Question

使用R中的虹膜数据集，我试图将NaïveBayes分类器拟合到虹膜训练数据中，这样我就可以为朴素贝叶斯分类器生成训练数据集（预测与实际）的混淆矩阵，什么是NaïveBayes分类器的错误分类率？

到目前为止，这是我的代码：

 iris$spl=sample.split(iris,SplitRatio=0.8)
 train=subset(iris, iris$spl==TRUE)
 test=subset(iris, iris$spl==FALSE)

 iris.nb <- naiveBayes(Species~.,data = train)
 iris.nb

 nb_test_predict <- predict(iris.nb, train)

有关如何解决此问题的任何建议吗？

Answer 1

包caret包括confusionMatrix函数，该函数返回非常完整的输出。

library(e1071)
library(caTools)
library(caret)

iris$spl = sample.split(iris, SplitRatio = 0.8)
train <- subset(iris, iris$spl == TRUE)
test <- subset(iris, iris$spl == FALSE)

iris.nb <- naiveBayes(Species ~ ., data = train)

nb_train_predict <- predict(iris.nb, test[ , names(test) != "Species"])

cfm <- confusionMatrix(nb_train_predict, test$Species)
cfm

# Confusion Matrix and Statistics
# 
# Reference
# Prediction   setosa versicolor virginica
# setosa         17          0         0
# versicolor      0         14         1
# virginica       0          2        16
# 
# Overall Statistics
# 
# Accuracy : 0.94            
# 95% CI : (0.8345, 0.9875)
# No Information Rate : 0.34            
# P-Value [Acc > NIR] : < 2.2e-16       
# 
# Kappa : 0.9099          
# Mcnemar's Test P-Value : NA              
# 
# Statistics by Class:
# 
#                      Class: setosa Class: versicolor Class: virginica
# Sensitivity                   1.00            0.8750           0.9412
# Specificity                   1.00            0.9706           0.9394
# Pos Pred Value                1.00            0.9333           0.8889
# Neg Pred Value                1.00            0.9429           0.9688
# Prevalence                    0.34            0.3200           0.3400
# Detection Rate                0.34            0.2800           0.3200
# Detection Prevalence          0.34            0.3000           0.3600
# Balanced Accuracy             1.00            0.9228           0.9403

将混淆矩阵显示为ggplot graphic：

library(ggplot2)
library(scales)

ggplotConfusionMatrix <- function(m){
  mytitle <- paste("Accuracy", percent_format()(m$overall[1]),
                   "Kappa", percent_format()(m$overall[2]))
  p <-
    ggplot(data = as.data.frame(m$table) ,
           aes(x = Reference, y = Prediction)) +
    geom_tile(aes(fill = log(Freq)), colour = "white") +
    scale_fill_gradient(low = "white", high = "steelblue") +
    geom_text(aes(x = Reference, y = Prediction, label = Freq)) +
    theme(legend.position = "none") +
    ggtitle(mytitle)
  return(p)
}

ggplotConfusionMatrix(cfm)

Answer 2

您似乎在寻找table功能

library(e1071)
library(caTools)
spl=sample.split(iris$Species,SplitRatio=0.8)
train=subset(iris, spl==TRUE)
test=subset(iris, spl==FALSE)

iris.nb <- naiveBayes(Species~.,data = train)
iris.nb

nb_train_predict <- predict(iris.nb, train[,!names(train) %in% "Species"])
table(nb_train_predict,train$Species)

输出

nb_train_predict setosa versicolor virginica
      setosa         40          0         0
      versicolor      0         38         4
      virginica       0          2        36

误分类率可以计算为

mean(nb_train_predict != train$Species)

如何产生混淆矩阵并找出NaïveBayes分类器的错误分类率？

2 个答案: