/spell-correct.py

http://spell-correct-in-go.googlecode.com/ · Python · 36 lines · 27 code · 9 blank · 0 comment · 20 complexity · 5dfa56ceca4676533ad7a94d5161de51 MD5 · raw file

  1. from datetime import datetime
  2. import re, collections
  3. def words(text): return re.findall('[a-z]+', text.lower())
  4. def train(features):
  5. model = collections.defaultdict(lambda: 1)
  6. for f in features:
  7. model[f] += 1
  8. return model
  9. NWORDS = train(words(file('big.txt').read()))
  10. alphabet = 'abcdefghijklmnopqrstuvwxyz'
  11. def edits1(word):
  12. splits = [(word[:i], word[i:]) for i in range(len(word) + 1)]
  13. deletes = [a + b[1:] for a, b in splits if b]
  14. transposes = [a + b[1] + b[0] + b[2:] for a, b in splits if len(b)>1]
  15. replaces = [a + c + b[1:] for a, b in splits for c in alphabet if b]
  16. inserts = [a + c + b for a, b in splits for c in alphabet]
  17. return set(deletes + transposes + replaces + inserts)
  18. def known_edits2(word):
  19. return set(e2 for e1 in edits1(word) for e2 in edits1(e1) if e2 in NWORDS)
  20. def known(words): return set(w for w in words if w in NWORDS)
  21. def correct(word):
  22. candidates = known([word]) or known(edits1(word)) or known_edits2(word) or [word]
  23. return max(candidates, key=NWORDS.get)
  24. startTime = datetime.now()
  25. for i in range(100):
  26. correct('korrecter')
  27. print datetime.now() - startTime