make_klink_circles: Generate Multiple Interlocked Circles in High-Dimensional Space
Description
This function generates \(k\) interlocked circular clusters in a
\(p\)-dimensional space. The circles are constructed using
gen_multicluster(), with each circle positioned in a different
coordinate plane and slightly offset so that they interlock with a
central circle (hub-like structure).
Usage
make_klink_circles(n = c(200, 100), p = 4, k = 2, offset = 0.5)
Value
A data frame (or tibble, depending on gen_multicluster())
containing the generated points and cluster assignments.
Arguments
- n
An integer vector of length \(k\) giving the number of points
in each circle. Default is c(200, 100).
- p
Integer, the dimensionality of the embedding space. Must be
at least 3. Default is 4.
- k
Integer, the number of circles to generate. Default is 2.
- offset
Numeric, the amount of positional shift applied to each
circle along the second coordinate axis to prevent complete overlap.
Default is 0.5.
Examples
Run this code# Generate two interlocked circles in 4-D
twolink_circles <- make_klink_circles()
Run the code above in your browser using DataLab