In each iteration, samples one minority class element x1, then one of x1's nearest neighbors: x2.
Both points are now interpolated / convex-combined, resulting in a new virtual data point x3
for the minority class.<p></p><p>The method handles factor features, too. The gower distance is used for nearest neighbor
calculation, see <code><a href="/link/daisy?package=mlr&version=2.0&to=cluster" rd-options="cluster" data-mini-rdoc="cluster::daisy">daisy</a></code>.
For interpolation, the new factor level for x3
is sampled from the two given levels of x1 and x2 per feature.</p>

smote

Interface to a large number of classification and regression
    techniques, including machine-readable parameter descriptions. There is
    also a n experimental extension for surival analysis and cost-sensitive
    learning. Generic resampling, including cross-validation, bootstrapping and
    subsampling. Hyperparameter tuning with modern optimization techniques.
    Filter and wrapper methods for feature selection. Extension of basic
    learners with additional operations common in machine learning.

smote: Synthetic Minority Oversampling Technique to handle class imbalancy in binary classification.

Description

Usage

Arguments

Value

See Also