Skip to main content
The text elements considered keywords are configurable. They can include detected noun phrases (compound nouns like ‘the quick brown fox’), any automatically recognized entities (people, products, events), or any lexical category of word, such as nouns, verbs, adjectives etc.

Usage

The following examples show how the step can be used in a recipe.

Examples

  • Example 1
  • Example 2
  • Signature
To extract all kinds of nouns only, i.e. entities, compound nouns, simple nouns and proper nouns (names):
extract_keywords(ds.text, ds.lang,
  {
    "keywords": {
      "entities": true,
      "noun_phrases": true,
      "pos_tags": ["NOUN", "PROPN"]
    }
  }) -> (ds.keywords)

Inputs & Outputs

The following are the inputs expected by the step and the outputs it produces. These are generally columns (ds.first_name), datasets (ds or ds[["first_name", "last_name"]]) or models (referenced by name e.g. "churn-clf").
text
column[text]
required
A text column to extract keywords from.
*lang
column[category]
An (optional) column identifying the languages of the corresponding texts. It is used to identify the correct model (spaCy) to use for each text. If the dataset doesn’t contain such a column yet, it can be created using the infer_language step. Ideally, languages should be expressed as two-letter ISO 639-1 language codes, such as “en”, “es” or “de” for English, Spanish or German respectively. We also detect fully spelled out names such as “english”, “German”, “allemande” etc., but it is not guaranteed that we will recognize all possible spellings correctly always, so ISO codes should be preferred.Alternatively, if all texts are in the same language, it can be identified with the language parameter instead.
keywords
column[list[category]]
required
Lists containing the keywords mentioned in each text.

Configuration

The following parameters can be used to configure the behaviour of the step by including them in a json object as the last “input” to the step, i.e. step(..., {"param": "value", ...}) -> (output).

Parameters

extended_language_support
boolean
default:"false"
Whether to enable support for additional languages. By default, Arabic (“ar”), Catalan (“ca”), Basque (“eu”), and Turkish (“tu”) are not enabled, since they’re supported only by a different class of language models (stanfordNLP’s Stanza) that is much slower than the rest. This parameter can be used to enable them.
min_language_freq
[number, integer]
default:"0.02"
Minimum number (or proportion) of texts to include a language in processing. Any texts in a language with fewer documents than these will be ignored. Can be useful to speed up processing when there is noise in the input languages, and when ignoring languages with a small number of documents only is acceptable. Values smaller than 1 will be interpreted as a proportion of all texts, and values greater than or equal to 1 as an absolute number of documents.
  • number
  • integer
{_}
number
number.Values must be in the following range:
0 < {_} < 1
language
[string, null]
The language of inputs texts. If all texts are in the same language, it can be specified here instead of passing it as an input column. The language will be used to identify the correct spaCy model to parse and analyze the texts. For allowed values, see the comment regarding the lang column above.
keywords
object
Configure how keywords are extracted. Define the text elements considered keywords and their minimum or maximum frequency in the dataset to be included in the result.
pos_tags
array[string]
default:"['NOUN', 'PROPN', 'ADJ']"
Part-Of-Speech (POS) tags. Which lexical units (nouns, verbs etc.) to include as keywords. See spaCy’s universal part-of-speech tags for a detailed table of allowed values.
Item
string
Each item in array.Values must be one of the following:ADJ ADP ADV AUX CONJ CCONJ DET INTJ NOUN NUM PART PRON PROPN PUNCT SCONJ SYM VERB
entities
boolean
default:"true"
Whether or not to include any detected entities (people, places, events, etc.).
noun_phrases
boolean
default:"true"
Whether or not to include compound noun phrases (such as ‘the quick red fox’).
frequency_filter
object
Filter keywords based on the number of texts they occur in. Filter conditions can be applied globally or per language.
min_rows
integer
default:"2"
Minimum number of rows. Keywords not occurring in at least these many rows (texts) will be excluded.Values must be in the following range:
0min_rows < inf
max_rows
number
default:"0.5"
Maximum proportion of rows. Keywords occurring in more than this proportion of rows (texts) will be excluded.Values must be in the following range:
0max_rows1
keep_top_n
integer
Whether to always include the n most frequent keywords. I.e. independent of any other filter conditions. Set to null to ignore.Values must be in the following range:
0keep_top_n < inf
filter_top_n
integer
default:"0"
Whether to exclude n most frequent keywords. I.e. independent of any other filter conditions.Values must be in the following range:
0filter_top_n < inf
by_lang
boolean
default:"false"
Filter per language. Apply filter conditions separately to texts grouped by language, rather than across all texts.
I