library(MASS)
library(randomForest)
library(caret)
set.seed(10)
## ½ÇÇè¿ë µ¥À̾î Áغñ
test = rbind(iris[21:65,],iris[91:110,])
training = rbind(rbind(iris[1:20,],iris[66:90,]),iris[111:150,])
## ÇнÀ
rf.fit = randomForest(Species ~ ., data= training, mtry = floor(sqrt(ncol(iris))), ntree = 500, importance = T)
## supervisory·Î ÇÏ·Á¸é
rf.fit = randomForest(Species ~ ., data= iris, mtry = max(1,floor(ncol(iris)/3)), ntree = 500, importance = T)
rf.fit
## ¼º´É Æò°¡
y_pred = predict(rf.fit, test)
y_pred
confusionMatrix(y_pred, test$Species)