natural
Version:
General natural language (tokenizing, stemming (English, Russian, Spanish), part-of-speech tagging, sentiment analysis, classification, inflection, phonetics, tfidf, WordNet, jaro-winkler, Levenshtein distance, Dice's Coefficient) facilities for node.
262 lines (216 loc) • 9.14 kB
JavaScript
/*
Copyright (c) 2018, Domingo Martín Mancera
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.
*/
'use strict'
const Stemmer = require('./stemmer_es')
// Inherit from the utility class in stemmer_es
class PorterStemmer extends Stemmer {
isVowel (c) {
const regex = /[aeiouáéíóú]/gi
return regex.test(c)
}
nextVowelPosition (word, start = 0) {
const length = word.length
for (let position = start; position < length; position++) {
if (this.isVowel(word[position])) {
return position
}
}
return length
}
nextConsonantPosition (word, start = 0) {
const length = word.length
for (let position = start; position < length; position++) {
if (!this.isVowel(word[position])) {
return position
}
}
return length
}
endsIn (word, suffix) {
if (word.length < suffix.length) {
return false
}
return (word.slice(-suffix.length) === suffix)
}
endsInArr (word, suffixes) {
const matches = []
for (const i in suffixes) {
if (this.endsIn(word, suffixes[i])) {
matches.push(suffixes[i])
}
}
const longest = matches.sort(function (a, b) {
return b.length - a.length
})[0]
if (longest) {
return longest
} else {
return ''
}
}
removeAccent (word) {
const accentedVowels = ['á', 'é', 'í', 'ó', 'ú']
const vowels = ['a', 'e', 'i', 'o', 'u']
for (const i in accentedVowels) {
word = word.replace(accentedVowels[i], vowels[i])
}
return word
}
stem (word) {
const length = word.length
word.toLowerCase()
if (length < 2) {
return this.removeAccent(word)
}
let r1, r2, rv
r1 = length
r2 = length
rv = length
// R1 is the region after the first non-vowel following a vowel, or is the null region
// at the end of the word if there is no such non-vowel.
for (let i = 0; i < (length - 1) && r1 === length; i++) {
if (this.isVowel(word[i]) && !this.isVowel(word[i + 1])) {
r1 = i + 2
}
}
// R2 is the region after the first non-vowel following a vowel in R1,
// or is the null region at the end of the word if there is no such non-vowel.
for (let i = r1; i < (length - 1) && r2 === length; i++) {
if (this.isVowel(word[i]) && !this.isVowel(word[i + 1])) {
r2 = i + 2
}
}
if (length > 3) {
if (!this.isVowel(word[1])) {
rv = this.nextVowelPosition(word, 2) + 1
} else if (this.isVowel(word[0]) && this.isVowel(word[1])) {
rv = this.nextConsonantPosition(word, 2) + 1
} else {
rv = 3
}
}
let r1Text = word.slice(r1)
let r2Text = word.slice(r2)
let rvText = word.slice(rv)
const originalWord = word
// Step 0: Attached pronoun
const pronounSuffix = ['me', 'se', 'sela', 'selo', 'selas', 'selos', 'la', 'le', 'lo', 'las', 'les', 'los', 'nos']
const pronounSuffixPre1 = ['iéndo', 'ándo', 'ár', 'ér', 'ír']
const pronounSuffixPre2 = ['iendo', 'ando', 'ar', 'er', 'ir']
const suffix = this.endsInArr(word, pronounSuffix)
if (suffix !== '') {
let preSuffix = this.endsInArr(rvText.slice(0, -suffix.length), pronounSuffixPre1)
if (preSuffix !== '') {
word = this.removeAccent(word.slice(0, -suffix.length))
} else {
preSuffix = this.endsInArr(rvText.slice(0, -suffix.length), pronounSuffixPre2)
if (preSuffix !== '' || (this.endsIn(word.slice(0, -suffix.length), 'uyendo'))) {
word = word.slice(0, -suffix.length)
}
}
}
if (word !== originalWord) {
r1Text = word.slice(r1)
r2Text = word.slice(r2)
rvText = word.slice(rv)
}
const wordAfter0 = word
let suf = null
if ((suf = this.endsInArr(r2Text, ['anza', 'anzas', 'ico', 'ica', 'icos', 'icas', 'ismo', 'ismos',
