The goal of similars is to match strings to the most similar string in another vector.
You can install the released version of similiars with:
remotes::install_github("davidsjoberg/similiars")The main function included is find_most_similar_string which returns a
vector of the same length as the vector to be matched against another.
library(similiars)
x <- c("battman", "Y-men", "Supergrl")
find_most_similar_string(x, superheroes)
#> [1] "Batman" "X-Men" "Supergirl"The other function is find_string_distance. It returns a list of
dataframes for each word and how similar each string is to all matched
strings. If you are not sure if there is a good match to be found it is
good practise to use this function before using
find_most_similar_string.
find_string_distance(c("aple", "abna"),
c("banana", "apple", "pineapple"))
#> $aple
#> # A tibble: 3 x 3
#> input_string string string_distance
#> <chr> <chr> <dbl>
#> 1 aple apple 1
#> 2 aple banana 5
#> 3 aple pineapple 5
#>
#> $abna
#> # A tibble: 3 x 3
#> input_string string string_distance
#> <chr> <chr> <dbl>
#> 1 abna banana 3
#> 2 abna apple 4
#> 3 abna pineapple 7If you want manual control over which should be replaced you can create
a case_when code chunk to use in your script.
# Misspelled vector
supr_heros <- c("Btman", "Supergrl", "Tor", "Capitan 'Murica")
# Corrected vector
super_heroes <- find_most_similar_string(supr_heros, superheroes)
# Make case_when skeleton to replace
case_when_skeleton(supr_heros, super_heroes)
#> case_when(
#> supr_heros == "Btman" ~ "Batman",
#> supr_heros == "Supergrl" ~ "Supergirl",
#> supr_heros == "Tor" ~ "Thor",
#> supr_heros == "Capitan 'Murica" ~ "Captain America",
#> T ~ supr_heros
#> )If you find anything you want to improve, submit an issue :)