R 实现朴素贝叶斯算法

朴素贝叶斯分类算法是一种常用的分类要领，应用很是遍及，譬如垃圾邮件判定，电子商务反作弊（作弊卖家等等）。

# 数据集来自Tom Mitchell’s book “Machine
Learning”.

#界说数据矩阵matrix，matrix(vector, nrow=r, ncol=c,
byrow=logical_value, 
dimnames=list(char_vector_rownames, char_vector_colnames))

#nrow暗示行数

#ncol暗示列数

#byrow暗示矩阵组织方法，按行可能按列

#dimnames暗示行标识，列标识
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);
 

#统计yes，no呈现的概率                 

prior.yes = sum(data[,5] ==
“yes”) / length(data[,5]);
prior.no = sum(data[,5] == “no”)
/ length(data[,5]);

#输入条件向量

###################################################
naive.bayes.prediction <- function(condition.vec)
{
###################################################
  # Calculate unnormlized posterior probability
for playtennis = yes.
  playtennis.yes <-
   
sum((data[,1] == condition.vec[1]) & (data[,5] ==
“yes”)) / sum(data[,5] ==
“yes”) * # P(outlook = f_1 |
playtennis = yes)
   
sum((data[,2] == condition.vec[2]) & (data[,5] ==
“yes”)) / sum(data[,5] ==
“yes”) * # P(temperature = f_2 |
playtennis = yes)
   
sum((data[,3] == condition.vec[3]) & (data[,5] ==
“yes”)) / sum(data[,5] ==
“yes”) * # P(humidity = f_3 |
playtennis = yes)
   
sum((data[,4] == condition.vec[4]) & (data[,5] ==
“yes”)) / sum(data[,5] ==
“yes”) * # P(wind = f_4 |
playtennis = yes)
     
prior.yes; # P(playtennis = yes)

  # Calculate unnormlized posterior probability
for playtennis = no.
  playtennis.no <-
   
sum((data[,1] == condition.vec[1]) & (data[,5] ==
“no”)) / sum(data[,5] ==
“no”) * # P(outlook = f_1 |
playtennis = no)
   
sum((data[,2] == condition.vec[2]) & (data[,5] ==
“no”)) / sum(data[,5] ==
“no”) * # P(temperature = f_2 |
playtennis = no)
   
sum((data[,3] == condition.vec[3]) & (data[,5] ==
“no”)) / sum(data[,5] ==
“no”) * # P(humidity = f_3 |
playtennis = no)
   
sum((data[,4] == condition.vec[4]) & (data[,5] ==
“no”)) / sum(data[,5] ==
“no”) * # P(wind = f_4 |
playtennis = no)
     
prior.no; # P(playtennis = no)

  return(list(post.pr.yes =
playtennis.yes,
             
post.pr.no = playtennis.no,
             
prediction = ifelse(playtennis.yes >=
               
playtennis.no, “yes”,
“no”)));
}

naive.bayes.prediction(c(“rain”,”hot”,”high”,”strong”));

naive.bayes.prediction(c(“sunny”,”mild”,”normal”,”weak”));

naive.bayes.prediction(c(“overcast”,”mild”,”normal”,”weak”));

 

#p#分页标题#e#

执行后，输出功效：

> naive.bayes.prediction(c("rain","hot","high","strong"));
$post.pr.yes
[1] 0.005291005

$post.pr.no
[1] 0.02742857

$prediction
[1] "no"

> naive.bayes.prediction(c("sunny","mild","normal","weak"));
$post.pr.yes
[1] 0.02821869

$post.pr.no
[1] 0.006857143

$prediction
[1] "yes"

> naive.bayes.prediction(c("overcast","mild","normal","weak"));
$post.pr.yes
[1] 0.05643739

$post.pr.no
[1] 0

$prediction
[1] "yes"

本文转载自：http://blog.sina.com.cn/s/blog_61c4630901013qoj.html