Method to trim and adjust objects by applying thresholds, minimum frequencies etc. It can be applied to context, features, context, partition and partition_bundle objects.

trim(.Object, ...)

# S4 method for class 'TermDocumentMatrix'
trim(
  .Object,
  terms_to_drop,
  docs_to_keep,
  min_count,
  min_doc_length,
  verbose = TRUE,
  ...
)

# S4 method for class 'DocumentTermMatrix'
trim(
  .Object,
  terms_to_drop,
  docs_to_keep,
  min_count,
  min_doc_length,
  verbose = TRUE,
  ...
)

punctuation

Format

An object of class character of length 13.

Arguments

.Object

The object to be trimmed

...

further arguments

terms_to_drop

A character vector with terms to exclude from matrix (terms used as stopwords).

docs_to_keep

A character vector with documents to keep.

min_count

A numeric value with a minimum value of total term frequency across documents to exclude rare terms from matrix.

min_doc_length

A numeric value with minimum total of the summed-up occurrence of tokens in a document. Exclude documents below this value and filter out short documents. Note that the min_doc_length filter is applied before filtering for min_count and terms_to_keep, and that these filters will reduce document lengths.

verbose

A logical value, whether to output progress messages.

Author

Andreas Blaette

Examples

use("RcppCWB", corpus = "REUTERS")
#> ✔ corpus loaded: REUTERS
dtm <- corpus("REUTERS") %>%
  split(s_attribute = "id") %>%
  as.DocumentTermMatrix(p_attribute = "word", verbose = FALSE)
#> ℹ s-attribute "id" has values: yes
#> ℹ get regions and get values
#> ✔ get regions and get values [14ms]
#> 
#> ℹ instantiate objects (n = 20)
#> ✔ instantiate objects (n = 20) [10ms]
#> 
#> ℹ assign names
#> ✔ assign names [9ms]
#> 
trim(dtm, min_doc_length = 100)
#> ℹ minimum document length 100 - dropped 7 docs (keeping 13)
#> <<DocumentTermMatrix (documents: 13, terms: 1192)>>
#> Non-/sparse entries: 2031/13465
#> Sparsity           : 87%
#> Maximal term length: 15
#> Weighting          : term frequency (tf)