rpart
rpart copied to clipboard
bethatkinson
Recursive Partitioning and Regression Trees
[How to interpret xerror of rpart (method="class") when losses are assigned](https://stackoverflow.com/questions/37736868/how-to-interpret-xerror-of-rpart-method-class-when-losses-are-assigned)  Strangely initial xerror is extremely small and it is 0.1000 (I see this is 1/loss of class 0),...
why it is the predicted probability of survival, not die? "Plotting rpart trees with the rpart.plot package" p3 
https://cran.r-project.org/web/packages/rpart/vignettes/longintro.pdf  What's the meaning of “54”？ What's the meaning of “0.6301370 0.3698630”？I guess it is the proportion of each class，but how to determine the corresponding class？
https://cran.r-project.org/web/packages/rpart/vignettes/longintro.pdf p20 
https://cran.r-project.org/web/packages/rpart/vignettes/longintro.pdf p18  How to adjust the probabilities for missing data?
Cost for mis-classification can be used as below: fit
https://cran.r-project.org/web/packages/rpart/vignettes/longintro.pdf p20：  it is said in p22：  I know the default impurity index is gini，but I can not get to know exactly how to calculate it
pages 11/12 of the rpart vignette covers this topic. https://cran.r-project.org/web/packages/rpart/vignettes/longintro.pdf ________________________________ From: lg21c ***@***.***> Sent: Saturday, February 17, 2024 8:52 AM To: bethatkinson/rpart ***@***.***> Cc: Subscribed ***@***.***> Subject: [EXTERNAL] [bethatkinson/rpart]...
pages 34 of the rpart vignette covers this topic. https://cran.r-project.org/web/packages/rpart/vignettes/longintro.pdf The last split decreases the cross-validated error. 