We are proud to announce the version 3.0 release of the quanteda package, just over a year following our last major release of v2.0. Version 3.0 is a significant update that makes quanteda and its growing family of extension packages more solid, more consistent, and more extensible.

Main changes

Modularisation

We have now separated the textplot_*() functions from the main package into a separate package quanteda.textplots, and the textstat_*() functions from the main package into a separate package quanteda.textstats. This completes the modularisation begun in v2 with the move of the textmodel_*() functions to the separate package quanteda.textmodels](https://github.com/quanteda/quanteda.textmodels). The quanteda package now consists only of core functions for textual data processing and management.

Dependencies

v3 has a much lighter dependency footprint. Its package dependency structure is now greatly reduced, by eliminating some unnecessary package dependencies, by modularising the quanteda packages, and by addressing complex downstream dependencies in packages such as stopwords. v3 should serve as a more lightweight and more consistent platform upon which other text analysis package developers can build.

Non-standard evaluation

v3 brings a new, consistent implementation of direct evaluation within docvars for by and groups arguments:

  • The *_sample() functions’ argument by, and groups in the *_group() functions, now take unquoted document variable (docvar) names directly, similar to the way the subset argument works in the *_subset() functions.
  • For groups, the default is now docid(x), which is now documented more completely. See ?groups and ?docid.
  • The by = "document" formerly sampled from docid(x), but this functionality is now removed. Instead, use by = docid(x) to replicate this functionality.
  • Quoted docvar names no longer work, as these will be evaluated literally. This may break some existing code, but it makes the usage of the by and groups arguments consistent with how other functions in R work.

Deprecation of previous “shortcut” functions

To enforce a more consistent workflow and one that users have to control more explicitly, we now require objects that operate on tokens to work only with tokenised inputs. Previously, some functions such as dfm() and kwic() worked directly with untokenised inputs, such as character or corpus objects, and tokenised these on the fly, with additional arguments to tokens() fed via ....

In v3, we now restrict such functions to tokens() objects. This may be less convenient in that users previously could run dfm() or kwic() on a corpus object directly. The new usage, by contrast, requires users to take more direct control of tokenization options, or to substitute the alternative tokeniser of their choice (and then coercing it to tokens via as.tokens()). This also allows our function behaviour to be more consistent, with each function performing a single task, rather than combining functions (such as tokenisation and constructing a matrix).

The most commonly used shortcut involved constructing a dfm directly from a character or corpus object. Formerly, this would construct a tokens object internally before creating the dfm, and allowed passing arguments to tokens() via .... This is now deprecated, although still functional with a warning.

What are the advantages of skipping the shortcut approach? Requiring the text processing steps to be explicit means that users will be in greater control of the consequences of the sequencing of these steps. Some processing steps are sequence-dependent, for instance when a user removes stopwords and also applies a stemmer.

txt <- "Because during the concert it's very loud."
tokens(txt, remove_punct = TRUE) %>%
    tokens_remove(stopwords("en")) %>%
    tokens_wordstem()
## Tokens consisting of 1 document.
## text1 :
## [1] "concert" "loud"

is different from:

txt <- "Because during the concert it's very loud."
tokens(txt, remove_punct = TRUE) %>%
    tokens_wordstem() %>%
    tokens_remove(stopwords("en"))
## Tokens consisting of 1 document.
## text1 :
## [1] "Becaus"  "dure"    "concert" "veri"    "loud"

While the first is probably what nearly all users want, by removing these from options previously hard-wired via options to dfm() now requires the user to choose the sequence. Now, users must either creating a tokens object first, or pipe the tokens return to dfm() using %>%. This has always been possible prior to v3, of course, but now it is required.

We have also deprecated direct character or corpus inputs to kwic(), since this also requires a tokenised input.

These are deprecations rather than removals, since in v3.x the deprecated arguments and methods still work, but with deprecation warnings. We strongly encourage users to switch to the new workflow, as the deprecated arguments will be removed in the next major release.

Other new features

The full list of new features is the following:

  • dfm() has a new argument, remove_padding, for removing the “pads” left behind after removing tokens with padding = TRUE. (For other extensive changes to dfm(), see “Deprecated” below.)

  • tokens_group(), formerly internal-only, is now exported.

  • corpus_sample(), dfm_sample(), and tokens_sample() now work consistently.

  • The kwic() return object structure has been redefined, and built with an option to use a new function index() that returns token spans following a pattern search.

