Deck 9: Predictive Data Mining

Question 1

Misclassifying an actual __________ observation as a(n) __________ observation is known as a false positive.

A)Class 0, Class 1
B)Class 1, Class 0
C)error, accuracy
D)false, true

Answer 1

Class 0, Class 1

Question 2

__________ is dividing the sample data into three sets for training, validation, and testing of the data mining algorithm performance.

A)Data sampling
B)Data partitioning
C)Data preparation
D)Model assessment

Answer 2

Data partitioning

Question 3

A(n) __________ is often displayed as a row of values in a spreadsheet or database in which the columns correspond to the variables.

A)record
B)data point
C)classification
D)location

Answer 3

record

Question 4

The percent of misclassified records out of the total records in the validation data is known as the

A)overall error rate.
B)error.
C)accuracy.
D)class.

Question 5

As we increase the cutoff value, _______ error will decrease and _________ error will rise.

A)Class 0, Class 1
B)Class 1, Class 0
C)false, true
D)None of these are correct.

Question 6

__________ involves descriptive statistics, data visualization, and clustering.

A)Data exploration
B)Data partitioning
C)Data preparation
D)Model assessment

Question 7

__________ is one minus the Class 0 error rate.

A)Sensitivity
B)Specificity
C)Accuracy
D)Cutoff value

Question 8

__________ is the manipulation of the data with the goal of putting it in a form suitable for formal modeling.

A)Data sampling
B)Data partitioning
C)Data preparation
D)Model assessment

Question 9

__________ is the step in data mining that includes addressing missing and erroneous data, reducing the number of variables, defining new variables, and data exploration.

A)Data sampling
B)Data partitioning
C)Data preparation
D)Model assessment

Question 10

__________ is NOT a step of the data mining process.

A)Data sampling
B)Data partitioning
C)Model construction
D)Supervised learning

Question 11

Determine a freshman's likely first-year grade point average from the student's Scholastic Aptitude Test (SAT) score, high school grade point average, and number of extra-curricular activities. This is an example of

A)classification of a categorical outcome.
B)estimation of a continuous outcome.
C)prediction of a categorical outcome.
D)unsupervised learning.

Question 12

A characteristic or quantity of interest that can take on different values is a(n)

A)variable.
B)observation.
C)record.
D)quality.

Question 13

__________ is a category of data mining techniques in which an algorithm learns how to classify or estimate an outcome variable of interest.

A)Supervised learning
B)Unsupervised learning
C)Dimension reduction
D)Data sampling

Question 14

Applying descriptive statistics and data visualization to the training set to understand the data and assist in the selection of an appropriate technique is a part of

A)data exploration.
B)data partitioning.
C)data preparation.
D)model assessment.

Question 15

Data mining methods for classifying or estimating an outcome based on a set of input variables is referred to as

A)supervised learning.
B)unsupervised learning.
C)dimension reduction.
D)data sampling.

Question 16

The set of recorded values of variables associated with a single entity is a(n)

A)observation.
B)data point.
C)classification.
D)location.

Question 17

Estimation methods are also referred to as

A)prediction methods.
B)clustering methods.
C)association methods.
D)supervised methods.

Question 18

Data used to build a data mining model is called

A)validation data.
B)training data.
C)test data.
D)exploration data.

Question 19

Classifying a record as belonging to one class when it belongs to another class is referred to as a(n)

A)overall error rate.
B)error.
C)accuracy.
D)class.

Question 20

__________ is a method of extracting data relevant to the business problem under consideration. It is the first step in the data mining process.

A)Data sampling
B)Data partitioning
C)Model construction
D)Model assessment

Question 21

A __________ classifies a categorical outcome variable by splitting observations into groups via a sequence of hierarchical rules.

A)regression tree
B)scatter chart
C)classification tree
D)confusion matrix

Question 22

The x-axis of a lift chart shows

A)the number of actual Class 1 records identified.
B)the ratio of decile mean to overall mean.
C)the number of actual Class 1 records.
D)the ratio of the overall mean to the decile mean.

