利用R语言分析挖掘Titanic数据集(一)

时间：2021-09-10 22:11:47

简介

一个实际的数据挖掘项目包括6个阶段

1)提出正确的问题

问题本身确定了挖掘的对向与目标

2)数据采集

利用文件的i/o函数，JDBC/ODBC,网络爬虫技术从不同的系统，例如文件，数据库或internet采集数据，称为原始数据。由于原始数据存在存在格式的无序性与差异性问题，要利用分析工具与可视化程序来处理它们。

3)数据清洗

包括数据解析，排序，合并，筛选，缺失值插补以其其它各种数据转化和数据组织过程，最终得到一个合适数据分析的数据结构。

4)基础数据分析

进行基本的探索性数据分析，包括计算数据的汇总，采用基本的统计，聚类以及可视化方法来帮助用户更好的理解数据的特征，还可以通过图形来展现发现数据的主要性质，变化趋势，以及孤立点等。

5)高级数据分析

我们可以通过描述性统计得到有关数据特征的一个大概特征。但是我们希望从中得到一个大致推论，让用户以此为依据根据输入参数预测数据特征，这就必须借用机器学习的方法基于训练数据生成预测模型，在根据预测模型根据给定输入预测输出。

6)模型评估

为了评估生成的模型是否在给定领域能够得到最优的结果，还要进行模型的筛选。该任务通常包括多个步骤，包括参数的预处理，参数调优，机器学习算法切换。
下列样例子，我们将根据titanic获救乘客的数据，进行一个简单的数据挖掘，具体内容是从数据源kaggle取得数据（可能要fc),完成数据清洗，执行基本的数据分析分析，判断哪些属性对逃生概率有重要的影响。再执行深度数据挖掘，构建分类算法，根据给定的输入数据预测逃生概率。最后进行模型评估并取得预测模型。

从csv读取数据

1)从kaggle下载数据

https://www.kaggle.com/c/titanic/data

2)设置路径
> setwd("d:/R-TT")
> getwd()
[1] "d:/R-TT"
3)使用read.csv读取数据
#读取之后用str（）查看，""之间没有空格
> train.data = read.csv("titanic.csv",na.strings = c("NA",""))
> str(train.data)
'data.frame':   891 obs. of  12 variables:
 $ PassengerId: int  1 2 3 4 5 6 7 8 9 10 ...
 $ Survived   : int  0 1 1 1 0 0 0 0 1 1 ...
 $ Pclass     : int  3 1 3 1 3 3 1 3 3 2 ...
 $ Name       : Factor w/ 891 levels "Abbing, Mr. Anthony",..: 109 191 358 277 16 559 520 629 417 581 ...
 $ Sex        : Factor w/ 2 levels "female","male": 2 1 1 1 2 2 2 2 1 1 ...
 $ Age        : num  22 38 26 35 35 NA 54 2 27 14 ...
 $ SibSp      : int  1 1 0 1 0 0 0 3 0 1 ...
 $ Parch      : int  0 0 0 0 0 0 0 1 2 0 ...
 $ Ticket     : Factor w/ 681 levels "110152","110413",..: 524 597 670 50 473 276 86 396 345 133 ...
 $ Fare       : num  7.25 71.28 7.92 53.1 8.05 ...
 $ Cabin      : Factor w/ 147 levels "A10","A14","A16",..: NA 82 NA 56 NA NA 130 NA NA NA ...
 $ Embarked   : Factor w/ 3 levels "C","Q","S": 3 1 3 3 3 2 3 3 3 1 ...
> train.data$Survived = factor(train.data$Survived)
> train.data$Pclass = factor(train.data$Pclass)
> str(train.data)
'data.frame':   891 obs. of  12 variables:
 $ PassengerId: int  1 2 3 4 5 6 7 8 9 10 ...
 $ Survived   : Factor w/ 2 levels "0","1": 1 2 2 2 1 1 1 1 2 2 ...
 $ Pclass     : Factor w/ 3 levels "1","2","3": 3 1 3 1 3 3 1 3 3 2 ...
 $ Name       : Factor w/ 891 levels "Abbing, Mr. Anthony",..: 109 191 358 277 16 559 520 629 417 581 ...
 $ Sex        : Factor w/ 2 levels "female","male": 2 1 1 1 2 2 2 2 1 1 ...
 $ Age        : num  22 38 26 35 35 NA 54 2 27 14 ...
 $ SibSp      : int  1 1 0 1 0 0 0 3 0 1 ...
 $ Parch      : int  0 0 0 0 0 0 0 1 2 0 ...
 $ Ticket     : Factor w/ 681 levels "110152","110413",..: 524 597 670 50 473 276 86 396 345 133 ...
 $ Fare       : num  7.25 71.28 7.92 53.1 8.05 ...
 $ Cabin      : Factor w/ 147 levels "A10","A14","A16",..: NA 82 NA 56 NA NA 130 NA NA NA ...
 $ Embarked   : Factor w/ 3 levels "C","Q","S": 3 1 3 3 3 2 3 3 3 1 ...
4)注意

