Skip to content

step_tfidf creates a specification of a recipe step that will convert a token variable into multiple variables containing the term frequency-inverse document frequency of tokens.


  role = "predictor",
  trained = FALSE,
  columns = NULL,
  vocabulary = NULL,
  res = NULL,
  smooth_idf = TRUE,
  norm = "l1",
  sublinear_tf = FALSE,
  prefix = "tfidf",
  keep_original_cols = FALSE,
  skip = FALSE,
  id = rand_id("tfidf")



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 recipes::selections() for more details.


For model terms created by this step, what analysis role should they be assigned?. By default, the function assumes that the new columns created by the original variables will be used as predictors in a model.


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


A character string of variable names that will be populated (eventually) by the terms argument. This is NULL until the step is trained by recipes::prep.recipe().


A character vector of strings to be considered.


The words that will be used to calculate the term frequency will be stored here once this preprocessing step has be trained by prep.recipe().


TRUE smooth IDF weights by adding one to document frequencies, as if an extra document was seen containing every term in the collection exactly once. This prevents division by zero.


A character, defines the type of normalization to apply to term vectors. "l1" by default, i.e., scale by the number of words in the document. Must be one of c("l1", "l2", "none").


A logical, apply sublinear term-frequency scaling, i.e., replace the term frequency with 1 + log(TF). Defaults to FALSE.


A character string that will be the prefix to the resulting new variables. See notes below.


A logical to keep the original variables in the output. Defaults to FALSE.


A logical. Should the step be skipped when the recipe is baked by recipes::bake.recipe()? While all operations are baked when recipes::prep.recipe() 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 = FALSE.


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


An updated version of recipe with the new step added to the sequence of existing steps (if any).


It is strongly advised to use step_tokenfilter before using step_tfidf to limit the number of variables created; otherwise you may run into memory issues. A good strategy is to start with a low token count and increase depending on how much RAM you want to use.

Term frequency-inverse document frequency is the product of two statistics: the term frequency (TF) and the inverse document frequency (IDF).

Term frequency measures how many times each token appears in each observation.

Inverse document frequency is a measure of how informative a word is, e.g., how common or rare the word is across all the observations. If a word appears in all the observations it might not give that much insight, but if it only appears in some it might help differentiate between observations.

The IDF is defined as follows: idf = log(1 + (# documents in the corpus) / (# documents where the term appears))

The new components will have names that begin with prefix, then the name of the variable, followed by the tokens all separated by -. The variable names are padded with zeros. For example if prefix = "hash", and if num_terms < 10, their names will be hash1 - hash9. If num_terms = 101, their names will be hash001 - hash101.


When you tidy() this step, a tibble with columns terms (the selectors or variables selected), token (name of the tokens), weight (the calculated IDF weight) is returned.

Case weights

The underlying operation does not allow for case weights.

See also

step_tokenize() to turn characters into tokens

Other Steps for Numeric Variables From Tokens: step_lda(), step_texthash(), step_tf(), step_word_embeddings()


# \donttest{

tate_rec <- recipe(~., data = tate_text) %>%
  step_tokenize(medium) %>%

tate_obj <- tate_rec %>%

bake(tate_obj, tate_text)
#> # A tibble: 4,284 × 956
#>        id artist             title     year tfidf_medium_1 tfidf_medium_10
#>     <dbl> <fct>              <fct>    <dbl>          <dbl>           <dbl>
#>  1  21926 Absalon            Proposa…  1990              0               0
#>  2  20472 Auerbach, Frank    Michael   1990              0               0
#>  3  20474 Auerbach, Frank    Geoffrey  1990              0               0
#>  4  20473 Auerbach, Frank    Jake      1990              0               0
#>  5  20513 Auerbach, Frank    To the …  1990              0               0
#>  6  21389 Ayres, OBE Gillian Phaëthon  1990              0               0
#>  7 121187 Barlow, Phyllida   Untitled  1990              0               0
#>  8  19455 Baselitz, Georg    Green V…  1990              0               0
#>  9  20938 Beattie, Basil     Present…  1990              0               0
#> 10 105941 Beuys, Joseph      Joseph …  1990              0               0
#> # … with 4,274 more rows, and 950 more variables: tfidf_medium_100 <dbl>,
#> #   tfidf_medium_11 <dbl>, tfidf_medium_12 <dbl>, tfidf_medium_13 <dbl>,
#> #   tfidf_medium_133 <dbl>, tfidf_medium_14 <dbl>, tfidf_medium_15 <dbl>,
#> #   tfidf_medium_151 <dbl>, tfidf_medium_16 <dbl>,
#> #   tfidf_medium_160 <dbl>, tfidf_medium_16mm <dbl>,
#> #   tfidf_medium_18 <dbl>, tfidf_medium_19 <dbl>, tfidf_medium_2 <dbl>,
#> #   tfidf_medium_20 <dbl>, tfidf_medium_2000 <dbl>, …

tidy(tate_rec, number = 2)
#> # A tibble: 1 × 4
#>   terms  token weight id         
#>   <chr>  <chr>  <dbl> <chr>      
#> 1 medium NA        NA tfidf_0tRhE
tidy(tate_obj, number = 2)
#> # A tibble: 952 × 4
#>    terms  token weight id         
#>    <chr>  <chr>  <dbl> <chr>      
#>  1 medium 1       7.26 tfidf_0tRhE
#>  2 medium 10      7.26 tfidf_0tRhE
#>  3 medium 100     7.26 tfidf_0tRhE
#>  4 medium 11      7.67 tfidf_0tRhE
#>  5 medium 12      7.67 tfidf_0tRhE
#>  6 medium 13      8.36 tfidf_0tRhE
#>  7 medium 133     8.36 tfidf_0tRhE
#>  8 medium 14      6.75 tfidf_0tRhE
#>  9 medium 15      6.57 tfidf_0tRhE
#> 10 medium 151     8.36 tfidf_0tRhE
#> # … with 942 more rows
# }