sample_generator: The function to generate 2-dimensional dataset

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

The function to generate 2-dimensional dataset

Description

The function to generate 2-dimensional dataset given the number of instances and the ratio between the number of negative instances to total instances. The positive instances will be distributed uniformly as the circle in the center while negative instances are around over the domain. The random positive outcasts are also generated. The dataset is used to show the difference between datasets generated by each sampling technique.

Usage

sample_generator(n, ratio = 0.8, xlim = c(0, 1), ylim = c(0, 1),
   radius = 0.25, overlap = -0.05, outcast_ratio = 0.01)

Arguments

n	The number of instances in the dataset
ratio	The ratio of negative instances to the total number of instances
xlim	The range of values in the first dimension
ylim	The range of values in the second dimension
radius	The radius of the circle of positive instances
overlap	The gap between the set of positive and negative instances
outcast_ratio	The ratio of outcast to be generate in this dataset.

Value

A 2-dimensional dataset with the 3rd column as its target class vector.

Author(s)

Wacharasak Siriseriwan <[e-mail]>

Examples

	data_example = sample_generator(5000,ratio = 0.80)
	plot(data_example[data_example[,3]=="n",1],
	data_example[data_example[,3]=="n",2],col="yellow")
	points(data_example[data_example[,3]=="p",1],
	data_example[data_example[,3]=="p",2],col="red",pch=14)

smotefamily documentation built on May 29, 2024, 7:54 a.m.