Learn R Programming

smotefamily (version 1.4.0)

A Collection of Oversampling Techniques for Class Imbalance Problem Based on SMOTE

Description

A collection of various oversampling techniques developed from SMOTE is provided. SMOTE is a oversampling technique which synthesizes a new minority instance between a pair of one minority instance and one of its K nearest neighbor. Other techniques adopt this concept with other criteria in order to generate balanced dataset for class imbalance problem.

Copy Link

Version

Install

install.packages('smotefamily')

Monthly Downloads

15,569

Version

1.4.0

License

GPL (>= 3)

Maintainer

Wacharasak Siriseriwan

Last Published

March 14th, 2024

Functions in smotefamily (1.4.0)

ANS

Adaptive Neighbor Synthetic Majority Oversampling TEchnique
kncount

Counting the number of each class in K nearest neighbor
gap

The function to provide a random number which is used as a location of synthetic instance
knearest

The function to find n_clust nearest neighbors of each instance, always removing the index of that instance if it is reported.
n_dup_max

The function to calculate the maximum round each sampling is repeated
sample_generator

The function to generate 2-dimensional dataset
SMOTEfamily

SMOTE family package for Data Generation
SMOTE

Synthetic Minority Oversampling TEchnique
Borderline-SMOTE

Borderline-SMOTE
RSLS

Relocating Safe-level SMOTE
SLS

Safe-level SMOTE
ADASYN

Adaptive Synthetic Sampling Approach for Imbalanced Learning
DBSMOTE

Density-based SMOTE