python实现中文分词FMM算法实例

本文实例讲述了python实现中文分词FMM算法。分享给大家供大家参考。具体分析如下：

FMM算法的最简单思想是使用贪心算法向前找n个，如果这n个组成的词在词典中出现，就ok，如果没有出现，那么找n-1个...然后继续下去。假如n个词在词典中出现，那么从n+1位置继续找下去，直到句子结束。

import re  
def PreProcess(sentence,edcode="utf-8"):  
 sentence = sentence.decode(edcode)  
 sentence=re.sub(u"[。，,！……!《》<>\"':：？\?、\|“”‘'；]"," ",sentence)  
 return sentence  
def FMM(sentence,diction,result = [],maxwordLength = 4,edcode="utf-8"):
 i = 0
 sentence = PreProcess(sentence,edcode)  
 length = len(sentence)  
 while i < length:
   # find the ascii word  
   tempi=i  
   tok=sentence[i:i+1]  
   while re.search("[0-9A-Za-z\-\+#@_\.]{1}",tok)<>None:  
     i= i+1
     tok=sentence[i:i+1]  
   if i-tempi>0:  
     result.append(sentence[tempi:i].lower().encode(edcode))
   # find chinese word  
   left = len(sentence[i:])  
   if left == 1:  
     """go to 4 step over the FMM"""
     """should we add the last one? Yes, if not blank"""
     if sentence[i:] <> " ":  
       result.append(sentence[i:].encode(edcode))  
     return result  
   m = min(left,maxwordLength)  
   for j in xrange(m,0,-1):  
     leftword = sentence[i:j+i].encode(edcode)
    #  print leftword.decode(edcode)  
     if LookUp(leftword,diction):
       # find the left word in dictionary
       # it's the right one  
       i = j+i  
       result.append(leftword)  
       break
     elif j == 1:  
       """only one word, add into result, if not blank"""
       if leftword.decode(edcode) <> " ":  
         result.append(leftword)  
       i = i+1
     else:  
       continue
 return result  
def LookUp(word,dictionary):  
 if dictionary.has_key(word):  
   return True
 return False
def ConvertGBKtoUTF(sentence):  
 return sentence.decode('gbk').encode('utf-8')
dictions = {}  
dictions["ab"] = 1
dictions["cd"] = 2
dictions["abc"] = 1
dictions["ss"] = 1
dictions[ConvertGBKtoUTF("好的")] = 1
dictions[ConvertGBKtoUTF("真的")] = 1
sentence = "asdfa好的是这样吗vasdiw呀真的daf dasfiw asid是吗？"
s = FMM(ConvertGBKtoUTF(sentence),dictions)  
for i in s:  
 print i.decode("utf-8")
test = open("test.txt","r")  
for line in test:  
 s = FMM(CovertGBKtoUTF(line),dictions)  
 for i in s:  
   print i.decode("utf-8")

运行结果如下：

asdfa
好的
是
这
样
吗
vasdiw
呀
真的
daf
dasfiw
asid
是
吗
？

希望本文所述对大家的Python程序设计有所帮助。