Question 23

A(n) __________ matrix displays a model's correct and incorrect classification.

A)cumulative lift
B)confusion
C)decile-wise lift chart
D)ROC curve

Question 24

_________ attempts to classify a categorical outcome as a linear function of explanatory variables.

A)Linear regression
B)Logistic regression
C)Classification model
D)Supervised learning

Question 25

How many Class 1's are correctly classified as Class 1 in the Table below? $$\begin{array} { | l | c | c | } \hline { \text { Confusion Matrix } } \\\hline & { \text { Predicted Class } } \\\hline \text { Actual Class } & \mathbf { 1 } & \mathbf { 0 } \\\hline \mathbf { 1 } & 221 & 100 \\\hline \mathbf { 0 } & 30 & 3,000 \\\hline\end{array}$$ ?

A)221
B)100
C)30
D)3,000

Question 26

Which of the following is a commonly used supervised learning method?

A)k-means clustering
B)k-nearest neighbors
C)hierarchical clustering
D)association rule development

Question 27

The impurity of a group of observations is based on the variance of the outcome value for the observations in the group for

A)regression trees.
B)time-series plots.
C)classification trees.
D)cumulative lift charts.

Question 28

How many Class 1's are incorrectly classified as Class 0? $$\begin{array} { | l | c | c | } \hline { \text { Confusion Matrix } } \\\hline & { \text { Predicted Class } } \\\hline \text { Actual Class } & \mathbf { 1 } & \mathbf { 0 } \\\hline \mathbf { 1 } & 221 & 100 \\\hline \mathbf { 0 } & 30 & 3,000 \\\hline\end{array}$$ ?

A)221
B)100
C)30
D)3,000

Question 29

An observation classified as part of a group with a characteristic when it actually does not have the characteristic is termed as a(n)

A)false negative.
B)false positive.
C)residual.
D)outlier.

Question 30

The y-axis of a decile chart shows

A)number of important class records identified.
B)ratio of decile mean to overall mean.
C)the number of actual Class 1 records.
D)the ratio of the overall mean to the decile mean.

Question 31

__________ compares the number of actual Class 1 observations identified if considered in decreasing order of their estimated probability if randomly classified.

A)Cumulative lift
B)Confusion
C)Decile-wise lift chart
D)ROC curve

Question 32

__________ refers to the scenario in which the analyst builds a model that does a great job of explaining the sample of data on which it is based but fails to accurately predict outside the sample data.

A)Underfitting
B)Overfitting
C)Oversampling
D)Undersampling

Question 33

A test set is the data set used to

A)build the data mining model.
B)estimate performance of candidate models on unseen data.
C)estimate performance of the final model on unseen data.
D)show counts of actual versus predicted class values.

Question 34

One minus the overall error rate is often referred to as the __________ of the model.

A)sensitivity
B)accuracy
C)specificity
D)cutoff value

Question 35

Separate error rates with respect to the false negative and false positive cases are computed to take into account the

A)asymmetric costs in misclassification.
B)symmetric weights of these two cases.
C)distortions due to outliers.
D)effect of sampling error.

Question 36

__________ is a measure of the heterogeneity of observations in a classification tree.

A)Sensitivity
B)Specificity
C)Accuracy
D)Impurity

Question 37

__________ is a generalization of linear regression for predicting a categorical outcome variable.

A)Multiple linear regression
B)Logistic regression
C)Discriminant analysis
D)Cluster analysis

Question 38

In the k-nearest neighbors method, when the value of k is set to 1

A)the classification or prediction of a new observation is based solely on the single most similar observation from the training set.
B)the new observation's class is naïvely assigned to the most common class in the training set.
C)the new observation's prediction is used to estimate the anticipated error rate on future data over the entire training set.
D)the classification or prediction of a new observation is subject to the smallest possible classification error.