na.strings 代表缺失数据的值(转化为NA)，不参与运算

3.根据数据类型进行转化

#将int型数据转化成factor型.与上面对比有两列数据转化为factor
> train.data$Survived = factor(train.data$Survived)
> train.data$Pclass = factor(train.data$Pclass)
> str(train.data)
'data.frame':   891 obs. of  12 variables:
 $ PassengerId: int  1 2 3 4 5 6 7 8 9 10 ...
 $ Survived   : Factor w/ 2 levels "0","1": 1 2 2 2 1 1 1 1 2 2 ...
 $ Pclass     : Factor w/ 3 levels "1","2","3": 3 1 3 1 3 3 1 3 3 2 ...
 $ Name       : Factor w/ 891 levels "Abbing, Mr. Anthony",..: 109 191 358 277 16 559 520 629 417 581 ...
 $ Sex        : Factor w/ 2 levels "female","male": 2 1 1 1 2 2 2 2 1 1 ...
 $ Age        : num  22 38 26 35 35 NA 54 2 27 14 ...
 $ SibSp      : int  1 1 0 1 0 0 0 3 0 1 ...
 $ Parch      : int  0 0 0 0 0 0 0 1 2 0 ...
 $ Ticket     : Factor w/ 681 levels "110152","110413",..: 524 597 670 50 473 276 86 396 345 133 ...
 $ Fare       : num  7.25 71.28 7.92 53.1 8.05 ...
 $ Cabin      : Factor w/ 147 levels "A10","A14","A16",..: NA 82 NA 56 NA NA 130 NA NA NA ...
 $ Embarked   : Factor w/ 3 levels "C","Q","S": 3 1 3 3 3 2 3 3 3 1 ...

