如何粘贴数据框行中的文本，只保留 R 中的唯一值

Question

我有一个数据框，其中每一行代表一个人，列代表他们的名字。一些值是 NA 或重复的。数据看起来像下面的数据框。

Name	Name1	Name2	Name3	Name4
Tom	Tom	Thomas	Tom	Tommy
Jim	NA	James	NA	Jimmy
Dave	Dave	David	NA	Davey
Tim	NA	Timothy	Tim	Timmy
Rob	Rob	NA	Rob	Robby
Sam	NA	NA	Sam	NA

我想合并每一行中的唯一名称并将它们放入一个新列中，其中每个名称只出现一次。我知道我可以使用粘贴功能生成一列，其中所有文本值都显示如下：

Name	Name1	Name2	Name3	Name4	unique
Tom	Tom	Thomas	NA	Tommy	Tom, Tom, Thomas, NA, Tommy

但我不希望相同的文本在唯一列中出现多次。 如何合并行数据，使每个名称在新的 $unique 单元格中只出现一次？

Name	Name1	Name2	Name3	Name4	unique
Tom	Tom	Thomas	Tom	Tommy	Tom, Thomas, Tommy
Jim	NA	James	NA	Jimmy	Jim, James, Jimmy
Dave	Dave	David	NA	Davey	Dave, David, Davey
Tim	NA	Timothy	Tim	Timmy	Tim, Timothy, Timmy
Rob	Rob	NA	Rob	Robby	Rob, Robert, Robby
Sam	NA	NA	Sam	NA	Sam

Answer 1

applying unique row-wise, na.omit 并折叠 toString.

transform(dat, unique=apply(dat, 1, \(x) toString(na.omit(unique(x)))))
#   Name Name1   Name2 Name3 Name4              unique
# 1  Tom   Tom  Thomas   Tom Tommy  Tom, Thomas, Tommy
# 2  Jim  <NA>   James  <NA> Jimmy   Jim, James, Jimmy
# 3 Dave  Dave   David  <NA> Davey  Dave, David, Davey
# 4  Tim  <NA> Timothy   Tim Timmy Tim, Timothy, Timmy
# 5  Rob   Rob    <NA>   Rob Robby          Rob, Robby
# 6  Sam  <NA>    <NA>   Sam  <NA>                 Sam

如果愿意，您还可以实施 sort。

---

数据：

dat <- structure(list(Name = c("Tom", "Jim", "Dave", "Tim", "Rob", "Sam"
), Name1 = c("Tom", NA, "Dave", NA, "Rob", NA), Name2 = c("Thomas", 
"James", "David", "Timothy", NA, NA), Name3 = c("Tom", NA, NA, 
"Tim", "Rob", "Sam"), Name4 = c("Tommy", "Jimmy", "Davey", "Timmy", 
"Robby", NA)), class = "data.frame", row.names = c(NA, -6L))

Answer 2

df%>%
 rowid_to_column()%>%
 left_join(pivot_longer(.,-rowid)%>%
 group_by(rowid)%>%    
 summarise(value=toString(na.omit(unique(value))), .groups = 'drop'))
  rowid Name Name1   Name2 Name3 Name4               value
1     1  Tom   Tom  Thomas   Tom Tommy  Tom, Thomas, Tommy
2     2  Jim  <NA>   James  <NA> Jimmy   Jim, James, Jimmy
3     3 Dave  Dave   David  <NA> Davey  Dave, David, Davey
4     4  Tim  <NA> Timothy   Tim Timmy Tim, Timothy, Timmy
5     5  Rob   Rob    <NA>   Rob Robby          Rob, Robby
6     6  Sam  <NA>    <NA>   Sam  <NA>                 Sam

Answer 3

使用tidyverse

library(dplyr)
df1 %>% 
 rowwise %>% 
  mutate(unique = toString(unique(na.omit(c_across(everything()))))) %>% 
  ungroup

-输出

# A tibble: 6 × 6
  Name  Name1 Name2   Name3 Name4 unique             
  <chr> <chr> <chr>   <chr> <chr> <chr>              
1 Tom   Tom   Thomas  Tom   Tommy Tom, Thomas, Tommy 
2 Jim   <NA>  James   <NA>  Jimmy Jim, James, Jimmy  
3 Dave  Dave  David   <NA>  Davey Dave, David, Davey 
4 Tim   <NA>  Timothy Tim   Timmy Tim, Timothy, Timmy
5 Rob   Rob   <NA>    Rob   Robby Rob, Robby         
6 Sam   <NA>  <NA>    Sam   <NA>  Sam