similar to: Multiple regression trees

I'm looking for an analogy to SPSS and "variable lable" and "value lable" in R. I have a copy of S+4.0 and can't find any info in their docs, and don't find it in R-intro either. In SPSS, for each variable, there is a name like VAR001 and a variable lable (a longer descriptive string) like Respondent ID and for many of the values of other variables there are

Hi, I am using rpart decision trees to analyze customer churn. I am finding that the decision trees created are not effective because they are not able to recognize factors that influence churn. I have created an example situation below. What do I need to do to for rpart to build a tree with the variable experience? My guess is that this would happen if rpart used the loss matrix while creating

I have a class with 732 members, so using rpart.plot is giving me a tiny plot in the middle of the window. Is there a good way to modify the plot, or replace the long list with something like "group1"? -- View this message in context: http://r.789695.n4.nabble.com/Plotting-rpart-trees-with-long-list-of-class-members-tp4635671.html Sent from the R help mailing list archive at

Jonathan, Try making a list instead of an array. See ?list. Also, did you look into random forests? I'm not sure what you want to do, but there might be methods there to do some of the work for you. Sean On 3/19/04 1:12 PM, "Jonathan Williams" <jonathan.williams at pharmacology.oxford.ac.uk> wrote: > I would like to collect the trees grown by rpart fits in an array,

Hi, I am somewhat new to R so this question may be foolish, but is it possible to add decision trees into a list, array or vector in R? I am trying to build a collection (ensemble) of decision trees. Every time a new instance arrive I need to get the prediction of each decision tree. I have tried to add a decision tree into a variable but without luck. Is a special package needed perhaps? This

Hi! I'm trying to construct and plot a decision tree to class a set of only 8 objects and tried to use the rpart and tree function, but get a error message both times: rpart: fit is not a tree, just a root tree: cannot plot singlenode tree I read in the post 'question regression trees' that rpart doesn't split a set of fewer than 20 objects...so I guess the same holds true for

I would like to know if someone has done work on trees with multivariate response. I need something like rpart but for vector responses. If someone has code that he/she is willing to share, I would be grateful. If not, even guidelines for writing my own starting from rpart would be welcomed. ft. -- Fernando TUSELL e-mail: Departamento de

Hi, I am working on a regression tree in Rpart that uses a continuous response variable that is ordered. I read a previous response by Pfr. Ripley to a inquiry regarding the ability of rpart to handle ordinal responses in 2003. At that time rpart was unable to implement an algorithm to handle ordinal responses. Has there been any effort to rectify this in recent years? Thanks! Stacey On

Hello Everyone, I'm doing a Classification trees with categorical explanatory variables using library rpart and I would like to do a prediction for some data imputs. I don't know where's a function or how can I do it?. Is there someone can help ?? ¿. Here's the code that I'm using. library(rpart) fit <- rpart(Kyphosis ~ Age + Number + Start, data=kyphosis) plot(fit)

Hi, I have done an analysis using 'rpart' to construct a Classification Tree. I am wanting to retain the output in tree form so that it is easily interpretable. However, I am wanting to compare the 'accuracy' of the tree to a Random Forest to estimate how much predictive ability is lost by using one simple tree. My understanding is that the error automatically displayed by the two

Hi, I am using the packages tree and rpart to build a classification tree to predict a 0/1 outcome. The package rpart has the advantage that the function plotcp gives a visual representation of the cross-validation results with a horizontal line indicating the 1 standard error rule, i.e. the recommendation to select the most parsimonious model (the smallest tree) whose error is not more than one

I am familiar with the rpart and tree packages for classification and regression trees. However, quite a bit of the research in the transportation community relating to decision trees uses the C4.5 family of algorithms by Quinlan. Are there any plans to make a C4.5 (or a derivative of it) available to R? If not, then I might use the WEKA Java package ( http://www.cs.waikato.ac.nz/ml/weka) that

Hi, I am building a regression tree (method=anova) by using rpart package and as a final result I get the final leaves characterized by different means and standard deviations for the dependent variable. However, differently from the classification tree for categorical variables I cannot find a way to assign each observation to a leaf, i.e. I can find no frame whcih contains the observation id

Hello, This is my first time addressing such a big audience so apologies in advance in case I fail to formulate this question. I am working with 13 species of trees, and the data I have are: 1 continuous (phenolic concentration in xylem and in phloem) and 2 categorical variables: lineage (3 subclades) and habitat (fire and non fire). I am trying to see how species can be splitted

Hi netters, I want to learn a decision tree from a series of instances (learning data). The packages tree or rpart can do this quite well, but the scoring functions (splitting criteria) are fixed in these packages, like gini or something. However, I'm going to use another scoring function. At first I wanna modify the R code of tree or rpart and put my own scoring function in. But it

Dear R, Short verse of the question: Is there R code which will calculate surrogate splits and/or delta impurity for decision trees at each node? Long Version: I have local, legacy code which I use to calculate my decision trees. I would like to switch to R, but as I understand it surrogate splits are not implemented. Surrogate splits and feature ranking are described in Breiman et al

Hi, I am trying to use CART to find an ideal cut-off value for a simple diagnostic test (ie when the test score is above x, diagnose the condition). When I put in the model fit=rpart(outcome ~ predictor1(TB144), method="class", data=data8) sometimes it gives me a tree with multiple nodes for the same predictor (see below for example of tree with 1 or multiple nodes). Is there a way

Hi dear R-help memebers, When building a CART model (specifically classification tree) using rpart, it is sometimes obvious that there are variables (X's) that are meaningful for predicting some of the outcome (y) variables - while other predictors are relevant for other outcome variables (y's only). *How can it be estimated, which explanatory variable is "used" for which of

Dear all, Though labeling the x and y axis in the plot command seems to be straight forward, I can not get it to work if I do the following: ## Creating example data edata <- c(1,2,1,2) edata <- matrix(edata, 2, 2, byrow = T) colnames(edata) <- c("a", "b") edata <- data.frame(edata) ## plot data par(mfrow<- c(1,2)) plot(edata$a, edata$b, xlab =

--- included message ---- Thus, my question is: *What common measures exists for ranking/measuring variable importance of participating variables in a CART model? And how can this be computed using R (for example, when using the rpart package)* ---end ---- Consider the following printout from rpart summary(rpart(time ~ age + ph.ecog + pat.karno, data=lung)) Node number 1: 228 observations,