R语言——一秒决策树分析 R数据分析

决策树分析 1.文档读取

data <- read.csv(file.choose(),header=TRUE) View(data)

2.建立训练数据与测试数据

set.seed(12345) select <- sample(1:nrow(data),nrow(data)*0.7) train <- data[select,] test <- data[-select,]

【R语言——一秒决策树分析】3、建立模型

library(rpart) library(rpart.plot)CART.tree <- rpart(Class ~ ., data=https://www.it610.com/article/train, control=rpart.control(minsplit=2, cp=0)) rpart.plot(CART.tree)

3、模型检验及评价

CART.Prediction <- predict(CART.tree, newdata=https://www.it610.com/article/test, type='class')Results <- table(Prediction=CART.Prediction, Actual=test$Class) ResultsCorrect_Rate <- sum(diag(Results)) / sum(Results) Correct_Rate

4、查找CP并重组检测模型

CART.tree <- prune(CART.tree, cp=0.03) rpart.plot(CART.tree)CART.Prediction <- predict(CART.tree, newdata=https://www.it610.com/article/test, type='class')results <- table(Prediction=CART.Prediction, Actual=test$Class) resultsCorrect_Rate <- sum(diag(results)) / sum(results) Correct_Rate