Survived(0=NO,1=YES)和Pclass(1=1st,2=2nd,3=3rd)都属于定类变量，我们使用factor函数将这两个变量转化为类型因子

4.检查缺失值

1)做准备

将属性Survived和Pclass转换为因子类型
R用NA(not available)代表缺失值，用NaN代表（not a number）代表不存在的值

2)进行操作
#用is.na()判断当成属于值是否包含NA值
 is.na(train.data$Age)
  [1] FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
 [16] FALSE FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE  TRUE  TRUE
 [31] FALSE  TRUE  TRUE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE
 [46]  TRUE  TRUE  TRUE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE
 [61] FALSE FALSE FALSE FALSE  TRUE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
 [76] FALSE  TRUE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE
 [91] FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE
[106] FALSE FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
[121] FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE
[136] FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
[151] FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE  TRUE  TRUE FALSE FALSE FALSE FALSE FALSE
[166] FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE
[181]  TRUE  TRUE FALSE FALSE FALSE  TRUE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
[196] FALSE  TRUE FALSE  TRUE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
[211] FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE
[226] FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE
[241]  TRUE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE
[256] FALSE  TRUE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE
[271]  TRUE FALSE FALSE FALSE  TRUE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE
[286] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE  TRUE FALSE
[301]  TRUE  TRUE FALSE  TRUE  TRUE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
[316] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE
[331]  TRUE FALSE FALSE FALSE  TRUE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
[346] FALSE FALSE  TRUE FALSE FALSE FALSE  TRUE FALSE FALSE  TRUE FALSE FALSE FALSE  TRUE  TRUE
[361] FALSE FALSE FALSE FALSE  TRUE FALSE FALSE  TRUE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE
[376]  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE  TRUE FALSE
[391] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
[406] FALSE FALSE FALSE FALSE  TRUE  TRUE  TRUE FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE FALSE
[421]  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE  TRUE FALSE FALSE  TRUE FALSE FALSE FALSE
[436] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE
[451] FALSE  TRUE FALSE FALSE  TRUE FALSE FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE
[466] FALSE  TRUE FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE
[481] FALSE  TRUE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE
[496]  TRUE FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE
[511] FALSE  TRUE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE  TRUE
[526] FALSE FALSE  TRUE FALSE FALSE FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE
[541] FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE
[556] FALSE FALSE  TRUE FALSE FALSE  TRUE FALSE FALSE  TRUE  TRUE FALSE FALSE FALSE  TRUE FALSE
[571] FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE  TRUE
[586] FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE  TRUE FALSE FALSE  TRUE FALSE  TRUE FALSE
[601] FALSE  TRUE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE  TRUE  TRUE FALSE
[616] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE
[631] FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE  TRUE FALSE
[646] FALSE FALSE FALSE  TRUE FALSE  TRUE FALSE FALSE  TRUE FALSE FALSE  TRUE FALSE FALSE FALSE
[661] FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE
[676] FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
[691] FALSE FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
[706] FALSE FALSE FALSE FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE
[721] FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE
[736] FALSE FALSE FALSE  TRUE  TRUE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
[751] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE
[766] FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE  TRUE FALSE  TRUE FALSE
[781] FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE  TRUE  TRUE FALSE
[796] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
[811] FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
[826]  TRUE  TRUE FALSE  TRUE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE  TRUE
[841] FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE
[856] FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE
[871] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE
[886] FALSE FALSE FALSE  TRUE FALSE FALSE
#is.na()进行了标记，用sum()统计缺失值总数
> sum(is.na(train.data$Age) == TRUE)
[1] 177
#用缺失总数除以非缺失数来计算缺失比例
> sum(is.na(train.data$Age) == TRUE)/length(train.data$Age)
[1] 0.1986532
#用sapply函数来计算所有属性缺失值的比例：
> sapply(train.data, function(df){
+   sum(is.na(df==TRUE))/length(df)
+ })
PassengerId    Survived      Pclass        Name         Sex         Age       SibSp       Parch 
0.000000000 0.000000000 0.000000000 0.000000000 0.000000000 0.198653199 0.000000000 0.000000000 
     Ticket        Fare       Cabin    Embarked 
0.000000000 0.000000000 0.771043771 0.002244669 
#除了观察缺失值的比例，我们可以使用Amelia包对缺失值进行可视化处理
> library(Rcpp)
> library(Amelia)
#使用missmap函数绘制缺失值示意图
> missmap(train.data,main = "MISSINGMAP")

利用R语言分析挖掘Titanic数据集(一)

5.插补缺失值

1）执行以下操作增补缺失值

#首先列出出发港口的分布。增加了useNA = "alays"的参数设置，展示train.data中最大的两个港口个数
> table(train.data$Embarked,useNA = "always")

   C    Q    S <NA> 
168   77  644    2 
#将其中两个缺失值处理为概率最大的两个港口
> train.data$Embarked[which(is.na(train.data$Embarked))] = 'S'
> table(train.data$Embarked,useNA = "always")

   C    Q    S <NA> 
168   77  646    0 
#AGE有很多缺失值，考虑到age与称乎Name在很大的相关性，我们可以根据将其所属年龄的平均值进行插补。
#首先将name 转化成character
>train.data$Name = as.character(train.data$Name)

#strsplit（）把字符串按照某个规则进行拆分，\\s表示   空格,回车,换行等空白符,  +号表示一个或多个的意思,这里我们只摘抄了后面几行展示分类规则。
>strsplit(train.data$Name,"\\s+")

.......
[[883]]
[1] "Dahlberg," "Miss."     "Gerda"     "Ulrika"   

[[884]]
[1] "Banfield," "Mr."       "Frederick" "James"    