  • The punctuation regular expression and that for matching social media usernames has now been redefined so that the valid Twitter username @_ is now counted as a “tag” rather than as “punctuation”.

  • The data object data_corpus_inaugural has been updated to include the Biden 2021 inaugural address.

  • A new system of validators for input types now provides better argument type and value checking, with more consistent error messages for invalid types or values.

  • Upon startup, we now message the console with the Unicode and ICU version information. Because we removed our redefinition of View() (see below), the former conflict warning is now gone.

  • as.character.corpus() now has a use.names = TRUE argument, similar to as.character.tokens() (but with a different default value).

Deprecations

In addition to the deprecation of convenience shortcuts noted above, the following usages are also deprecated in v3.

  • dfm(): As of version 3, only tokens objects are supported as inputs to dfm() (as well as a dfm as input). Calling dfm() for character or corpus objects is still functional, but issues a warning. Convenience passing of arguments to tokens() via ... for dfm() is also deprecated, but undocumented, and functions only with a warning. Users should now create a tokens object (using tokens() from character or corpus inputs before calling dfm().

  • kwic(): As of version 3, only tokens objects are supported as inputs to kwic(). Calling kwic() for character or corpus objects is still functional, but issues a warning. Passing arguments to tokens() via ... in kwic() is now disabled. Users should now create a tokens object (using tokens() from character or corpus inputs before calling kwic()).

  • (as noted above) Shortcut arguments to dfm() are now deprecated. These are still active, with a warning, although they are no longer documented. These are:

    • stem: use tokens_wordstem() or dfm_wordstem() instead.
    • select, remove: use tokens_select() / dfm_select() or tokens_remove() / dfm_remove() instead.
    • dictionary, thesaurus: use tokens_lookup() or dfm_lookup() instead.
    • valuetype, case_insensitive: these are disabled; for the deprecated arguments that take these qualifiers, they are fixed to the defaults "glob" and TRUE.
    • groups: use tokens_group() or dfm_group() instead.

  • texts() and texts<- are deprecated.

    • Use as.character.corpus() to turn a corpus into a simple named character vector.
    • Use corpus_group() instead of texts(x, groups = ...) to aggregate texts by a grouping variable.
    • Use [<- instead of texts()<- for replacing texts in a corpus object. To replace all of the texts in a corpus, while keeping it a corpus, use [], e.g., to replace all texts in a five-document corpus:
    corp <- data_corpus_inaugural[1:5]
# equivalent to old usage of texts(corp) <- LETTERS[1:5]
corp[] <- LETTERS[1:5]

Removals

Finally, we have removed some previously deprecated arguments and functions, and moved others to different packages.

  • The textplot_*() and textstat_*() functions are now moved to quanteda.textplots and quanteda.textstats, respectively.

  • The following functions have been removed:

    • all methods for defunct corpuszip objects.
    • View() functions – so no more namespace conflict warnings on startup!
    • as.wfm() and as.DocumentTermMatrix() (the same functionality is available via convert())
    • metadoc() and metacorpus()
    • corpus_trimsentences() (replaced by corpus_trim())
    • all of the tortl functions.

  • dfm objects can no longer be used as a pattern in dfm_select() (formerly deprecated).

  • dfm_sample():

    • no longer has a margin argument. Instead, dfm_sample() now samples only on documents, the same as corpus_sample() and tokens_sample(); and
    • no longer works with by = "document" – use by = docid(x) instead.

  • dictionary_edit(), char_edit(), and list_edit() are removed.

  • dfm_weight(): the formerly deprecated "scheme" options are now removed.

  • tokens(): The formerly deprecated options remove_hyphens and remove_twitter are now removed. (Use split_hyphens instead, and the default tokenizer always now preserves Twitter and other social media tags.)

  • Special versions of head() and tail() for corpus, dfm, and fcm objects are now removed, since the base methods work fine for these objects. The main consequence was the removal of the nf option from the methods for dfm and fcm objects, which limited the number of features. This can be accomplished using the index operator [ instead, or for printing, by specifying print(x, max_nfeat = 6L) (for instance).

Acknowledgements

We wish to thank the CRAN maintainers, especially Kurt Hornik, for their patience and assistance in preparing this tricky release, which involved some tricky tests and a refresh of all of the quanteda packages at the same time.

We also thank the authors of the quanteda’s numerous reverse-dependent packages for working with us to update their code to work with the changes introduced in version 3. We hope the new infrastructure and more consistent usage will provide an even more solid base on which your code can build.

comments powered by Disqus