How should I plot a multi-split (non binary split) decision tree run from RWeka J48() function?

Question

I try to build a multi-split tree with the option of J48() function:

mytree = J48(class ~ ., data=train.set, na.action = NULL, control= Weka_control(U=TRUE, B=FALSE))

and it seems to work fine. But I cannot plot the decision tree by any of the following codes:

if(require("partykit", quietly = TRUE)) plot(mytree)

by using the plot tool in the partykit package. The manual says it only plots the binary-split tree. Or the code:

prp(mytree)

by using the plot function in the rpart.plot package. Because R shows the warning message

Error in prp(mytree) : Not an rpart object

Can anyone tell me how to plot this tree?

Since you do not provide your data, it is very difficult to help you. I was able to construct a non-binary tree using J48 and plot it. Please type dput(mytree) and paste the results into your question so that we can try to figure out why your tree is causing problems. — G5W

d.lee d.lee · Accepted Answer · 2018-08-27T04:00:24

The data I use is the heart disease data from UCI repository, and it looks like > head(Hungarian) age sex cp trestbps chol fbs restecg thalach exang oldpeak slope ca thal num 1 28 1 2 130 132 0 2 185 0 0 <NA> NA <NA> 0 2 29 1 2 120 243 0 0 160 0 0 <NA> NA <NA> 0 3 29 1 2 140 NA 0 0 170 0 0 <NA> NA <NA> 0 4 30 0 1 170 237 0 1 170 0 0 <NA> NA 6 0 5 31 0 2 100 219 0 1 150 0 0 <NA> NA <NA> 0 6 32 0 2 105 198 0 0 165 0 0 <NA> NA <NA> 0 The class variable is the variable num (0 or 1). And the result of typing dput(mytree) is

structure(list(classifier = new("jobjRef", jobj = <pointer: 0x0de0e848>, 
jclass = "java/lang/Object"), predictions = structure(c(1L, 
1L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 1L, 2L, 1L, 1L, 2L, 
1L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 2L, 1L, 1L, 1L, 2L, 1L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 2L, 1L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 1L, 2L, 2L, 
2L, 1L, 1L, 2L, 1L, 1L, 2L, 1L, 1L, 1L, 1L, 2L, 1L, 1L, 1L, 1L, 
1L, 1L, 2L, 1L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 2L, 1L, 2L, 1L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 2L, 1L, 1L, 
1L, 1L, 1L, 2L, 1L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 2L, 1L, 1L, 2L, 
1L, 1L, 1L, 1L, 2L, 2L, 2L, 1L, 1L, 1L, 2L, 1L, 1L, 2L, 1L, 1L, 
1L, 1L, 2L, 2L, 1L, 1L, 1L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 2L, 2L, 1L, 1L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 
1L, 1L, 1L, 2L, 2L, 1L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
2L, 1L, 1L, 1L, 2L, 2L, 1L, 1L, 1L, 2L, 1L, 2L), .Label = c("0", 
"1"), class = "factor"), call = J48(formula = num ~ ., data = train.set, 
    na.action = NULL, control = Weka_control(U = TRUE, B = FALSE)), 
    handlers = list(data = function (mf) 
    {
        terms <- attr(mf, "terms")
        if (any(attr(terms, "order") > 1L)) 
            stop("Interactions are not allowed.")
        factors <- attr(terms, "factors")
        varnms <- rownames(factors)[c(TRUE, rowSums(factors)[-1L] > 
            0)]
        mf[, sub("^`(.*)`$", "\\1", varnms), drop = FALSE]
    }), levels = c("0", "1"), terms = num ~ age + sex + cp + 
        trestbps + chol + fbs + restecg + thalach + exang + oldpeak + 
        slope + ca + thal), class = c("J48", "Weka_tree", "Weka_classifier"
))

Thanks a lot.

How should I plot a multi-split (non binary split) decision tree run from RWeka J48() function?

2 Answers