运行并用分类数据绘制随机森林作为 r 中的特征向量

Question

我有如下数据。它有三列。第一列是分类数据，第二列是数字，最后一列是我的 class 标签。我想运行我的数据上的随机森林并绘制树以及变量重要性。我的目标是找到哪个 subject_result 是最重要的，其次是什么，还可以看到树。

使用此代码有错误

library(randomForest)  
randomForest(ENSC_Disc~.,data = df)

Error in randomForest.default(m, y, ...) : 
  NA/NaN/Inf in foreign function call (arg 1)

同样使用rpart和ctree return错误。

data.frame(stringsAsFactors=FALSE,
   subject_result = c("ENSCPassed", "CHEMPassed", "ENSCPassed", "OTHERPassed",
                      "ENSCPassed", "MATHPassed", "ENSCPassed", "OTHERPassed",
                      "OTHERPassed", "OTHERPassed", "PHYSPassed", "CHEMPassed",
                      "MATHPassed", "ENSCPassed", "CMPTPassed", "OTHERPassed",
                      "CMPTPassed"),
     semester_num = c(9L, 4L, 16L, 7L, 7L, 2L, 8L, 11L, 4L, 12L, 1L, 4L, 3L,
                      11L, 8L, 11L, 12L),
        ENSC_Disc = c(1, 1, 1, 0, 1, 0, 1, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0)
)

Answer 1

这是使用 caret.

的尝试

library(tidyverse)
library(caret)
df<-data.frame(stringsAsFactors=FALSE,
           subject_result = c("ENSCPassed", "CHEMPassed", "ENSCPassed", "OTHERPassed",
                              "ENSCPassed", "MATHPassed", "ENSCPassed", "OTHERPassed",
                              "OTHERPassed", "OTHERPassed", "PHYSPassed", "CHEMPassed",
                              "MATHPassed", "ENSCPassed", "CMPTPassed", "OTHERPassed",
                              "CMPTPassed"),
           semester_num = c(9L, 4L, 16L, 7L, 7L, 2L, 8L, 11L, 4L, 12L, 1L, 4L, 3L,
                            11L, 8L, 11L, 12L),
           ENSC_Disc = c(1, 1, 1, 0, 1, 0, 1, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0)
)
set.seed(233)
str(df)
df$ENSC_Disc<-as.factor(df$ENSC_Disc)
fit.rf<-train(ENSC_Disc~.,data=df,metric="Accuracy",method="rf",
              trControl=trainControl(method="cv",number=5))

第一个变量重要性：

plot(varImp(fit.rf))

实际的树：这并不像我想的那样有效。更好的方法是使用 library(rattle) 但这只适用于 "rpart" 而不是 "rf" 但这里是：

plot(fit.rf$finalModel)

运行并用分类数据绘制随机森林作为 r 中的特征向量

Running and plotting Random Forest with categorical data as a feature vector in r

plot

r

random-forest

运行 并用分类数据绘制随机森林作为 r 中的特征向量

Running and plotting Random Forest with categorical data as a feature vector in r

plot

r

random-forest

运行并用分类数据绘制随机森林作为 r 中的特征向量