Machine Learning - Initialize Model

Question 1

One-Class Support Vector Machine:

Answer 1

Anomaly Detection

Creates a one-class support vector machine model for anomaly detection.

This module is particularly useful in scenarios where you have a lot of “normal” data and not many cases of the anomalies you are trying to detect. For example, if you need to detect fraudulent transactions, you might not have many examples of fraud that you could use to train a typical classification model, but you might have many examples of good transactions.

You use the One-Class Support Vector Model module to create the model, and then train the model using the Train Anomaly Detection Model. The dataset that you use for training can contain all or mostly normal cases.

You can then apply different metrics to identify potential anomalies. For example, you might use a large dataset of good transactions to identify cases that possibly represent fraudulent transactions.

Question 2

PCA-Based Anomaly Detection:

Answer 2

Anomaly Detection

Creates an anomaly detection model by using Principal Component Analysis.

This module helps you build a model in scenarios where it is easy to obtain training data from one class, such as valid transactions, but difficult to obtain sufficient samples of the targeted anomalies.

For example, to detect fraudulent transactions, very often you don’t have enough examples of fraud to train on, but have many examples of good transactions. The PCA-Based Anomaly Detection module solves the problem by analyzing available features to determine what constitutes a “normal” class, and applying distance metrics to identify cases that represent anomalies. This let you train a model using existing imbalanced data.

Question 3

Multiclass Decision Forest:

Answer 3

Classification

Creates a multiclass classification model by using the decision forest algorithm.

Question 4

Multiclass Decision Jungle:

Answer 4

Classification

Creates a multiclass classification model by using the decision jungle algorithm.

Question 5

Multiclass Logistic Regression:

Answer 5

Classification

Creates a multiclass logistic regression classification model.

Question 6

Multiclass Neural Network:

Answer 6

Classification

Creates a multiclass classification model by using a neural network algorithm.

Question 7

One-vs-All Multiclass:

Answer 7

Classification

Creates a multiclass classification model from an ensemble of binary classification models.

Question 8

Two-Class Averaged Perceptron:

Answer 8

Classification

Creates an averaged perceptron binary classification model.

Question 9

Two-Class Bayes Point Machine:

Answer 9

Classification

Creates a Bayes point machine binary classification model.

Question 10

Two-Class Boosted Decision Tree:

Answer 10

Classification

Creates a binary classifier by using a boosted decision tree algorithm.

Question 11

Two-Class Decision Forest:

Answer 11

Classification

Creates a two-class classification model by using the decision forest algorithm.

Question 12

Two-Class Decision Jungle:

Answer 12

Classification

Creates a two-class classification model by using the decision jungle algorithm.

Question 13

Two-Class Locally Deep Support Vector Machine:

Answer 13

Classification

Creates a binary classification model by using the locally deep Support Vector Machine algorithm.

Question 14

Two-Class Logistic Regression:

Answer 14

Classification

Creates a two-class logistic regression model.

Question 15

Two-Class Neural Network:

Answer 15

Classification

Creates a binary classifier by using a neural network algorithm.

Question 16

Two-Class Support Vector Machine:

Answer 16

Classification

Creates a binary classification model by using the Support Vector Machine algorithm.

Question 17

K-Means Clustering:

Answer 17

Clustering

Configures and initializes a K-means clustering model.

Question 18

Bayesian Linear Regression:

Answer 18

Creates a Bayesian linear regression model.

Question 19

Boosted Decision Tree Regression:

Answer 19

Creates a regression model by using the Boosted Decision Tree algorithm.

Question 20

Decision Forest Regression:

Answer 20

Creates a regression model by using the decision forest algorithm.

Question 21

Fast Forest Quantile Regression:

Answer 21

Creates a quantile regression model.

Question 22

Linear Regression:

Answer 22

Creates a linear regression model.

Question 23

Neural Network Regression:

Answer 23

Creates a regression model by using a neural network algorithm.

Question 24

Ordinal Regression:

Answer 24

Creates an ordinal regression model.

Question 25

Poisson Regression:

Answer 25

Creates a regression model that assumes data has a Poisson distribution.