基于分组添加标注：为R数据框新增rel_from列的实现需求

示例数据 #

先构造一个符合需求的测试数据集：

mydata <- data.frame(
  Name = c("Alice", "Alice", "Bob", "Bob", "Charlie", "Charlie", "Dave", "Dave", "Eve", "Eve", "Eve"),
  Dep = c("HR", "HR", "Eng", "Eng", "Fin", "Fin", "Ops", "Ops", "Mar", "Mar", "Mar"),
  Loc = c("NY", "NY", "LA", "LA", "Chicago", "Chicago", "Boston", "Boston", "Dallas", "Dallas", "Dallas"),
  rel = c("Ind", "Boss", "Ind", "Dir", "Ind", "Boss", "Ind", "Dir", "Ind", "Boss", "Dir")
)

处理方案（使用dplyr） #

利用dplyr的分组和条件判断功能实现需求，步骤如下：

1. 加载dplyr包（未安装先运行install.packages("dplyr")）

library(dplyr)

2. 分组并添加rel_from列

mydata_processed <- mydata %>%
  # 按Name、Dep、Loc分组
  group_by(Name, Dep, Loc) %>%
  # 根据分组内的rel值，为Ind行赋值rel_from
  mutate(
    rel_from = case_when(
      # 非Ind行设为NA
      rel != "Ind" ~ NA_character_,
      # 同时存在Boss和Dir → 标注Both
      "Boss" %in% rel & "Dir" %in% rel ~ "Both",
      # 仅存在Boss → 标注Boss
      "Boss" %in% rel ~ "Boss",
      # 仅存在Dir → 标注Dir
      "Dir" %in% rel ~ "Dir",
      # 兜底情况（理论上不会触发）
      TRUE ~ NA_character_
    )
  ) %>%
  # 取消分组
  ungroup()

输出结果 #

运行上述代码后，得到的处理结果如下：

# A tibble: 11 × 5
   Name    Dep   Loc     rel   rel_from
   <chr>   <chr> <chr>   <chr> <chr>   
 1 Alice   HR    NY      Ind   Boss    
 2 Alice   HR    NY      Boss  NA      
 3 Bob     Eng   LA      Ind   Dir     
 4 Bob     Eng   LA      Dir   NA      
 5 Charlie Fin   Chicago Ind   Boss    
 6 Charlie Fin   Chicago Boss  NA      
 7 Dave    Ops   Boston  Ind   Dir     
 8 Dave    Ops   Boston  Dir   NA      
 9 Eve     Mar   Dallas  Ind   Both    
10 Eve     Mar   Dallas  Boss  NA      
11 Eve     Mar   Dallas  Dir   NA      

代码逻辑说明 #

• group_by(Name, Dep, Loc)：将数据按指定的三个列分组，确保每个组内是同一Name、Dep、Loc的记录
• case_when()：按顺序判断条件，为rel_from赋值：
  • 首先排除非Ind行，设为NA
  • 优先判断是否同时存在Boss和Dir，满足则标注"Both"
  • 其次判断是否存在Boss，满足则标注"Boss"
  • 最后判断是否存在Dir，满足则标注"Dir"
• ungroup()：处理完成后取消分组，恢复普通data.frame结构

内容的提问来源于stack exchange，提问作者akang