Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!
Get Started for Free

scutr

Balancing Multiclass Datasets for Classification Tasks

Imbalanced training datasets impede many popular classifiers. To balance training data, a combination of oversampling minority classes and undersampling majority classes is useful. This package implements the SCUT (SMOTE and Cluster-based Undersampling Technique) algorithm as described in Agrawal et. al. (2015) <doi:10.5220/0005595502260234>. Their paper uses model-based clustering and synthetic oversampling to balance multiclass training datasets, although other resampling methods are provided in this package.

Functions (13)

SCUT

SMOTE and cluster-based undersampling technique.

bullseye

An imbalanced dataset with a minor class centered around the origin with a majority class surrounding the center.

imbalance

An imbalanced dataset with randomly placed normal distributions around the origin. The nth class has n * 10 observations.

oversample_smote

Oversample a dataset by SMOTE.

resample_random

Randomly resample a dataset.

sample_classes

Stratified index sample of different values in a vector.

undersample_hclust

Undersample a dataset by hierarchical clustering.

undersample_kmeans

Undersample a dataset by kmeans clustering.

undersample_mclust

Undersample a dataset by expectation-maximization clustering

undersample_mindist

Undersample a dataset by iteratively removing the observation with the lowest total distance to its neighbors of the same class.

undersample_tomek

Undersample a dataset by removing Tomek links.

validate_dataset

Validate a dataset for resampling.

wine

Type and chemical analysis of three different kinds of wine.

scutr

Balancing Multiclass Datasets for Classification Tasks

v0.1.2
MIT + file LICENSE
Authors
Keenan Ganz [aut, cre]
Initial release

We don't support your browser anymore

Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.