Find similar records

[PDD]

Hello
I have the following columns in my data table: surname, first name, postcode, Street ....
I found the duplicates in my data table with:
duplicate_indexes <- which(duplicated(my_data[c('surname', 'first_name')]),)
duplicates <- my_data[duplicate_indexes,]

Currently under duplicates I have only records which have exactly the same Information in each column of my data table (surname, first name, postcode, street are exactly the same) .

Now I would like also to find the records from my data table, which have similarities, e.g. surname or first name are the same and postcode is also the same. how can I do that?

Many thanks for your ideas.

[nirgrahamuk]

Consider these examples :

(mydata <- data.frame(surname=c(rep(LETTERS[1:5],2),"E"),
                     firstname=c(rep(letters[1:4],2),"d","a","x"),
                     postcode=c(rep(1:3,3),1,1)))

library(dplyr)
# dupliate over named variables only , i.e. surname & firstname
group_by(
  mydata,
  surname,
  firstname
) %>%
  mutate(
    n = n()
  ) %>%
  filter(n > 1)

#duplicated over postcode
group_by(
  mydata,
postcode
) %>%
  mutate(
    n = n()
  ) %>%
  filter(n > 1)

#surname or first name are the same and postcode is also the same
(surname_and_postcode_dup <- 
group_by(
  mydata,
  surname,postcode) %>%
  mutate(
    n = n()
  ) %>%
  filter(n > 1) )

(firstname_and_postcode_dup <- 
group_by(
  mydata,
  firstname,postcode) %>%
  mutate(
    n = n()
  ) %>%
  filter(n > 1))

(sur_or_first_and_post <-
  bind_rows(surname_and_postcode_dup,
            firstname_and_postcode_dup))

[PDD]

waw exactly what I need, thank you very much, you also showed me what I can do with my data