Text Analysis

/**
 * Text class for multi-sentence analysis and transformation
 */
class Text {
  /** Array of parsed sentences */
  sentences: Sentence[];
  /** Original input text */
  raw_text: string;
  /** Contraction expansion/contraction methods */
  contractions: Contractions;
}

/**
 * Get formatted text with original spacing and punctuation
 * @returns {string} Reconstructed text
 */
text(): string;

/**
 * Get normalized text with cleaned whitespace
 * @returns {string} Normalized text
 */
normal(): string;

/**
 * Get root/lemmatized form of all words
 * @returns {string} Root form text
 */
root(): string;

const nlp = require('nlp_compromise');

const doc = nlp.text('She walks quickly. He ran slowly.');

console.log(doc.text());   // 'She walks quickly. He ran slowly.'
console.log(doc.normal()); // 'she walks quickly he ran slowly'
console.log(doc.root());   // 'she walk quickly he run slowly'

/**
 * Get all terms from all sentences
 * @returns {Term[]} Array of all terms
 */
terms(): Term[];

/**
 * Get part-of-speech tags for all sentences
 * @returns {string[][]} Array of tag arrays for each sentence
 */
tags(): string[][];

const doc = nlp.text('Dogs bark loudly.');
const terms = doc.terms();
console.log(terms[0].text); // 'Dogs'
console.log(terms[0].tag);  // 'Plural'

const tags = doc.tags();
console.log(tags[0]); // ['Plural', 'Verb', 'Adverb']

/**
 * Find patterns across all sentences in the text
 * @param {string} pattern - Search pattern (supports grammatical patterns)
 * @param {object} [options] - Matching options
 * @returns {Result[]} Array of matching results
 */
match(pattern, options): Result[];

/**
 * Replace patterns across all sentences
 * @param {string} pattern - Pattern to replace
 * @param {string} replacement - Replacement text
 * @param {object} [options] - Replacement options
 * @returns {Text} Modified text object
 */
replace(pattern, replacement, options): Text;

const doc = nlp.text('The big dog barked. A small cat meowed.');

// Find adjective-noun patterns
const matches = doc.match('[Adjective] [Noun]');
console.log(matches[0].text()); // 'big dog'
console.log(matches[1].text()); // 'small cat'

// Replace patterns
doc.replace('[Adjective] dog', 'tiny puppy');
console.log(doc.text()); // 'The tiny puppy barked. A small cat meowed.'

/**
 * Convert all verbs to past tense
 * @returns {Text} Text with past tense verbs
 */
to_past(): Text;

/**
 * Convert all verbs to present tense
 * @returns {Text} Text with present tense verbs
 */
to_present(): Text;

/**
 * Convert all verbs to future tense
 * @returns {Text} Text with future tense verbs
 */
to_future(): Text;

/**
 * Negate all sentences in the text
 * @returns {Text} Text with negated sentences
 */
negate(): Text;

const doc = nlp.text('She walks to school. He runs home.');

console.log(doc.to_past().text());    // 'She walked to school. He ran home.'
console.log(doc.to_future().text());  // 'She will walk to school. He will run home.'
console.log(doc.negate().text());     // 'She doesn\'t walk to school. He doesn\'t run home.'

/**
 * Extract all person entities from text
 * @returns {Person[]} Array of Person objects
 */
people(): Person[];

/**
 * Extract all place entities from text
 * @returns {Place[]} Array of Place objects
 */
places(): Place[];

/**
 * Extract all organization entities from text
 * @returns {Organization[]} Array of Organization objects
 */
organizations(): Organization[];

/**
 * Extract all date entities from text
 * @returns {Date[]} Array of Date objects
 */
dates(): Date[];

/**
 * Extract all numeric values and measurements
 * @returns {Value[]} Array of Value objects
 */
values(): Value[];

/**
 * Extract topics and keywords with frequency counts
 * @returns {Topic[]} Array of Topic objects with counts
 */
topics(): Topic[];

const doc = nlp.text(`
  Tony Hawk visited Apple Inc. in Cupertino on March 15th.
  He spent $500 on new skateboard equipment.
`);

const people = doc.people();
console.log(people[0].text);        // 'Tony Hawk'
console.log(people[0].firstName);   // 'Tony'

const places = doc.places();
console.log(places[0].text);        // 'Cupertino'

const orgs = doc.organizations();
console.log(orgs[0].text);          // 'Apple Inc.'

const dates = doc.dates();  
console.log(dates[0].text);         // 'March 15th'

const values = doc.values();
console.log(values[0].number);      // 500
console.log(values[0].unit);        // 'dollar'

/**
 * Extract all nouns from text
 * @returns {Noun[]} Array of Noun objects
 */
nouns(): Noun[];

/**
 * Extract all adjectives from text  
 * @returns {Adjective[]} Array of Adjective objects
 */
adjectives(): Adjective[];

/**
 * Extract all verbs from text
 * @returns {Verb[]} Array of Verb objects
 */
verbs(): Verb[];

/**
 * Extract all adverbs from text
 * @returns {Adverb[]} Array of Adverb objects
 */
adverbs(): Adverb[];

const doc = nlp.text('The quick brown fox jumps quickly over the lazy dog.');

const nouns = doc.nouns();
console.log(nouns.map(n => n.text)); // ['fox', 'dog']

const adjectives = doc.adjectives();
console.log(adjectives.map(a => a.text)); // ['quick', 'brown', 'lazy']

const verbs = doc.verbs();
console.log(verbs.map(v => v.text)); // ['jumps']

const adverbs = doc.adverbs();
console.log(adverbs.map(a => a.text)); // ['quickly']

/**
 * Contractions object for expansion/contraction operations
 */
interface Contractions {
  /**
   * Expand all contractions in text (he'd → he would)
   * @returns {Text} Text with expanded contractions
   */
  expand(): Text;
  
  /**
   * Contract appropriate phrases (he would → he'd)
   * @returns {Text} Text with contracted phrases
   */
  contract(): Text;
}

const doc = nlp.text("He'd like to go. She's very smart.");

console.log(doc.contractions.expand().text());
// "He would like to go. She is very smart."

const expanded = nlp.text("He would like to go. She is very smart.");
console.log(expanded.contractions.contract().text());
// "He'd like to go. She's very smart."