. 
A)  type = 1 argument is the number of observations that fall into each node displayed. 
B)  type = 1 argument places the number of cases under each decision node in the diagram. 
C)  type = 1 argument allows for all nodes, except leaf nodes, to be labeled in the diagram. 
D)  type = 1 argument allows for the predicting variable to be displayed in root node.

In a decision tree, the recursive process of partitions continues and only terminates when the Gini index reaches 0.5.

Small changes in the training set, while using the CART algorithm, will result in drastically different trees.

In a random forest model, as a guideline the user needs to select a number of the random features for each tree. If there are 196 predictor variables in the data, each tree will randomly select how many features to be included in the tree?

In R, to determine the number of splits in the default classification tree, the rpart function uses what to determine when to stop growing the tree?

Ensemble tree models combine multiple single-tree models to reduce the variation in prediction error. Of the strategies, which may lead to overfitting?

A subset with the highest degree of impurity is when a 50% and 50% split occur between classes.

Using the following chart for age and income, determine the split points for income.

When using the CART algorithm, the Gini index is used in the classification tree, however in a regression tree, _____ is used to measure impurity.

If 73% of the cases belong to Class 0 and 27% belong to Class 1, what is the Gini index?

When generating a single regression tree visually, the prp function is used. Based on the following example code, what does setting type = 1 mean? >prp(default_tree, type = 1, extra = 1, under = TRUE)

Introduction to Business Analytics

Data Management and Wrangling

Data Visualization and Summary Measures

Probability and Probability Distributions

Statistical Inference

Regression Analysis

Advanced Regression Analysis

Introduction to Data Mining

Supervised Data Mining: K-Nearest Neighbors and Naãve Bayes

Unsupervised Data Mining

Forecasting With Time Series Data

Introduction to Prescriptive Analytics

Filters

Exam 10: Supervised Data Mining: Decision Trees

In a decision tree, the recursive process of partitions continues and only terminates when the Gini index reaches 0.5.

Small changes in the training set, while using the CART algorithm, will result in drastically different trees.

In a random forest model, as a guideline the user needs to select a number of the random features for each tree. If there are 196 predictor variables in the data, each tree will randomly select how many features to be included in the tree?

In R, to determine the number of splits in the default classification tree, the rpart function uses what to determine when to stop growing the tree?

Ensemble tree models combine multiple single-tree models to reduce the variation in prediction error. Of the strategies, which may lead to overfitting?

A subset with the highest degree of impurity is when a 50% and 50% split occur between classes.

Using the following chart for age and income, determine the split points for income.

When using the CART algorithm, the Gini index is used in the classification tree, however in a regression tree, _____ is used to measure impurity.

If 73% of the cases belong to Class 0 and 27% belong to Class 1, what is the Gini index?

When generating a single regression tree visually, the prp function is used. Based on the following example code, what does setting type = 1 mean? >prp(default_tree, type = 1, extra = 1, under = TRUE)

Introduction to Business Analytics

Data Management and Wrangling

Data Visualization and Summary Measures

Probability and Probability Distributions

Statistical Inference

Regression Analysis

Advanced Regression Analysis

Introduction to Data Mining

Supervised Data Mining: K-Nearest Neighbors and Naãve Bayes

Unsupervised Data Mining

Forecasting With Time Series Data

Introduction to Prescriptive Analytics

Filters