collapse factor levels using stringdist

step_collapse_stringdist() creates a specification of a recipe step that will collapse factor levels that have a low stringdist between them.

Usage

step_collapse_stringdist(
  recipe,
  ...,
  role = NA,
  trained = FALSE,
  distance = NULL,
  method = "osa",
  options = list(),
  results = NULL,
  columns = NULL,
  skip = FALSE,
  id = rand_id("collapse_stringdist")
)

Arguments

recipe

A recipe object. The step will be added to the sequence of operations for this recipe.

...

One or more selector functions to choose which variables are affected by the step. See selections() for more details. For the tidy method, these are not currently used.

role

Not used by this step since no new variables are created.

trained

A logical to indicate if the quantities for preprocessing have been estimated.

distance

Integer, value to determine which strings should be collapsed with which. The value is being used inclusive, so 2 will collapse levels that have a string distance between them of 2 or lower.

method

Character, method for distance calculation. The default is "osa", see stringdist::stringdist-metrics.

options

List, other arguments passed to stringdist::stringdistmatrix() such as weight, q, p, and bt, that are used for different values of method.

results

A list denoting the way the labels should be collapses is stored here once this preprocessing step has be trained by prep().

columns

A character string of variable names that will be populated (eventually) by the terms argument.

skip

A logical. Should the step be skipped when the recipe is baked by bake()? While all operations are baked when prep() is run, some operations may not be able to be conducted on new data (e.g. processing the outcome variable(s)). Care should be taken when using skip = TRUE as it may affect the computations for subsequent operations.

id

A character string that is unique to this step to identify it.

Value

An updated version of recipe with the new step added to the sequence of existing steps (if any). For the tidy method, a tibble with columns terms (the columns that will be affected) and base.

Tidying

When you tidy() this step, a tibble with columns "terms" (the column being modified), "from" (the old levels), "to" (the new levels), and "id".

Case weights

The underlying operation does not allow for case weights.

Examples

library(recipes)
library(tibble)
data0 <- tibble(
  x1 = c("a", "b", "d", "e", "sfgsfgsd", "hjhgfgjgr"),
  x2 = c("ak", "b", "djj", "e", "hjhgfgjgr", "hjhgfgjgr")
)

rec <- recipe(~., data = data0) %>%
  step_collapse_stringdist(all_predictors(), distance = 1) %>%
  prep()

rec %>%
  bake(new_data = NULL)
#> # A tibble: 6 × 2
#>   x1        x2       
#>   <fct>     <fct>    
#> 1 a         ak       
#> 2 a         b        
#> 3 a         djj      
#> 4 a         b        
#> 5 sfgsfgsd  hjhgfgjgr
#> 6 hjhgfgjgr hjhgfgjgr

tidy(rec, 1)
#> # A tibble: 11 × 4
#>    terms from      to        id                       
#>    <chr> <chr>     <chr>     <chr>                    
#>  1 x1    a         a         collapse_stringdist_F7qKG
#>  2 x1    b         a         collapse_stringdist_F7qKG
#>  3 x1    d         a         collapse_stringdist_F7qKG
#>  4 x1    e         a         collapse_stringdist_F7qKG
#>  5 x1    hjhgfgjgr hjhgfgjgr collapse_stringdist_F7qKG
#>  6 x1    sfgsfgsd  sfgsfgsd  collapse_stringdist_F7qKG
#>  7 x2    ak        ak        collapse_stringdist_F7qKG
#>  8 x2    b         b         collapse_stringdist_F7qKG
#>  9 x2    e         b         collapse_stringdist_F7qKG
#> 10 x2    djj       djj       collapse_stringdist_F7qKG
#> 11 x2    hjhgfgjgr hjhgfgjgr collapse_stringdist_F7qKG

rec <- recipe(~., data = data0) %>%
  step_collapse_stringdist(all_predictors(), distance = 2) %>%
  prep()

rec %>%
  bake(new_data = NULL)
#> # A tibble: 6 × 2
#>   x1        x2       
#>   <fct>     <fct>    
#> 1 a         ak       
#> 2 a         ak       
#> 3 a         djj      
#> 4 a         ak       
#> 5 sfgsfgsd  hjhgfgjgr
#> 6 hjhgfgjgr hjhgfgjgr

tidy(rec, 1)
#> # A tibble: 11 × 4
#>    terms from      to        id                       
#>    <chr> <chr>     <chr>     <chr>                    
#>  1 x1    a         a         collapse_stringdist_m6FIF
#>  2 x1    b         a         collapse_stringdist_m6FIF
#>  3 x1    d         a         collapse_stringdist_m6FIF
#>  4 x1    e         a         collapse_stringdist_m6FIF
#>  5 x1    hjhgfgjgr hjhgfgjgr collapse_stringdist_m6FIF
#>  6 x1    sfgsfgsd  sfgsfgsd  collapse_stringdist_m6FIF
#>  7 x2    ak        ak        collapse_stringdist_m6FIF
#>  8 x2    b         ak        collapse_stringdist_m6FIF
#>  9 x2    e         ak        collapse_stringdist_m6FIF
#> 10 x2    djj       djj       collapse_stringdist_m6FIF
#> 11 x2    hjhgfgjgr hjhgfgjgr collapse_stringdist_m6FIF