node-nltk-stopwords

A node module exposing nltk stopwords corpora and provide utility functions for removing stopwords

Usage

npm install --save nltk-stopwords

var stopwords = require('nltk-stopwords')

// let's load english stopwords
var english = stopwords.load('english')

// Remove stopwords from a sentence
stopwords.remove("A sentence can not be without stopwords", english)
// This should output
// 'A sentence without stopwords'

// You can also pass a string of language in the second parameter, `stopwords.remove` will handle stopword loading.
stopwords.remove("J'essaye de trouver un bon example", "french")
// This should output
// "J'essaye trouver bon example"

By default, the sentence is tokenized by whitespace. You can also pass an array of tokenized string instead.

stopwords.remove([ 'A', 'sentence', 'can', 'not', 'be', 'without', 'stopwords' ], english)
// This should output 
// [ 'A', 'sentence', 'without', 'stopwords' ]

If a string input is passed, stopwords.remove will return a string, if an array input is passed instead, the resulting return value will also be an array.