R | factor 因子

factor 因子

在R中数据的分类是用因子数据类型（factor）来表示的。如性别。

> #数据框 data.frame
> ID <- c(1:4)
> age <- c(25,34,28,72)
> treatment <- c("type1","type2","type3","type1")
> status <- c("poor","stable","poor","stable")
> pdata <- data.frame(ID,age,treatment,status)
> colnames <- c("ID","age","treatment","status")
> rownames <- pdata[1]
> pdata
  ID age treatment status
1  1  25     type1   poor
2  2  34     type2 stable
3  3  28     type3   poor
4  4  72     type1 improve
> summary(pdata)
       ID            age         treatment        
 Min.   :1.00   Min.   :25.00   Length:4          
 1st Qu.:1.75   1st Qu.:27.25   Class :character  
 Median :2.50   Median :31.00   Mode  :character  
 Mean   :2.50   Mean   :39.75                     
 3rd Qu.:3.25   3rd Qu.:43.50                     
 Max.   :4.00   Max.   :72.00                     
    status         
 Length:4          
 Class :character  
 Mode  :character          

我们想按照status对患者进行分类，所以将status由字符串改为factor

> pdata$status <- as.factor(pdata$status)
> class(pdata$status)
[1] "factor"
> summary(pdata)
       ID            age         treatment            status 
 Min.   :1.00   Min.   :25.00   Length:4           poor  :2  
 1st Qu.:1.75   1st Qu.:27.25   Class :character   stable:1  
 Median :2.50   Median :31.00   Mode  :character   improve:1        
 Mean   :2.50   Mean   :39.75                                
 3rd Qu.:3.25   3rd Qu.:43.50                                
 Max.   :4.00   Max.   :72.00                                

status是一个因子向量，表明了每个患者的病情是恶化、稳定、好转，类别总共有三大类。

因为通过factor对患者进行了分类，所以可以对不同类别进行作图：

> library(ggplot2)
> ggplot(data = pdata, aes(x = status)) + geom_bar()

image.png

levels()函数

对factor进行排序、增删，有利于清洗数据、作图、比较

> levels(pdata$status)
[1] "improve" "poor"    "stable" 

批量修改其中某一类别患者的status，如把improve改为stable

> pdata$status[pdata$status=="improve"] <-"stable"
> pdata
  ID age treatment status
1  1  25     type1   poor
2  2  34     type2 stable
3  3  28     type3   poor
4  4  72     type1 stable

此时检查还有哪些levels

> levels(pdata$status)
[1] "improve" "poor"    "stable" 

improve这一类还在。如果我们想要删掉这一类别，则droplevels()

> pdata$status <- droplevels(pdata$status)
> levels(pdata$status)
[1] "poor"   "stable"

此时没有对应患者的status被删掉

---

几种对levels的赋值方法

levels(f1)
levels(f2) <- value
attr(x, "levels") <- value

gl()函数：generate factor levels, 生成因子levels
gl(n,k,labels = c(), ordered = T)
n: 有几类
k: 每类重复几个
labels: 每类起什么名字
ordered：是否排序

> ## assign individual levels
> x <- gl(2, 4, 8)
> levels(x)[1] <- "low"
> levels(x)[2] <- "high"
> x
[1] low  low  low  low  high high high high
Levels: low high

> ## or as a group
> y <- gl(2, 4, 8)
> levels(y) <- c("low", "high")
> y
[1] low  low  low  low  high high high high
Levels: low high

将某几类归为一类

> ## combine some levels
> z <- gl(3, 2, 12, labels = c("apple", "salad", "orange"))
> z
 [1] apple  apple  salad  salad  orange orange apple  apple  salad 
[10] salad  orange orange
Levels: apple salad orange
> levels(z) <- c("fruit", "veg", "fruit")
> z
 [1] fruit fruit veg veg fruit fruit fruit fruit veg veg fruit fruit
Levels: fruit veg

重复出现的类别归为一类

> ## same, using a named list
> z <- gl(3, 2, 12, labels = c("apple", "salad", "orange"))
> z
 [1] apple  apple  salad  salad  orange orange apple  apple  salad 
[10] salad  orange orange
Levels: apple salad orange
> levels(z) <- list("fruit" = c("apple","orange"), "veg"   = "salad")
> z
 [1] fruit fruit veg   veg   fruit fruit fruit fruit veg   veg  
[11] fruit fruit
Levels: fruit veg

加入data.frame中没有出现的组别

> ## we can add levels this way:
> f <- factor(c("a","b"))
> levels(f) <- c("c", "a", "b")
> f
[1] c a
Levels: c a b

对组别的命名

> f <- factor(c("a","b"))
> levels(f) <- list(C = "C", A = "a", B = "b")
> f
[1] A B
Levels: C A B

将数值类型转换成因子

直接用as.numeric转换会有问题，转换后的内容不是你想要的：

> f <- factor(c(3.4,1.2,5))
> f
[1] 3.4 1.2 5  
Levels: 1.2 3.4 5
> as.numeric(f)
[1] 2 1 3

正确的方式是

> f <- factor(c(3.4,1.2,5))
> f <- levels(f)[f]
> f <- as.numeric(f)
> f
[1] 3.4 1.2 5.0