改进后的基于web领域术语抽取

#!/usr/bin/env python
# -*- coding: gb2312 -*-
from gensim.models  import word2vec  #上面的编码变成 utf8,gb2312都可以
import gensim
import os


class MySentences(object):
    def __init__(self, dirname):
        self.dirname = dirname

    def __iter__(self):
        for fname in os.listdir(self.dirname):
            for line in open(os.path.join(self.dirname, fname)):
                yield line.split()

sentences = MySentences('D:\\手机-预处理后\\')#('D:\\手机-正文\\花卉\\')
model = gensim.models.Word2Vec(sentences,min_count=2,size=500)
model.save('D:\\model\\mymodelphone')
new_model = gensim.models.Word2Vec.load('D:\\model\\mymodelphone')

list=new_model.most_similar('小米',topn=20)
#print(list[1][0])
#for s in list:
 #   t=s[0]
    #print t.decode('utf8')  ##这样就好了。不需要encode
  #  print (t)

#list=new_model.most_similar('牡丹')
#print(list[1][0])

for s in list:
    t=s[0]
    print t,s[1]#.decode('utf8')  ##这样就好了。不需要encode
    print s[1]
                



    
    
'''
#读取当个文档，进行训练，进行测试

sentences = []
#sentences=word2vec.Text8Corpus("D:\\test.txt")

f = open("D:\\test.txt", "r")
for line in f.readlines():
    # print "create %s"%line
     sentences.append(line.split(" "))


f.close()

#print sentences
model = word2vec.Word2Vec(sentences,min_count=3,size=50)

# 如何打印出来 汉字,easy啊。直接打印list[0][0]

list=model.most_similar('普吉岛')
#print(list[1][0])
for s in list:
    t=s[0]
    print t.decode('utf8')  ##这样就好了。不需要encode


#s="蒙牛 -- 伊利 间的余弦距离："
#print s.decode('utf-8')
#print model.similarity('旅游','中国')
'''