Text Analytics ต้องนับความถี่คำ บทความนี้เสนอ 3 วิธี
1. collections.Counter (แนะนำ)
from collections import Counter
text = "Learn and practice and learn to practice"
freq = Counter(text.split()).most_common()
print(freq)
# [('and', 2), ('practice', 2), ('Learn', 1), ('learn', 1), ('to', 1)]
2. NLTK FreqDist
from nltk import FreqDist
text = "Learn and practice and learn to practice"
words = text.split()
fdist = FreqDist(words)
print(fdist.most_common())
# [('and', 2), ('practice', 2), ('Learn', 1), ('learn', 1), ('to', 1)]
3. Dictionary + count()
text = "Learn and practice and learn to practice"
words = text.split()
wfreq = [words.count(w) for w in words]
print(dict(zip(words, wfreq)))
# {'Learn': 1, 'and': 2, 'practice': 2, 'learn': 1, 'to': 1}
เปรียบเทียบ
| วิธี | Case Sensitive | Performance | Dependencies |
|---|---|---|---|
| Counter | Yes | O(n) | Built-in |
| FreqDist | Yes | O(n) | NLTK |
| Dict + count | Yes | O(n²) | Built-in |
แนะนำ: ใช้ Counter สำหรับ General Purpose, ใช้ FreqDist สำหรับ NLP Pipeline
Tip: ใช้ text.lower().split() ถ้าต้องการ Case-insensitive