'able', 'ables', 'ible', 'ibles', 'ista', 'istas', 'oso', 'osa',
'osos', 'osas', 'amiento', 'amientos', 'imiento', 'imientos'])) !== '') {
word = word.slice(0, -suf.length)
} else if ((suf = this.endsInArr(r2Text, ['icadora', 'icador', 'icación', 'icadoras', 'icadores', 'icaciones',
'icante', 'icantes', 'icancia', 'icancias', 'adora', 'ador', 'ación',
'adoras', 'adores', 'aciones', 'ante', 'antes', 'ancia', 'ancias'])) !== '') {
word = word.slice(0, -suf.length)
} else if ((suf = this.endsInArr(r2Text, ['logía', 'logías'])) !== '') {
word = word.slice(0, -suf.length) + 'log'
} else if ((suf = this.endsInArr(r2Text, ['ución', 'uciones'])) !== '') {
word = word.slice(0, -suf.length) + 'u'
} else if ((suf = this.endsInArr(r2Text, ['encia', 'encias'])) !== '') {
word = word.slice(0, -suf.length) + 'ente'
} else if ((suf = this.endsInArr(r2Text, ['ativamente', 'ivamente', 'osamente', 'icamente', 'adamente'])) !== '') {
word = word.slice(0, -suf.length)
} else if ((suf = this.endsInArr(r1Text, ['amente'])) !== '') {
word = word.slice(0, -suf.length)
} else if ((suf = this.endsInArr(r2Text, ['antemente', 'ablemente', 'iblemente', 'mente'])) !== '') {
word = word.slice(0, -suf.length)
} else if ((suf = this.endsInArr(r2Text, ['abilidad', 'abilidades', 'icidad', 'icidades', 'ividad', 'ividades', 'idad', 'idades'])) !== '') {
word = word.slice(0, -suf.length)
} else if ((suf = this.endsInArr(r2Text, ['ativa', 'ativo', 'ativas', 'ativos', 'iva', 'ivo', 'ivas', 'ivos'])) !== '') {
word = word.slice(0, -suf.length)
}
if (word !== wordAfter0) {
r1Text = word.slice(r1)
r2Text = word.slice(r2)
rvText = word.slice(rv)
}
const wordAfter1 = word
if (wordAfter0 === wordAfter1) {
// Do step 2a if no ending was removed by step 1.
suf = this.endsInArr(rvText, ['ya', 'ye', 'yan', 'yen', 'yeron', 'yendo', 'yo', 'yó', 'yas', 'yes', 'yais', 'yamos'])
if (suf !== '' && (word.slice(-suf.length - 1, -suf.length) === 'u')) {
word = word.slice(0, -suf.length)
}
if (word !== wordAfter1) {
r1Text = word.slice(r1)
r2Text = word.slice(r2)
rvText = word.slice(rv)
}
const wordAfter2a = word
// Do Step 2b if step 2a was done, but failed to remove a suffix.
if (wordAfter2a === wordAfter1) {
if ((suf = this.endsInArr(rvText, ['arían', 'arías', 'arán', 'arás', 'aríais', 'aría', 'aréis',
'aríamos', 'aremos', 'ará', 'aré', 'erían', 'erías', 'erán',
'erás', 'eríais', 'ería', 'eréis', 'eríamos', 'eremos', 'erá',
'eré', 'irían', 'irías', 'irán', 'irás', 'iríais', 'iría', 'iréis',
'iríamos', 'iremos', 'irá', 'iré', 'aba', 'ada', 'ida', 'ía', 'ara',
'iera', 'ad', 'ed', 'id', 'ase', 'iese', 'aste', 'iste', 'an',
'aban', 'ían', 'aran', 'ieran', 'asen', 'iesen', 'aron', 'ieron',
'ado', 'ido', 'ando', 'iendo', 'ió', 'ar', 'er', 'ir', 'as', 'abas',
'adas', 'idas', 'ías', 'aras', 'ieras', 'ases', 'ieses', 'ís', 'áis',
'abais', 'íais', 'arais', 'ierais', ' aseis', 'ieseis', 'asteis',
'isteis', 'ados', 'idos', 'amos', 'ábamos', 'íamos', 'imos', 'áramos',
'iéramos', 'iésemos', 'ásemos'])) !== '') {
word = word.slice(0, -suf.length)
} else if ((suf = this.endsInArr(rvText, ['en', 'es', 'éis', 'emos'])) !== '') {
word = word.slice(0, -suf.length)
if (this.endsIn(word, 'gu')) {
word = word.slice(0, -1)
}
}
}
}
r1Text = word.slice(r1)
r2Text = word.slice(r2)
rvText = word.slice(rv)
if ((suf = this.endsInArr(rvText, ['os', 'a', 'o', 'á', 'í', 'ó'])) !== '') {
word = word.slice(0, -suf.length)
} else if ((this.endsInArr(rvText, ['e', 'é'])) !== '') {
word = word.slice(0, -1)
rvText = word.slice(rv)
if (this.endsIn(rvText, 'u') && this.endsIn(word, 'gu')) {
word = word.slice(0, -1)
}
}
return this.removeAccent(word)
}
}
module.exports = new PorterStemmer()