详解Python 字符串相似性的几种度量方法 - python基础

易学Python (Anthony Briggs著) 中文pdf扫描版[40MB]

×Ö·û´®µÄÏàËÆÐÔ±È½ÏÓ¦ÓÃ³¡ºÏºÜ¶à£¬ÏñÆ´Ð´¾À´í¡¢ÎÄ±¾È¥ÖØ¡¢ÉÏÏÂÎÄÏàËÆÐÔµÈ¡£

ÆÀ¼Û×Ö·û´®ÏàËÆ¶È×î³£¼ûµÄ°ì·¨¾ÍÊÇ£º°ÑÒ»¸ö×Ö·û´®Í¨¹ý²åÈë¡¢É¾³ý»òÌæ»»ÕâÑùµÄ±à¼²Ù×÷£¬±ä³ÉÁíÍâÒ»¸ö×Ö·û´®£¬ËùÐèÒªµÄ×îÉÙ±à¼´ÎÊý£¬ÕâÖÖ¾ÍÊÇ±à¼¾àÀë£¨edit distance£©¶ÈÁ¿·½·¨£¬Ò²³ÆÎªLevenshtein¾àÀë¡£º£Ã÷¾àÀëÊÇ±à¼¾àÀëµÄÒ»ÖÖÌØÊâÇé¿ö£¬Ö»¼ÆËãµÈ³¤Çé¿öÏÂÌæ»»²Ù×÷µÄ±à¼´ÎÊý£¬Ö»ÄÜÓ¦ÓÃÓÚÁ½¸öµÈ³¤×Ö·û´®¼äµÄ¾àÀë¶ÈÁ¿¡£

ÆäËû³£ÓÃµÄ¶ÈÁ¿·½·¨»¹ÓÐ Jaccard distance¡¢J-W¾àÀë£¨Jaro¨CWinkler distance£©¡¢ÓàÏÒÏàËÆÐÔ£¨cosine similarity£©¡¢Å·ÊÏ¾àÀë£¨Euclidean distance£©µÈ¡£

python-Levenshtein Ê¹ÓÃ

Ê¹ÓÃ pip install python-Levenshtein Ö¸Áî°²×° Levenshtein

# -*- coding: utf-8 -*-
 
import difflib
# import jieba
import Levenshtein
 
str1 = "ÎÒµÄ¹Ç÷ÀÑ©°× Ò²³¤²»³öÇàïý"
str2 = "Ñ©µÄÈÕ×Ó ÎÒÖ»Ïëµ½Ñ©ÖÐÈ¥si"
 
# 1. difflib
seq = difflib.SequenceMatcher(None, str1,str2)
ratio = seq.ratio()
print 'difflib similarity1: ', ratio
 
# difflib È¥µôÁÐ±íÖÐ²»ÐèÒª±È½ÏµÄ×Ö·û
seq = difflib.SequenceMatcher(lambda x: x in ' ÎÒµÄÑ©', str1,str2)
ratio = seq.ratio()
print 'difflib similarity2: ', ratio
 
# 2. hamming¾àÀë£¬str1ºÍstr2³¤¶È±ØÐëÒ»ÖÂ£¬ÃèÊöÁ½¸öµÈ³¤×Ö´®Ö®¼ä¶ÔÓ¦Î»ÖÃÉÏ²»Í¬×Ö·ûµÄ¸öÊý
# sim = Levenshtein.hamming(str1, str2)
# print 'hamming similarity: ', sim
 
# 3. ±à¼¾àÀë£¬ÃèÊöÓÉÒ»¸ö×Ö´®×ª»¯³ÉÁíÒ»¸ö×Ö´®×îÉÙµÄ²Ù×÷´ÎÊý£¬ÔÚÆäÖÐµÄ²Ù×÷°üÀ¨ ²åÈë¡¢É¾³ý¡¢Ìæ»»
sim = Levenshtein.distance(str1, str2)
print 'Levenshtein similarity: ', sim
 
# 4.¼ÆËãÀ³ÎÄË¹Ì¹±È
sim = Levenshtein.ratio(str1, str2)
print 'Levenshtein.ratio similarity: ', sim
 
# 5.¼ÆËãjaro¾àÀë
sim = Levenshtein.jaro(str1, str2 )
print 'Levenshtein.jaro similarity: ', sim
 
# 6. Jaro¨CWinkler¾àÀë
sim = Levenshtein.jaro_winkler(str1 , str2 )
print 'Levenshtein.jaro_winkler similarity: ', sim

Êä³ö£º

difflib similarity1: 0.246575342466
difflib similarity2: 0.0821917808219
Levenshtein similarity: 33
Levenshtein.ratio similarity: 0.27397260274
Levenshtein.jaro similarity: 0.490208958959
Levenshtein.jaro_winkler similarity: 0.490208958959

转载请注明：谷谷点程序 » 详解Python 字符串相似性的几种度量方法