[[885]]
[1] "Sutehall," "Mr."       "Henry"     "Jr"       

[[886]]
[1] "Rice,"     "Mrs."      "William"   "(Margaret" "Norton)"  

[[887]]
[1] "Montvila," "Rev."      "Juozas"   

[[888]]
[1] "Graham,"  "Miss."    "Margaret" "Edith"   

[[889]]
[1] "Johnston,"  "Miss."      "Catherine"  "Helen"      "\"Carrie\""

[[890]]
[1] "Behr,"  "Mr."    "Karl"   "Howell"

[[891]]
[1] "Dooley," "Mr."     "Patrick"





#生成的strsplit是列表形式，unlist()取消列表，这里我们只摘抄了后面几行展示这个过程
>unlist(strsplit(train.data$Name,"\\s+"))
......
[3598] "Mr."                     "Henry"                   "Jr"                     
[3601] "Rice,"                   "Mrs."                    "William"                
[3604] "(Margaret"               "Norton)"                 "Montvila,"              
[3607] "Rev."                    "Juozas"                  "Graham,"                
[3610] "Miss."                   "Margaret"                "Edith"                  
[3613] "Johnston,"               "Miss."                   "Catherine"              
[3616] "Helen"                   "\"Carrie\""              "Behr,"                  
[3619] "Mr."                     "Karl"                    "Howell"                 
[3622] "Dooley,"                 "Mr."                     "Patrick"  

#用table计算其出现的频次，
table_words = table(unlist(strsplit(train.data$Name,"\\s+")))
#我们用（"\\."）做为一种正则表达式，做为一种筛选的条件，sort()进行排序
> sort(table_words [grep("\\.",names(table_words))],decreasing = TRUE)

      Mr.     Miss.      Mrs.   Master.       Dr.      Rev.      Col.    Major.     Mlle. 
517       182       125        40         7         6         2         2         2 
    Capt. Countess.      Don. Jonkheer.        L.     Lady.      Mme.       Ms.      Sir. 
1         1         1         1         1         1         1         1         1 
#为了找到包含缺失值的不同人群，我们使用stringr包提供的str_match函数来匹配包含符号“.的子字符串，然后使用cbind函数将列拼在一起，最后，用table函数来获得缺失值的统计信息，并对每种人群进行计数
> library(stringr)
> tb = cbind(train.data$Age,str_match(train.data$Name,"[a-zA-Z]+\\."))
#tb的左侧列出了年龄包括缺省值，右侧列出了正式表达形式
> tb

       [,1]   [,2]       

  [1,] "22"   "Mr."      

  [2,] "38"   "Mrs."     

  [3,] "26"   "Miss."    

  [4,] "35"   "Mrs."     

  [5,] "35"   "Mr."      

  [6,] NA     "Mr."      

  [7,] "54"   "Mr."      

  [8,] "2"    "Master."  

  [9,] "27"   "Mrs."     

 [10,] "14"   "Mrs."     

 [11,] "4"    "Miss."    

 [12,] "58"   "Miss."    

 [13,] "20"   "Mr."      

 [14,] "39"   "Mr."      

 [15,] "14"   "Miss."    

 [16,] "55"   "Mrs."     

 [17,] "2"    "Master."  

 [18,] NA     "Mr."      

 [19,] "31"   "Mrs."     

 [20,] NA     "Mrs."   

