forestRK (version 0.0-5)

mds.plot.forestRK: Makes 2D MDS (multidimensional scaling) ggplot of the test observations based on the predictions from a forestRK model.

Description

Plots 2D MDS (Multi-Dimensional Scaling) ggplot of the test observations based on the provided forestRK model, and each test observation is colour coded by their predicted class types.

The plot also has legends that tells user which colour pertains to which predicted class type.

The existing R functions dist and cmdscale were used in this function to compute the Multi-Dimensional Scales of the test data.

Usage

mds.plot.forestRK(pred.forestRK.object = pred.forestRK(),
 plot.title ="MDS Plot of Test Data Colour Coded by Forest RK Model Predictions",
 xlab ="First Coordinate", ylab = "Second Coordinate",
 colour.lab = "Predictions By The Random Forest RK Model")

Arguments

pred.forestRK.object

a pred.forestRK() object.

plot.title

an user specified title for the mds plot; the default is "MDS Plot of Test Data Colour Coded by Forest RK Model Predictions".

xlab

label for the x-axis of the plot; the default is "First Coordinate".

ylab

label for the y-axis of the plot; the default is "Second Coordinate".

colour.lab

label title for the legend that specifies categories for each colour; the default is "Predictions By The Random Forest RK Model".

Value

A multidimensional scaling ggplot (2D) of the test observations, colour coded by their predicted class types.

See Also

forestRK

Examples

# NOT RUN {
  ## example: iris dataset
  ## load the forestRK package
  library(forestRK)

  x.train <- x.organizer(iris[,1:4], encoding = "num")[c(1:25,51:75,101:125),]
  x.test <- x.organizer(iris[,1:4], encoding = "num")[c(26:50,76:100,126:150),]
  y.train <- y.organizer(iris[c(1:25,51:75,101:125),5])$y.new
  y.factor.levels <- y.organizer(iris[c(1:25,51:75,101:125),5])$y.factor.levels

  # min.num.obs.end.node.tree is set to 5 by default;
  # entropy is set to TRUE by default
  # typically the nbags and samp.size has to be much larger than 30 and 50
  pred.forest.rk <- pred.forestRK(x.test = x.test,
                                  x.training = x.train, y.training = y.train,
                                  nbags = 30, samp.size = 50,
                                  y.factor.levels = y.factor.levels)

  # generate a classical mds plot of test observations
  # and colour code them by the predicted class
  mds.plot.forestRK(pred.forest.rk)
# }