Analyses

SVM classification

Support-vector classifier with a holdout confusion matrix and, for two classes, an ROC.

Edit on GitHub

When to use it

Use SVM classification when the outcome is a category and you want a holdout confusion matrix without trees. Random forest of the same problem is random forest classification.

Assumptions

The outcome is categorical. Features are numeric. Classes with fewer than two rows are dropped. Kernel, C, and class weights are not exposed. Holdout fraction 0.25; HTTP and chat dispatch both honour test_fraction.

Running it in Tensr

ML → Classification → Support Vector Machines (Classification). In chat: “SVM of passed on hours and anxiety.”
Target is the categorical column. Features need at least one numeric column.

Options

Prop

Type

Reading the output

The same strong pass/fail sample and the same two features. Kernel and C are not user controls.

Actual \ Predicted01
0303
11314

Train accuracy = 0.722, test accuracy = 0.733, n = 240. Hours was built to move the pass rate strongly, so holdout accuracy should sit clearly above chance.

Reporting (APA 7)

Holdout classification of passed from hours and anxiety, n = 240. Test accuracy = 0.733, train accuracy = 0.722.

Coming from SPSS

SPSS has no SVM dialog. Tensr’s item is on the ML menu, Classification → Support Vector Machines (Classification). The path string stored for this item says ML → Classification → SVM.

Random forest classification and gradient boosting are the tree ensembles.