#将为含有缺失值的对应列找出来

> tb[is.na(tb[,1]),2]
  [1] "Mr."     "Mr."     "Mrs."    "Mr."     "Miss."   "Mr."     "Mrs."    "Miss."   "Mr."    
 [10] "Mr."     "Mr."     "Mr."     "Miss."   "Mr."     "Mr."     "Mr."     "Master." "Mr."    
 [19] "Mr."     "Miss."   "Mr."     "Mr."     "Mr."     "Mr."     "Miss."   "Mr."     "Mr."    
 [28] "Miss."   "Mrs."    "Mr."     "Mr."     "Master." "Mrs."    "Mr."     "Master." "Miss."  
 [37] "Mr."     "Mr."     "Mrs."    "Mr."     "Miss."   "Mr."     "Mr."     "Mr."     "Miss."  
 [46] "Miss."   "Miss."   "Miss."   "Mr."     "Mrs."    "Mr."     "Miss."   "Mr."     "Miss."  
 [55] "Mr."     "Mr."     "Mr."     "Mr."     "Miss."   "Mr."     "Miss."   "Mr."     "Miss."  
 [64] "Mr."     "Miss."   "Mrs."    "Mr."     "Mrs."    "Mr."     "Mr."     "Miss."   "Miss."  
 [73] "Mr."     "Mrs."    "Miss."   "Mrs."    "Mr."     "Mr."     "Miss."   "Mr."     "Mr."    
 [82] "Mr."     "Mrs."    "Mr."     "Mr."     "Mr."     "Mrs."    "Mr."     "Mr."     "Mr."    
 [91] "Mrs."    "Mr."     "Mr."     "Mr."     "Mr."     "Mr."     "Mr."     "Mr."     "Miss."  
[100] "Mr."     "Mr."     "Mr."     "Miss."   "Mr."     "Mr."     "Mr."     "Mr."     "Mr."    
[109] "Mr."     "Mr."     "Mrs."    "Mr."     "Mr."     "Mr."     "Mr."     "Mr."     "Mr."    
[118] "Miss."   "Mr."     "Miss."   "Mrs."    "Mr."     "Mr."     "Miss."   "Miss."   "Mr."    
[127] "Mr."     "Mr."     "Mr."     "Miss."   "Mr."     "Mr."     "Mr."     "Mr."     "Mr."    
[136] "Mr."     "Mr."     "Miss."   "Mr."     "Mr."     "Mrs."    "Mr."     "Miss."   "Mr."    
[145] "Miss."   "Master." "Mr."     "Mr."     "Miss."   "Mr."     "Mr."     "Mr."     "Mr."    
[154] "Mr."     "Dr."     "Mr."     "Mr."     "Mr."     "Mr."     "Mr."     "Mr."     "Miss."  
[163] "Mr."     "Mr."     "Mr."     "Mr."     "Mr."     "Mr."     "Mr."     "Mr."     "Mr."    
[172] "Mrs."    "Mr."     "Miss."   "Mr."     "Mr."     "Miss."  

#对其进行统计

> table(tb[is.na(tb[,1]),2])

    Dr. Master.   Miss.     Mr.    Mrs. 

1       4      36     119      17 

#如果某个人群包含缺失值，插补的方式是将每一类人群平均值计算出来（不包含缺失值）,grepl检索目标行命令，"Mr\\."的\\表示绝对匹配
> mean.mr = mean(train.data$Age[grepl("Mr\\.",train.data$Name)&!is.na(train.data$Age)])
> 
> mean.mrs = mean(train.data$Age[grepl("Mrs\\.",train.data$Name)&!is.na(train.data$Age)])
> 
> mean.dr = mean(train.data$Age[grepl("Dr\\.",train.data$Name)&!is.na(train.data$Age)])
> 
> mean.miss = mean(train.data$Age[grepl("Miss\\.",train.data$Name)&!is.na(train.data$Age)])
> 
> mean.master = mean(train.data$Age[grepl("Master\\.",train.data$Name)&!is.na(train.data$Age)])

#将每类人的人属性均值插补到缺失值中
> train.data$Age[grepl("Mr\\.",train.data$Name)&is.na(train.data$Age)] = mean.mr
> 
> train.data$Age[grepl("Mrs\\.",train.data$Name)&is.na(train.data$Age)] = mean.mrs
> 
> train.data$Age[grepl("Dr\\.",train.data$Name)&is.na(train.data$Age)] = mean.dr
> train.data$Age[grepl("Miss\\.",train.data$Name)&is.na(train.data$Age)] = mean.miss
>train.data$Age[grepl("Master\\.",train.data$Name)&is.na(train.data$Age)] = mean.master

#对于缺失值，我们考虑他们的身份，将其所属人群的年龄平均值来插补缺失值，不过。对于Cabin属性，由于该属性缺失太多，没有办法从其它参考属性来推断，进一步分析中不在尝试使用该属性。
标签：

相关文章