retext-keywords

retext-keywords

retext plugin to extract keywords and key-phrases.

Install

This package is ESM only: Node 12+ is needed to use it and it must be imported instead of required.

npm:

npm install retext-keywords

Use

Say we have the following file, example.txt, with the first four paragraphs on term Extraction from Wikipedia:

Terminology mining, term extraction, term recognition, or glossary extraction, is a subtask of information extraction. The goal of terminology extraction is to automatically extract relevant terms from a given corpus.

In the semantic web era, a growing number of communities and networked enterprises started to access and interoperate through the internet. Modeling these communities and their information needs is important for several web applications, like topic-driven web crawlers, web services, recommender systems, etc. The development of terminology extraction is essential to the language industry.

One of the first steps to model the knowledge domain of a virtual community is to collect a vocabulary of domain-relevant terms, constituting the linguistic surface manifestation of domain concepts. Several methods to automatically extract technical terms from domain-specific document warehouses have been described in the literature.

Typically, approaches to automatic term extraction make use of linguistic processors (part of speech tagging, phrase chunking) to extract terminological candidates, i.e. syntactically plausible terminological noun phrases, NPs (e.g. compounds "credit card", adjective-NPs "local tourist information office", and prepositional-NPs "board of directors" - in English, the first two constructs are the most frequent). Terminological entries are then filtered from the candidate list using statistical and machine learning methods. Once filtered, because of their low ambiguity and high specificity, these terms are particularly useful for conceptualizing a knowledge domain or for supporting the creation of a domain ontology. Furthermore, terminology extraction is a very useful starting point for semantic similarity, knowledge management, human translation and machine translation, etc.

…and our script, example.js, looks as follows:

import {readSync} from 'to-vfile'
import {toString} from 'nlcst-to-string'
import {retext} from 'retext'
import retextPos from 'retext-pos'
import retextKeywords from 'retext-keywords'

const file = readSync('example.txt')

retext()
  .use(retextPos) // Make sure to use `retext-pos` before `retext-keywords`.
  .use(retextKeywords)
  .process(file)
  .then((file) => {
    console.log('Keywords:')
    file.data.keywords.forEach((keyword) => {
      console.log(toString(keyword.matches[0].node))
    })

    console.log()
    console.log('Key-phrases:')
    file.data.keyphrases.forEach((phrase) => {
      console.log(phrase.matches[0].nodes.map((d) => toString(d)).join(''))
    })
  })

Now, running node example yields:

Keywords:
term
extraction
Terminology
web
domain

Key-phrases:
terminology extraction
terms
term extraction
knowledge domain
communities

API

This package exports no identifiers. The default export is retextKeywords.

unified().use(retextKeywords[, options])

Extract keywords and key-phrases from the document.

The results are stored on file.data: keywords at file.data.keywords and key-phrases at file.data.keyphrases. Both are lists.

A single keyword looks as follows:

{
  stem: 'term',
  score: 1,
  matches: [
    {node: Node, index: 5, parent: Node},
    // …
  ],
  // …
}

…and a key-phrase:

{
  score: 1,
  weight: 11,
  stems: ['terminolog', 'extract'],
  value: 'terminolog extract',
  matches:  [
    {nodes: [Node, Node, Node], parent: Node},
    // …
  ]
}

options.maximum

Try to detect at most maximum words and phrases (number, default: 5).

Note that actual counts may differ. For example, when two words have the same score, both will be returned. Or when too few words exist, less will be returned. the same goes for phrases.

Contribute

See contributing.md in retextjs/.github for ways to get started. See support.md for ways to get help.

This project has a code of conduct. By interacting with this repository, organization, or community you agree to abide by its terms.

License

MIT © Titus Wormer