smote_and_undersample: SMOTE oversampling and undersampling
In hyperSMURF: Hyper-Ensemble Smote Undersampled Random Forests

Description Usage Arguments Details Value See Also Examples

Function to both oversample by SMOTE the minority class and undersample the majority class

1	smote_and_undersample(data, y, fp = 1, ratio = 1, k = 5)

`data`	a data frame or matrix. Rows: examples; columns: features
`y`	a factor with the labels. 0:majority class, 1: minority class.
`fp`	multiplicative factor for the SMOTE oversampling of the minority class. If fp<1 no oversampling is performed.
`ratio`	ratio of the #majority/#minority
`k`	number of the nearest neighbours for SMOTE oversampling (def. 5)

If n is the number of examples of the minority class, then fp*n new synthetic examples are generated according to the SMOTE algorithm and ratio*(fp*n + n) negative examples are undersampled form the majority class.

A list with two entries:

`X`	a data frame including the original minority class examples plus the SMOTE oversampled and undersampled data
`Y`	a factor with the labels of the data frame

smote

1 2	d <- imbalanced.data.generator(n.pos=20, n.neg=1000, n.features=12, n.inf.features=2, sd=1, seed=1); res <- smote_and_undersample(d$data, d$label, fp = 2, ratio = 3);