Cassini Data
A Cassini data set with 1000 points in 2-dimensional space which are drawn from the uniform distribution on 3 structures. The two outer structures are banana-shaped; the “middle” structure in between them is a circle.
data("Cassini")A classed list with components
xa matrix with 1000 rows and 2 columns giving the coordinates of the points.
classesa factor indicating which structure the respective points belong to.
Instances of Cassini data sets can be created using function
mlbench.cassini in package mlbench.
The data set at hand was obtained using
library("mlbench")
set.seed(1234)
Cassini <- mlbench.cassini(1000)data("Cassini")
op <- par(mfcol = c(1, 2))
## Plot the data set:
plot(Cassini$x, col = as.integer(Cassini$classes),
xlab = "", ylab = "")
## Create a "random" k-means partition of the data:
set.seed(1234)
party <- kmeans(Cassini$x, 3)
## And plot that.
plot(Cassini$x, col = cl_class_ids(party),
xlab = "", ylab = "")
## (We can see the problem ...)
par(op)Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.