2020-12-22

ibrary(ipred)
library(rpart)
library(rpart.plot)
library(RColorBrewer)
library(rattle)
library(rpart.plot)
library(RColorBrewer)
library(mvtnorm)
library(modeltools)
library(stats4)
library(strucchange)
library(zoo)
library(ipred)
library(randomForest)
#读取数据
train1<-read.csv(“train.csv”,stringsAsFactors=FALSE)
test1<-read.csv(“test.csv”,stringsAsFactors=FALSE)
train1KaTeX parse error: Expected 'EOF', got '#' at position 9: Name[1] #̲设置测试数据中Survived…Survived <- NA
#将训练数据的和测试数据整合到一起
combi <- rbind(train1,test1)
#将combi中名字转成字符串格式
combi$Name<-as.character(combi$Name)
#读取第一个名字
combiKaTeX parse error: Expected 'EOF', got '#' at position 9: Name[1] #̲对名字进行拆分 strspli…Name[1], split=’[,.]’)
#索引1
strsplit(combiKaTeX parse error: Expected 'EOF', got '#' at position 29: …t='[,.]')[[1]] #̲索引2，挖出姓氏 strspl…Name[1], split=’[,.]’)[[1]][2]
#对所有的乘客姓名进行划分，并挖出姓氏
combi$Title<-sapply(combi$Name, FUN=function(x) {strsplit(x, split=’[,.]’)[[1]][2]})
combiKaTeX parse error: Double superscript at position 16: Title <- sub(' '̲, '', combiTitle)
#列出划分后的数据
table(combiKaTeX parse error: Expected 'EOF', got '#' at position 8: Title) #̲创建新的组合 combiTitle[combi$Titlecombi$Title[combi$Titlecombi$Title[combiKaTeX parse error: Expected 'EOF', got '#' at position 69: …r')] <- 'Lady' #̲创建因子 combiTitle <- factor(combi$Title)

#建立一个家庭组合变量
combi$FamilySize<-combi$SibSp + combiKaTeX parse error: Expected 'EOF', got '#' at position 11: Parch + 1 #̲对字符串进行处理 combiSurname <- sapply(combiKaTeX parse error: Expected 'EOF', got '#' at position 59: ….]')[[1]][1]}) #̲创建新的变量 combiFamilyID <- paste(as.character(combi$FamilySize),combi$Surname, sep="")
combi$FamilyID[combi$FamilySize <= 2] <- ‘Small’
#查看家族ID中的分布个数
table(combiKaTeX parse error: Expected 'EOF', got '#' at position 11: FamilyID) #̲转化为数据框 famIDs <…FamilyID))
famIDs <- famIDs[famIDs$Freq<=2,]combi$FamilyID[combi$FamilyID$Var1] <- ‘Small’
combi$FamilyID<-factor(combi$FamilyID)

#创建决策树
Agefit <- rpart(Age ~ Pclass + Sex + SibSp + Parch + Fare + Embarked + Title + FamilySize,
data=combi[!is.na(combiKaTeX parse error: Expected 'EOF', got '#' at position 25: …ethod="anova") #̲画出决策树图形 fancyRp…Age[is.na(combi$Age)]<-predict(Agefit,combi[is.na(combi$Age),])
#创建一个新变量
combi$FamilyID2<-combi$FamilyID
combi$FamilyID2<-as.character(combi$FamilyID2)
combi$FamilyID2[combi$FamilySize <= 3] <- ‘Small’
combi$FamilyID2<-factor(combi$FamilyID2)
#将合并后的数组进行分开
train_data <- data.frame(combi[1:891,])
test_data <- data.frame(combi[892:1309,])
#利用装袋技术进行建模
BAG=bagging((Survived) ~ Pclass+Age+Sex+SibSp+Parch + Fare + Embarked + Title + FamilySize + FamilyID,data = train_data,mfinal=900000)
#对于预测数据做回归预测
Predictio <- predict(BAG, test_data)

sub<- data.frame(PassengerId = test_data$PassengerId, Survive = Predictio)

sub$Survived[sub$Survive<0.5]<-0
sub$Survived[sub$Survive>=0.5]<-1
sub1=sub[,-2]
summary(sub1)
write.csv(sub1, file = “best.csv”, row.names = FALSE