Machine Learning with Viya® 3.4® Lesson 5: Support Vector Machines (SVG) and Additional Topics

Question 1

What is a dot product?

Answer 1

A dot product is a way to multiply vectors that result in a scalar, or a single number, as the answer. It is an element-by-element multiplication, and then a sum across the products.

Question 2

How is a support vector machine constructed in order to avoid the curse of dimensionality?

Answer 2

By using only the observations closest to the separating hyperplane

Question 3

How does using only the observations closest to the separating hyperplane avoid the curse of dimensionality?

Answer 3

By limiting the number of data points in the solution.

Question 4

What kind of information is in the Training Results table in an SVM run?

Answer 4

The Training Results table shows the parameters for the final Support Vector Machine model such as the number of support vectors and the bias.

Question 5

Where can you find the average square error on the VALIDATE partition?

Answer 5

In the Fit Statistics table on the Assessment tab.

Question 6

Where can you view the misclassification matrix?

Answer 6

The Output Window

Question 7

What are the two constraints used to solve for optimization in a support vector machine?

Answer 7

If the target variable equals 1, then H must be greater than or equal to 1. If the target is -1, then H must be less than or equal to -1.

Question 8

What’s a term for describing data points that are not linearly separable?

Answer 8

soft margin hyperplane

Question 9

What do you need to do when you encounter a soft margin hyperplane?

Answer 9

Account for errors that the separating hyperplane might make

Question 10

TRUE or FALSE: When the data are not linearly separable, the process of optimizing the location of the hyperplane must account for classification errors.

Answer 10

TRUE: When the data are not linearly separable, the hyperplane will misclassify some data points. In this situation, the process of optimizing the location of the hyperplane must account for these classification errors.

Question 11

What is a kernel function?

Answer 11

A kernel function operates as a dot product in a higher dimension (that is, in a feature space), but it is applied to the raw data.

Question 12

Suppose you are modeling data with a binary target and three inputs. The data are linearly separable. How many possible solutions exist that classify the target?

Answer 12

an infinite number of solutions can classify the binary target when the data are linearly separable

Question 13

What type of target variable is supported in a support vector machine in Model Studio?

Answer 13

Support vector machines are used exclusively with binary targets in Model Studio.

Question 14

What are the elements of a classifier model for a Support Vector Machine?

Answer 14

The classifier model (H) has two elements: a normal vector and a bias term

Question 15

What is the maximum-margin hyperplane in a two-dimensional input space?

Answer 15

the exact center of the thickest line that touches the innermost values of one target outcome and the innermost values of the other target outcome

Question 16

What are support vectors?

Answer 16

Support vectors are the points in the data that are closest to the maximum-margin hyperplane.

Question 17

In support vector machines, finding the separating hyperplane is an optimization problem with constraints that involve the values of the binary target.

a. True
b. False

Answer 17

A: True

Solving for the support vector machine is actually an optimization problem with two constraints. The first constraint is based on a target value of +1, and the second constraint is based on a target value of -1.

Question 18

How is a feature space is constructed?

Answer 18

A feature space is constructed by applying a nonlinear transformation to data so that linear separation exists in this higher-dimensional space

Question 19

What is a kernel function?

Answer 19

C: A kernel function is a math trick used to avoid having to calculate dot products on transformed data.

Question 20

What information is provided in the Local Interpretable Model-Agnostic Explanation (LIME) plot when the Model Interpretability feature is used in Model Studio?

Answer 20

A LIME plot creates a localized linear regression model around a particular observation based on a perturbed sample set of data.

Question 21

How is the Input Relative Importance table that appears in the results calculated when the Model Interpretability feature is used in Model Studio?

Answer 21

The Input Relative Importance table is calculated by depth-one decision trees using each input to estimate the predicted values of the model being interpreted.

Question 22

What are the three options used to increase the flexibility of a support vector machine model in Model Studio?

Answer 22

Penalty, kernel, and tolerance

Question 23

What is the penalty term?

Answer 23

The penalty is a term that accounts for misclassification errors in model optimization.

Question 24

What is tolerance?

Answer 24

The tolerance value balances the number of support vectors and model accuracy.

Question 25

Which of the following machine learning models is the easiest to interpret?

a. decision tree
b. neural network
c. support vector machine

Answer 25

Decision trees are highly interpretable because they are based on English rules, which are rules that use Boolean logic.

Question 26

What does the Penalty value do?

Answer 26

The Penalty value balances model complexity and training error.

Question 27

What is the risk associated with a larger Penalty value?

Answer 27

A larger Penalty value creates a more robust model at the risk of overfitting the training data.

Question 28

What does the Tolerance value do?

Answer 28

The Tolerance value balances the number of support vectors and model accuracy.

Question 29

What is the consequence of too large a Tolerance value?

Answer 29

A Tolerance value that is too large creates too few support vectors.

Question 30

What is the consequence of a Tolerance value that is too small?

Answer 30

A Tolerance value that is too small overfits the training data.

Question 31

What does an intersecting slope in an ICE Plot indicate?

Answer 31

An Intersecting slope indicates that there is an interaction between the plot variable and one or more complementary variables.

Question 32

Why is it useful to look among clusters for different relationships between the groups (or levels) of the categorical variable and the target when evaluating an ICE plot of a categorical input?

Answer 32

Significant differences in these relationships indicate group effects.

Question 33

Where is the largest possible margine of error in a minimum-maximum hyperplane?

Answer 33

This hyperplane has the largest possible margin of error on its positive and negative sides.

Question 34

What does the autoencoder method on the Feature Extraction node do?

Answer 34

The Autoencoder method builds a neural network that uses the inputs to reconstructs the inputs.

Question 35

How is an autoencoder network different than an MLP network?

Answer 35

An autoencoder network is like an MLP network except that its output layer is duplicated from the input layer.

Question 36

When you scale the input variables for a binary target using support vector machines, what happens to the inputs?

Answer 36

Values are scaled to range from 0 to 1.

Question 37

How are missing values for class variables handled when “Use missing” is specified for a Support Vector Machine node?

Answer 37

SVMs treat missing values as a separate category.

Question 38

What is the only Global Interpretability plot available in Model Studio?

Answer 38

Partial Dependence plots are based on an aggregation across all observations, thus they provide global interpretability

Question 39

Which model interpretability tools can be used to help interpret a machine learning model for a single observation?

Answer 39

Local Interpretable Model-Agnostic Explanations (LIME) plots, Kernel SHAP (Shapley) plots

Question 40

Where does the Open Source Code node execute?

Answer 40

CAS

Question 41

Which assessment measure should be used to determine the champion when predicting an interval target?

Answer 41

Average Squared Error

Question 42

Which assessment measure should be used to determine the champion for a decision focused model?

Answer 42

Misclassification Rate

Question 43

Which assessment measure would you use to determine the champion for a model used for ranking?

Answer 43

The ROC Index or Gini Coefficient