JavaScript实现英文单词单数化的精准解决方案技术问询

Great question! Singularization is way trickier than pluralization in English—those irregular edge cases always trip up regex-based solutions. Let's walk through your options to get better accuracy, based on your needs.

Your current function works for simple suffix-based plurals, but it fails because it doesn’t account for three key issues:

• Irregular internal changes (like analysis → analyses, where the root changes instead of just the suffix)
• Invariant nouns (words like series that stay the same in singular and plural)
• Suffix priority (longer, more specific suffixes like -ves need to be checked before -s to avoid mistakes like turning knives into knive)

Here’s a refined version that fixes these gaps:

function singularize(word) {
  // First handle common irregular plural-to-singular pairs
  const irregularPairs = new Map([
    ['analyses', 'analysis'],
    ['phenomena', 'phenomenon'],
    ['criteria', 'criterion'],
    ['media', 'medium'],
    ['data', 'datum'], // Note: "data" is often used as plural in casual speech, but its technical singular is "datum"
    ['children', 'child'],
    ['teeth', 'tooth'],
    ['feet', 'foot']
  ]);
  if (irregularPairs.has(word)) {
    return irregularPairs.get(word);
  }

  // Handle invariant nouns (no change between singular/plural)
  const invariantNouns = new Set([
    'series', 'species', 'deer', 'sheep', 'fish'
  ]);
  if (invariantNouns.has(word)) {
    return word;
  }

  // Suffix rules ordered by specificity (longer/more unique first)
  const suffixRules = [
    [/([^f])ves$/, '$1fe'], // e.g., knives → knife
    [/([^aeiou]y)ies$/, '$1y'], // e.g., babies → baby
    [/([aeiou]y)s$/, '$1y'], // e.g., boys → boy
    [/i$/, 'us'], // e.g., cacti → cactus
    [/zes$/, 'z'], // e.g., buzzes → buzz
    [/ses$/, 's'], // e.g., buses → bus
    [/es$/, ''], // e.g., boxes → box
    [/s$/, ''] // e.g., cats → cat
  ];

  // Test each rule in order and return the first match
  for (const [regex, replacement] of suffixRules) {
    if (regex.test(word)) {
      return word.replace(regex, replacement);
    }
  }

  // Fallback: if no rules apply, return the original word
  return word;
}

This will handle way more cases than your original function, though you’ll still need to add more irregular pairs if you run into edge cases like geese → goose or men → man.

Unfortunately, there’s no standard browser API that lets you directly access a built-in dictionary for singular/plural conversion. Browsers do have spell-checking dictionaries, but these are not exposed via JavaScript for lexical tasks like this.

The closest thing is Intl.PluralRules, but that’s for determining when to use a plural form based on a number (e.g., "1 cat" vs "2 cats")—it can’t convert an existing plural word back to its singular form.

If you need better accuracy than rule-based methods but don’t want to embed a huge full dictionary, a curated list of high-frequency irregular pairs is your best bet. You can expand the irregularPairs map from the first section to include more common words.

For even broader coverage, you can use open-source lists of English irregular plurals (many available as small JSON files) and load only the most frequently used entries. This balances accuracy and file size, avoiding the bloat of a complete dictionary.

For example, here’s how you might expand the irregular map to cover more cases:

const expandedIrregularPairs = new Map([
  ['analyses', 'analysis'],
  ['phenomena', 'phenomenon'],
  ['criteria', 'criterion'],
  ['media', 'medium'],
  ['data', 'datum'],
  ['children', 'child'],
  ['teeth', 'tooth'],
  ['feet', 'foot'],
  ['geese', 'goose'],
  ['mice', 'mouse'],
  ['men', 'man'],
  ['women', 'woman'],
  ['oxen', 'ox'],
  ['people', 'person']
]);

内容的提问来源于stack exchange，提问作者Julius