Predictive Data Mining

Question 1

Accuracy

Answer 1

Measure of classification success defined as 1 minus the overall error rate.

Question 2

Average error

Answer 2

The average difference between the actual values and the predicted values of observations in a data set.

Question 3

Bagging

Answer 3

An ensemble method that generates a committee of models based on random samples drawn with replacement and makes predictions based on the average prediction of the set of models.

Question 4

Bias

Answer 4

The tendency of a predictive model to overestimate or underestimate the value of a continuous outcome.

Question 5

Boosting

Answer 5

An ensemble method that iteratively samples from the original training data to generate individual models that target observations that were mispredicted in previously generated models. Its predictions are based on the weighted average of the predictions of the individual models, where the weights are proportional to the individual models’ accuracy.

Question 6

Class error rate

Answer 6

Percentage of observations of a given class misclassified by a model in a data set

Question 7

Classification confusion matrix

Answer 7

A matrix showing the counts of actual versus predicted class values.

Question 8

Classification tree

Answer 8

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

Question 9

Classification

Answer 9

A predictive data mining task requiring the prediction of an observation’s outcome class or category.

Question 10

Cumulative lift chart

Answer 10

A chart used to present how well a model performs in identifying observations most likely to be in a given class as compared with random classification.

Question 11

Cutoff value

Answer 11

The smallest value that the predicted probability of an observation can be for the observation to be classified as a given class.

Question 12

Decile-wise lift chart

Answer 12

A chart used to present how well a model performs at identifying observations for each of the top k deciles most likely to be in a given class versus a random selection.

Question 13

Ensemble method

Answer 13

A predictive data-mining approach in which a committee of individual classification or estimation models are generated and a prediction is made by combining these individual predictions.

Question 14

Estimation

Answer 14

A predictive data mining task requiring the prediction of an observation’s continuous outcome value.

Question 15

F1 score

Answer 15

A measure combining precision and sensitivity into a single metric.

Question 16

False negative

Answer 16

The misclassification of a positive observation as negative.

Question 17

False positive

Answer 17

The misclassification of a negative observation as positive

Question 18

Features

Answer 18

A set of input variables used to predict an observation’s outcome class or continuous outcome value.

Question 19

Impurity

Answer 19

Measure of the heterogeneity of observations in a classification tree.

Question 20

K-nearest neighbor (K-NN)

Answer 20

A classification method that classifies an observation based on the class of the k observations most similar or nearest to it.

Question 21

Logistic regression

Answer 21

A generalization of linear regression for predicting a categorical outcome variable.

Question 22

Mallow’s Cp statistic

Answer 22

A measure in which small values approximately equal to the number of coefficients suggest promising logistic regression models.

Question 23

Model overfitting

Answer 23

A situation in which a model explains random patterns in the data on which it is trained rather than just the relationships, resulting in training-set accuracy that far exceeds accuracy for the new data.

Question 24

Observation (record)

Answer 24

A set of observed values of variables associated with a single entity, often displayed as a row in a spreadsheet or database.

Question 25

Overall error rate

Answer 25

The percentage of observations misclassified by a model in a data set.

Question 26

Precision

Answer 26

The percentage of observations predicted to be in a given class that actually are in that class

Question 27

Random Forest

Answer 27

A variant of the bagging ensemble method that generates a committee of classification or regression trees based on different random samples but restricts each individual tree to a limited number of randomly selected features (variables)

Question 28

receiver operating characteristic (ROC) curve

Answer 28

A chart used to illustrate the tradeoff between a model’s ability to identify a given class’s observations and its complement class’s error rate.

Question 29

Regression tree

Answer 29

A tree that predicts values of a continuous outcome variable by splitting observations into groups via a sequence of hierarchical rules.

Question 30

Root mean square error

Answer 30

A measure of the accuracy of an estimation method defined as the square root of the sum of squared deviations between the actual values and predicted values of observations.

Question 31

Sensitivity; recall

Answer 31

The percentage of actual observations of a given class correctly identified, usually the positive class

Question 32

Specificity

Answer 32

The percentage of actual observations of a given class correctly identified; usually the negative class

Question 33

Supervised learning

Answer 33

Category of data-mining techniques in which an algorithm learns how to predict or classify an outcome variable of interest.

Question 34

Test set

Answer 34

Data set used to compute unbiased estimate of final predictive model’s accuracy.

Question 35

Training set

Answer 35

Data used to build candidate predictive models.

Question 36

Validation set

Answer 36

Data used to evaluate candidate predictive models.

Question 37

Variable (feature)

Answer 37

A characteristic or quantity of interest that can take on different values.