Introduction to Machine Learning with R by Scott V. Burger

Author:Scott V. Burger
Language: eng
Format: epub, mobi, pdf
Publisher: O'Reilly Media, Inc.
Published: 2018-03-26T04:00:00+00:00

multi <- data.frame(x1 = c(0.03, 0.24, 0.21, 0, 0, 0.23, 0.6, 0.64, 0.86, 0.77), x2 = c(0.07, 0.06, 0.19, 1.15, 0.95, 1, 0.81, 0.64, 0.44, 0.74), lab = c(1, 1, 1, 2, 2, 2, 3, 3, 3, 3)) plot(x2 ~ x1, pch = lab, cex = 2, data = multi, main = "Multi-Class Classification", xlab = "x", ylab = "y")

Figure 4-13. Multiclass data can be separated only by using a one-versus-many approach

There are three distinct classes of data, and you want to find some kind of lines that split them into their own categories, much like you did for the binary case. What this essentially boils down to is our simple binary test, but you change which group you’re comparing against. This is called a “one-versus-all” or “one-versus-many” test, in which you test three cases—triangles-versus-rest, circles-versus-rest, and crosses-versus-rest, as depicted in Figure 4-14:

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.