R語言樸素貝葉斯分類上機指導(dǎo)

樸素貝葉斯分類上機指導(dǎo)、實驗?zāi)康模?.掌握矩陣數(shù)據(jù)的輸入，輸出，以及矩陣和數(shù)據(jù)框數(shù)據(jù)的轉(zhuǎn)換，認識一下list型數(shù)據(jù)。2.理解并掌握樸素貝葉斯分類原理。3?會使用klaR包中的NaiveBayes函數(shù)實現(xiàn)貝葉斯分類算法。二、 實驗內(nèi)容：本實驗利用樸素貝葉斯分類方法對課本 P144頁playtennis數(shù)據(jù)集建立模型并預(yù)測。三、 實驗步驟：1?將課本P144頁playtennis數(shù)據(jù)以矩陣形式輸入，注意理解下面紅色代碼。data<-matrix(c("sunny","hot","high","weak","no","sunny","hot","high","strong","no","overcast","hot","high","weak","yes","rain","mild","high","weak","yes","rain","cool","normal","weak","yes","rain","cool","normal","strong","no","overcast","cool","normal","strong","yes","sunny","mild","high","weak","no","sunny","cool","normal","weak","yes","rain","mild","normal","weak","yes","sunny","mild","normal","strong","yes","overcast","mild","high","strong","yes","overcast","hot","normal","weak","yes","rain","mild","high","strong","no"),byrow=TRUE,dimnames=list(day=c(),condition=c("outlook","temperature","humidity","wind","playtennis")),nrow=14,ncol=5);#上網(wǎng)查詢dimnames的用法#輸出一下data數(shù)據(jù)：#將矩陣轉(zhuǎn)化成數(shù)據(jù)框>data1<-data.frame(data)>data1outlooktemperaturehumidity windplaytennis1sunnyhothighweakno2sunnyhothighstrongno3overcasthothighweakyes4rainmildhighweakyes5raincoolnormalweakyes6raincoolnormalstrongno7overcastcoolnormalstrongyes8sunnymildhighweakno9sunnycoolnormalweakyes10rainmildnormalweakyes11sunnymildnormalstrongyes12overcastmildhighstrongyes13overcasthotnormalweakyes14rainmildhighstrongno#將手動輸入的數(shù)據(jù)保存成txt文件>write.table(data1,file="playtennis.txt",sep="")#可以重新讀入保存的txt文件data2<-read.table("playtennis.txt",head=TRUE)2.理解并掌握樸素貝葉斯分類原理,讀懂下列代碼#算出去玩與不玩的先驗概率prior.yes<-sum(data2[,5]=="yes")/length(data2[,5]);}}prior.no<-sum(data2[,5]=="no")/length(data2[,5]);#建立樸素貝葉斯分類函數(shù)bayespre<-function(condition){post.yes<-sum((data2[,1]==condition[1])&(data2[,5]=="yes"))/sum(data2[,5]=="yes")*sum((data2[,2]==condition[2])&(data2[,5]=="yes"))/sum(data2[,5]=="yes")*sum((data2[,3]==condition[3])&(data2[,5]=="yes"))/sum(data2[,5]=="yes")*sum((data2[,4]==condition[4])&(data2[,5]=="yes"))/sum(data2[,5]=="yes")*prior.yes;post.no<-sum((data2[,1]==condition[1])&(data2[,5]=="no"))/sum(data2[,5]=="no")*sum((data2[,2]==condition[2])&(data2[,5]=="no"))/sum(data2[,5]=="no")*sum((data2[,3]==condition[3])&(data2[,5]=="no"))/sum(data2[,5]=="no")*sum((data2[,4]==condition[4])&(data2[,5]=="no"))/sum(data2[,5]=="no")*prior.no;return(list(prob.yes=post.yes,prob.no=post.no,prediction=ifelse(post.yes>=post.no,"yes","no")));#利用建立的樸素貝葉斯函數(shù)做預(yù)測bayespre(c("rain","hot","high","strong"))$prob.yes[1]0.005291005$prob.no[1]0.02742857$prediction[1]"no"#這些結(jié)果是多少？用函數(shù)bayespre()算一下，你能手動算出來嗎？把你的算式寫出來bayespre(c("sunny","mild","normal","weak"))bayespre(c("overcast","mild","normal","weak"))bayespre(c("sunny","cool","high","strong"))3使用klaR包中的NaiveBayes函數(shù)實現(xiàn)貝葉斯分類算法NaiveBayes()函數(shù)的語法和參數(shù)如下：NaiveBayes(formula,data, …,subset,na.action=na.pass)NaiveBayes(x,grouping,prior,usekernel=FALSE,fL=0, ?…)formula指定參與模型計算的變量，以公式形式給出，類似于 y=x1+x2+x3；data用于指定需要分析的數(shù)據(jù)對象；na.action指定缺失值的處理方法，默認情況下不將缺失值納入模型計算，也不會發(fā)生報錯信息，當設(shè)為“na.omit”時則會刪除含有缺失值的樣本；x指定需要處理的數(shù)據(jù)，可以是數(shù)據(jù)框形式，也可以是矩陣形式；grouping為每個觀測樣本指定所屬類別；prior可為各個類別指定先驗概率，默認情況下用各個類別的樣本比例作為先驗概率；usekernel指定密度估計的方法(在無法判斷數(shù)據(jù)的分布時，采用密度密度估計方法) ，默認情況下使用標準的密度估計，設(shè)為 TRUE時，則使用核密度估計方法；fL指定是否進行拉普拉斯修正， 默認情況下不對數(shù)據(jù)進行修正，當數(shù)據(jù)量較小時，可以設(shè)置該參數(shù)為1，即進行拉普拉斯修正。#通過抽樣建立訓(xùn)練樣本和測試樣本index<-sample(2,size=nrow(iris),replace=TRUE,prob=c(0.75,0.25))train<-iris[index==1,]test<-iris[index==2,]#加載R包并使用樸素貝葉斯算法library(MASS)library(klaR)#因子化train$Species<-as.factor(train$Species)res2<-NaiveBayes(Species~.,data=train)pre<-predict(res2,newdata=test[,1:4])#生成實際與預(yù)判交叉表和預(yù)判精度table(test$Species,pre$class)23110000920011sum(diag(table(test$Species,pre$class)))/sum(table(test$Species,pre$class))0.9375讀懂上面的例子。仿照上面的例子使用klaR包中的NaiveBayes函數(shù)建立playtennis?txt貝葉斯分類模型，并預(yù)測上面的例子。c("rain","hot","high","strong")c("sunny","mild","normal","weak")c("overcast","mild","normal","weak")c("sunny","cool","high","strong")#將上面的情況建立dataframe作為測試題test<-data.frame()test<-rbind(c("rain","