ngrams Method

The ngrams() method generates sequences of n consecutive words from text strings in the collection. This method is useful for analyzing text patterns, phrase frequencies, and language processing tasks.

Basic Syntax

ngrams(this: CollectionOperations<string>, n: number): CollectionOperations<string>

Examples

Basic Usage

import { collect } from 'ts-collect'

const text = collect([
  "The quick brown fox jumps"
])

const bigrams = text.ngrams(2)
console.log(bigrams.all())
// [
//   "The quick",
//   "quick brown",
//   "brown fox",
//   "fox jumps"
// ]

const trigrams = text.ngrams(3)
console.log(trigrams.all())
// [
//   "The quick brown",
//   "quick brown fox",
//   "brown fox jumps"
// ]

Pattern Analysis

interface PhraseFrequency {
  phrase: string
  count: number
}

const reviews = collect([
  "very good product quality",
  "good product durability",
  "product quality excellent"
])

const commonPhrases = reviews
  .ngrams(2)
  .frequency()
  .entries()
  .map(([phrase, count]) => ({ phrase, count }))
  .sortByDesc('count')

Real-world Example: E-commerce Review Analysis

class ProductReviewAnalyzer {
  private reviews: CollectionOperations<string>

  constructor(reviews: string[]) {
    this.reviews = collect(reviews)
  }

  findCommonPhrases() {
    return this.reviews
      .ngrams(2)
      .frequency()
      .entries()
      .filter(([_, count]) => count > 1)
      .map(([phrase, count]) => ({
        phrase,
        count,
        percentage: (count / this.reviews.count()) * 100
      }))
  }

  getKeyPhrases(minOccurrences: number = 2) {
    const bigramFrequency = this.reviews.ngrams(2).frequency()
    const trigramFrequency = this.reviews.ngrams(3).frequency()

    return {
      topBigrams: Array.from(bigramFrequency.entries())
        .filter(([_, count]) => count >= minOccurrences)
        .sort(([, a], [, b]) => b - a),
      topTrigrams: Array.from(trigramFrequency.entries())
        .filter(([_, count]) => count >= minOccurrences)
        .sort(([, a], [, b]) => b - a)
    }
  }
}

// Usage example
const analyzer = new ProductReviewAnalyzer([
  "Great product quality and durability",
  "Product quality is excellent",
  "Excellent customer service response",
  "Quick service response time"
])

console.log(analyzer.getKeyPhrases())

Return Value

• Returns a new collection containing n-gram strings
• Maintains original word order
• Preserves case of original text
• Empty input returns empty collection
• Handles multiple input strings
• Returns shorter n-grams for inputs smaller than n

Common Use Cases

1. Text Analysis

   • Finding common phrases
   • Analyzing writing patterns
   • Identifying repeated content
   • Language pattern detection

2. Content Optimization

   • SEO keyword analysis
   • Content uniqueness checking
   • Plagiarism detection
   • Phrase extraction

3. Customer Feedback Analysis

   • Review pattern analysis
   • Common complaint detection
   • Feature request identification
   • Sentiment pattern analysis

4. Quality Monitoring

   • Support response analysis
   • Documentation consistency
   • Content standardization
   • Pattern recognition

5. Search Enhancement

   • Query suggestion generation
   • Search term analysis
   • Autocomplete improvements
   • Related phrase detection