smote: SMOTE oversampling
In hyperSMURF: Hyper-Ensemble Smote Undersampled Random Forests

Description Usage Arguments Details Value See Also Examples

Function to oversample by SMOTE the minority class

1	smote(data, fp = 1, k = 5)

`data`	data frame or matrix of data including only the minority class. Rows: examples; columns: features
`fp`	multiplicative factor for the SMOTE oversampling of the minority class (def=1). If fp<1 no oversampling is performed.
`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 returned in addition to the original set of positives. If fp<1 no new data are generated and the original data set is returned

a data frame including the original minority class examples plus the SMOTE oversampled data

smote_and_undersample

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(d$data[d$label==1,], fp = 2, k = 3);