如何在 R 中的 dataframe 中創建具有三個級別的因子？

Question

我為 19 個不同的距離創建了一個因子，我需要確定三個級別，一個用於直接影響 (DirImp)，另一個用於我各自的間接影響距離，(Dist="1km_","2km_","3km_","4km_" ,"5km_","6km_","7km_","8km_","9km_","10km_","10km","20km","30km","40km","50km","60km"," 70km") 和其他到我的控制區域 (Contrl)，從距離 0 (DirImp) 開始，每公里增加公里，直到達到 10 公里，從這一點開始，每 10 公里增加一次，直到達到 70 公里，並且最后的距離是控制。

因此，為了澄清，在我的DataFrame中，我有一列（Dist），其中包含這些距離和其他列以及其他信息，我使用此代碼將其轉換為一個因子：

column Dist estructure:


levels(MY.DTAFRAME$Dist)
[1] "DirImp"   "10km"  "10km_" "1km_"  "20km"  "2km_"  "30km" 
[8] "3km_"  "40km"  "4km_"  "50km"  "5km_"  "60km"  "6km_" 
[15] "70km"  "7km_"  "8km_"  "9km_", "control" 

How I would like it to be:
level 1 = Direct impact ("DirImp")
level 2 = Distances ("1km_","2km_","3km_","4km_","5km_","6km_","7km_","8km_","9km_","10km_","10km","20km","30km","40km","50km","60km","70km")
level 3 = Contrl Area  ("Contrl")

Column Dist = ("DirImp", "1km_","2km_","3km_","4km_","5km_","6km_","7km_","8km_","9km_","10km_","10km","20km","30km","40km","50km","60km","70km", "control")

  MY.DATAFRAME$DistFact <- factor(MY.DATAFRAME$Dist, level ordered = TRUE)


  levels(MY.DTAFRAME$DistFact)
  [1] "DirImp"   "10km"  "10km_" "1km_"  "20km"  "2km_"  "30km" 
  [8] "3km_"  "40km"  "4km_"  "50km"  "5km_"  "60km"  "6km_" 
  [15] "70km"  "7km_"  "8km_"  "9km_", "control" 

Answer 1

問題要求的內容類似於以下內容嗎？

forcats::fct_collapse(y, 
                      DirImp = grep("DirImp", y, ignore.case = TRUE, value = TRUE), 
                      Distances = grep("km", y, ignore.case = TRUE, value = TRUE),
                      Control = grep("control", y, ignore.case = TRUE, value = TRUE)
                      )
# [1] Distances Distances Distances Distances Distances Distances
# [7] Distances Distances Distances Distances Distances Distances
#[13] Distances Distances Distances Distances Distances Distances
#[19] Distances Distances Distances Distances Distances Distances
#[25] Distances Distances Distances Distances Control   Distances
#Levels: DirImp Distances Control

或者，也許更具可讀性，

grep_tmp <- function(pattern, x){
  grep(pattern, x, ignore.case = TRUE, value = TRUE)
}

forcats::fct_collapse(y,
                      DirImp = grep_tmp("DirImp", y), 
                      Distances = grep_tmp("^\\d+km", y),
                      Control = grep_tmp("control", y)
                      )

數據

根據問題中發布的levels ，這里是示例數據。

set.seed(1234)
x <- scan(text = '"DirImp"   "10km"  "10km_" "1km_"  "20km"  "2km_"  "30km" 
"3km_"  "40km"  "4km_"  "50km"  "5km_"  "60km"  "6km_" 
"70km"  "7km_"  "8km_"  "9km_" "control"', what = character())

y <- factor(sample(x, 30, TRUE), levels = x)