knnor: K-Nearest Neighbor OveRsampling approach (KNNOR)

In aifeducation: Artificial Intelligence for Education

K-Nearest Neighbor OveRsampling approach (KNNOR)

Description

K-Nearest Neighbor OveRsampling approach (KNNOR)

Usage

knnor(dataset, k, aug_num, cycles_number_limit = 100L)

Arguments

dataset	list containing the following fields: embeddings: an 2-D array (matrix) with size batch x times*features labels: an 1-D array (vector) of integers with batch elements
k	⁠unsigned integer⁠ number of nearest neighbors
aug_num	⁠unsigned integer⁠ number of datapoints to be augmented
cycles_number_limit	⁠unsigned integer⁠ number of maximum try cycles

Value

Returns artificial points (⁠2-D array (matrix) with size ⁠aug_numxtimes*features')

References

Islam, A., Belhaouari, S. B., Rehman, A. U. & Bensmail, H. (2022). KNNOR: An oversampling technique for imbalanced datasets. Applied Soft Computing, 115, 108288. https://doi.org/10.1016/j.asoc.2021.108288

aifeducation documentation built on Nov. 19, 2025, 5:08 p.m.