R语言实现随机森林算法

library(xlsx)
data<-read.xlsx("E:\\数据分析\\data.xlsx",1)
#将数据集分为训练集和测试集,比例为7:3
train_sub = sample(nrow(data),7/10*nrow(data))
train_data = data[train_sub,]
test_data = data[-train_sub,]
library(pROC) #绘制ROC曲线
library(randomForest)
#数据预处理
train_data$buy= as.factor(train_data$buy)
test_data$buy= as.factor(test_data$buy)
wine_randomforest <- randomForest(buy ~  behavior+reason+money,
                                          data = train_data,
                                          ntree =500,
                                          mtry=3,
                                          importance=TRUE ,
                                          proximity=TRUE)
#查看变量的重要性
wine_randomforest$importance
varImpPlot(wine_randomforest, main = "variable importance")
#对测试集进行预测
pre_ran <- predict(wine_randomforest,newdata=test_data)
#将真实值和预测值整合到一起
obs_p_ran = data.frame(prob=pre_ran,obs=test_data$buy)
#输出混淆矩阵
table(test_data$buy,pre_ran,dnn=c("真实值","预测值"))
#绘制ROC曲线
ran_roc <- roc(test_data$ buy,as.numeric(pre_ran))
plot(ran_roc, print.auc=TRUE, auc.polygon=TRUE, grid=c(0.1, 0.2),grid.col=c("green", "red"), max.auc.polygon=TRUE,auc.polygon.col="skyblue", print.thres=TRUE,main='随机森林模型ROC曲线,mtry=3,ntree=500')
MDSplot(wine_randomforest,data$buy,k=2,palette=NULL,pch=20)
plot(wine_randomforest)
treesize(wine_randomforest,terminal = TRUE)

wine_randomforest2<- randomForest(buy ~  behavior+reason+money,
                                          data = train_data,
                                          ntree =200,
                                          mtry=3,
                                          importance=TRUE ,
                                          proximity=TRUE)
#改变树的个数再次进行拟合
#查看变量的重要性
wine_randomforest2$importance
varImpPlot(wine_randomforest2, main = "variable importance")
#对测试集进行预测
pre_ran <- predict(wine_randomforest2,newdata=test_data)
#将真实值和预测值整合到一起
obs_p_ran = data.frame(prob=pre_ran,obs=test_data$buy)
#输出混淆矩阵
table(test_data$buy,pre_ran,dnn=c("真实值","预测值"))
#绘制ROC曲线
ran_roc <- roc(test_data$ buy,as.numeric(pre_ran))
plot(ran_roc, print.auc=TRUE, auc.polygon=TRUE, grid=c(0.1, 0.2),grid.col=c("green", "red"), max.auc.polygon=TRUE,auc.polygon.col="skyblue", print.thres=TRUE,main='随机森林模型ROC曲线,mtry=3,ntree=200')
MDSplot(wine_randomforest2,data$buy,k=2,palette=NULL,pch=20)
plot(wine_randomforest2)
treesize(wine_randomforest2,terminal = TRUE)

数据集获取地址

链接：https://pan.baidu.com/s/1jDE37PoiDh8L5wfyekFFkw 
提取码：dr25