R - find and list duplicate rows based on two columns

user_4789486 0
Viewed 24611

Using R. Base package, dplyr, or data.table are all okay for me to use. My data is ~1000 rows x 20 columns. I expect about 300 duplicates.

I'd like to do something like the following, but with one alteration:

Match/group duplicate rows (indices)

I'd like to find, not fully duplicated rows, but rows duplicated in two columns. For example, given this input table:

File     T.N     ID     Col1     Col2
BAI.txt   T      1       sdaf    eiri
BAJ.txt   N      2       fdd     fds
BBK.txt   T      1       ter     ase
BCD.txt   N      1       twe     ase

If I want to find duplicates in T.N & ID only, I'd end up with the following table:

File     T.N     ID     Col1     Col2
BAI.txt   T      1       sdaf    eiri
BBK.txt   T      1       ter     ase
r
4 Answers

I have found this to be an easy and useful method.

tr <- tribble(~File,     ~TN,     ~ID,    ~Col1,     ~Col2,
              'BAI.txt',   'T',      1,       'sdaf',    'eiri',
              'BAJ.txt',   'N',     2,      'fdd',     'fds',
              'BBK.txt',   'T',      1,       'ter',     'ase',
              'BCD.txt',   'N',      1,       'twe',     'ase')

group_by(tr, TN, ID) %>% 
  filter(n() > 1)

Output:

# A tibble: 2 x 5
# Groups:   TN, ID [1]
  File    TN       ID Col1  Col2 
  <chr>   <chr> <dbl> <chr> <chr>
1 BAI.txt T         1 sdaf  eiri 
2 BBK.txt T         1 ter   ase
user_106878790

A simple solution is find_duplicates from hablar

library(dplyr)
library(data.table)
library(hablar)

df <- fread("
  File     T.N     ID     Col1     Col2
  BAI.txt   T      1       sdaf    eiri
  BAJ.txt   N      2       fdd     fds
  BBK.txt   T      1       ter     ase
  BCD.txt   N      1       twe     ase
            ")

df %>% 
  find_duplicates(T.N, ID)

which returns the rows with duplicates in T.N and ID:

  File    T.N      ID Col1  Col2 
  <chr>   <chr> <int> <chr> <chr>
1 BAI.txt T         1 sdaf  eiri 
2 BBK.txt T         1 ter   ase
user_78164820
Related