如何在R中绘制基于规则的决策树

Question

I am just starting with R and I am having a difficulty in plotting a rule based decision tree in R.

The problem is, I already have an output from hadoop MapReduce and it is in the form of a simple text file. Now I want to use this output from Hadoop and represent it graphically on R. The output file looks something like this.

1 overcast yes
1 rain 3 strong no
1 rain 3 weak yes
1 sunny 2 high no
1 sunny 2 normal yes

Is there a way that I can represent this graphically in R in something like,

http://web.cs.swarthmore.edu/~meeden/cs63/f05/figure3.1.jpg

Any help will be much appreciated. Thanks

Answer 1

Take a look at package 'rpart' . It's a package for recursive partitioning and decision trees. The following is obtained directly from the example from the help file ?rpart . The function expand.grid may be of some use to you here as well.

> example(rpart)

fit <- rpart(Kyphosis ~ Age + Number + Start, data = kyphosis)

fit2 <- rpart(Kyphosis ~ Age + Number + Start, data = kyphosis,
              parms = list(prior = c(.65,.35), split = "information"))

fit3 <- rpart(Kyphosis ~ Age + Number + Start, data = kyphosis,
              control = rpart.control(cp = 0.05))

par(mfrow = c(1,2), xpd = NA) # otherwise on some devices the text is clipped

plot(fit)
text(fit, use.n = TRUE)
plot(fit2)
text(fit2, use.n = TRUE)

Answer 2

First you have to convert the text file into a data frame. This might be a good start: Converting (web site) text file into data frame in R

Then you could use 'rpart' to build a tree. In addition to ' rpart' and it's prp() -function, you could also use the 'fancyRpartPlot' in the "rattle" -package to build fancier trees. Here is a good example.