cluster.assign.torus {ClusTorus}R Documentation

Clustering by connected components of ellipsoids

Description

cluster.assign.torus returns clustering assignment for data given icp.torus objects, which can be constructed with icp.torus.score.

Usage

cluster.assign.torus(
  data,
  icp.torus,
  level = 0.1,
  intersection.plot = TRUE,
  coord = NULL
)

Arguments

data

n x d matrix of toroidal data on [0, 2π)^d.

icp.torus

an object containing all values to compute the conformity score, which will be constructed with icp.torus.score.

level

a scalar in [0,1]. Default value is 0.1.

intersection.plot

boolean index. If TRUE, then plot the intersections of given ellipsoids. Default is TRUE.

coord

a 2-vector for prespecifing the coordinates. Default value is NULL and automatically generates all combinations of coordinates.

Value

clustering assignment for data, given icp.torus objects

References

S. Jung, K. Park, and B. Kim (2021), "Clustering on the torus by conformal prediction",

I. Gilitschenski and U. D. Hanebeck, "A robust computational test for overlap of two arbitrary-dimensional ellipsoids in fault-detection of Kalman filters"

See Also

icp.torus.score

Examples

data <- toydata1[, 1:2]
icp.torus <- icp.torus.score(data, method = "kmeans",
                             kmeansfitmethod = "general",
                             param = list(J = 4, concentration = 25))
level <- 0.1

cluster.assign.torus(data, icp.torus, level)

[Package ClusTorus version 0.1.3 Index]