使用jieba库进行分词

安装jieba就不说了,自行百度!

import jieba

将标题分词,并转为list

seg_list = list(jieba.cut(result.get("title"), cut_all=False))

所有标题使用空格连接,方便后面做自然语言处理

para = para + " ".join(seg_list)

将分词后的标题(使用空格分割的标题)放到一个list里面

summaryList.insert(0," ".join(seg_list))

统计词频

from nltk.tokenize import WordPunctTokenizer
import nltk tokenizer = WordPunctTokenizer()
#统计词频
sentences = tokenizer.tokenize(para)#此处将para转为list(16进制字符)
wordFreq=nltk.FreqDist(sentences)
for i in wordFreq:print i,wordFreq[i]

转化为词袋,这一步的输入是一系列的句子(词与词之间使用空格分开),构成的列表。得到的结果是句子中关键词构成的一个列表,称为词袋

#转换为词袋
vectorizer = CountVectorizer(min_df=1,max_df=50)
#summaryList 是一个列表,每一个元素是一个句子 词与词之间使用空格分开,默认不会处理单个词(即一个汉字的就会忽略)
#可以通过修改vectorizer的正则表达式,解决不处理单个字的问题
vectorizer.token_pattern='(?u)\\b\\w+\\b'
X = vectorizer.fit_transform(summaryList) print X.shape
nums,features=X.shape #帖子数量和词袋中的词数,通过X.getrow(i) 获取每个帖子对应的向量 print vectorizer print str(vectorizer.get_feature_names()).decode("unicode-escape")

一个计算欧式距离的函数

#计算欧式距离
def dist_raw(v1,v2):
delta=v1-v2
return sp.linalg.norm(delta.toarray())

计算新帖的向量

#测试
new_para='我要吃苹果不吃香蕉'
new_para_list=" ".join(list(jieba.cut(new_para, cut_all=False)))
new_vec=vectorizer.transform([new_para_list])#new_para_list 是一个句子,词之间使用空格分开
print 'new_vec:',new_vec

计算新帖字与原帖子的距离

for i in range(0,nums):
para = paras[i]
para_vec=X.getrow(i)
d=dist_raw(new_vec,para_vec)
print para," = ",d

所有代码:

#!/usr/bin/python
# -*- coding: utf-8 -*-
print 'test OK'
import sys
from nltk.tokenize import WordPunctTokenizer
import nltk
import jieba
from sklearn.feature_extraction.text import CountVectorizer
import scipy as sp reload(sys)
sys.setdefaultencoding("utf-8") tokenizer = WordPunctTokenizer()
summaryList = [];
file=open("./para.txt")
paras=file.readlines()
words=""
for para in paras:
print para
seg_list = list(jieba.cut(para, cut_all=False))
words +=" ".join(seg_list)
summaryList.insert(0," ".join(seg_list))
#para='I like eat apple because apple is red but because I love fruit'
#统计词频
sentences = tokenizer.tokenize(words)#此处将para转为list
#print sentences
wordFreq=nltk.FreqDist(sentences)
print str(wordFreq.keys()).decode("unicode-escape")
#print dir(wordFreq)
for i in wordFreq:
print i,wordFreq[i] print str(summaryList).decode("unicode-escape")
#转换为词袋
vectorizer = CountVectorizer(min_df=1,max_df=50)
#summaryList 是一个列表,每一个元素是一个句子 词与词之间使用空格分开,默认不会处理单个词(即一个汉字的就会忽略)
#可以通过修改vectorizer的正则表达式,解决不处理单个字的问题
vectorizer.token_pattern='(?u)\\b\\w+\\b'
X = vectorizer.fit_transform(summaryList)
print str(vectorizer.get_feature_names()).decode("unicode-escape")
print X.shape
nums,features=X.shape #帖子数量和词袋中的词数 #计算欧式距离
def dist_raw(v1,v2):
delta=v1-v2
return sp.linalg.norm(delta.toarray()) #测试
new_para='我要吃苹果不吃香蕉'
new_para_list=" ".join(list(jieba.cut(new_para, cut_all=False)))
new_vec=vectorizer.transform([new_para_list])#new_para_list 是一个句子,词之间使用空格分开
print 'new_vec:',new_vec for i in range(0,nums):
para = paras[i]
para_vec=X.getrow(i)
d=dist_raw(new_vec,para_vec)
print para," = ",d

版本二:

 #!/usr/bin/python
# -*- coding: utf-8 -*-
print 'test OK'
import sys
from nltk.tokenize import WordPunctTokenizer
import nltk
import jieba
from sklearn.feature_extraction.text import CountVectorizer
import scipy as sp reload(sys)
sys.setdefaultencoding("utf-8") tokenizer = WordPunctTokenizer()
summaryList = [];
file=open("./para.txt")
paras=file.readlines()
words=""
for para in paras:
print para
seg_list = list(jieba.cut(para, cut_all=False))
words +=" ".join(seg_list)
summaryList.insert(0," ".join(seg_list))
#para='I like eat apple because apple is red but because I love fruit'
#统计词频
sentences = tokenizer.tokenize(words)#此处将para转为list
#print sentences
wordFreq=nltk.FreqDist(sentences)
print str(wordFreq.keys()).decode("unicode-escape")
#print dir(wordFreq) print str(summaryList).decode("unicode-escape")
#转换为词袋
vectorizer = CountVectorizer(min_df=0,max_df=20)
#summaryList 是一个列表,每一个元素是一个句子 词与词之间使用空格分开,默认不会处理单个词(即一个汉字的就会忽略)
#可以通过修改vectorizer的正则表达式,解决不处理单个字的问题
#vectorizer.token_pattern='(?u)\\b\\w+\\b'
X = vectorizer.fit_transform(summaryList)
print str(vectorizer.get_feature_names()).decode("unicode-escape")
print X.shape
nums,features=X.shape #帖子数量和词袋中的词数 #计算欧式距离
def dist_raw(v1,v2):
delta=v1-v2
return sp.linalg.norm(delta.toarray()) #测试
new_para='夏季新款清新碎花雪纺连衣裙,收腰显瘦设计;小V领、小碎花、荷叶袖、荷叶边的结合使得这款连衣裙更显精致,清新且显气质。'
new_para_list=" ".join(list(jieba.cut(new_para, cut_all=False)))
new_vec=vectorizer.transform([new_para_list])#new_para_list 是一个句子,词之间使用空格分开
#print 'new_vec:',new_vec.toarray() minDis = 9999
title=""
for i in range(0,nums):
para = summaryList[i]
para_vec=X.getrow(i)
d=dist_raw(new_vec,para_vec)
#print X.getrow(i).toarray(),' = ',d
if(minDis > d):
minDis = d
title = para
print title," = ",d
print new_para_list
print title

运行结果:

aaarticlea/png;base64,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" alt="" />

python机器学习——分词的更多相关文章

  1. 常用python机器学习库总结

    开始学习Python,之后渐渐成为我学习工作中的第一辅助脚本语言,虽然开发语言是Java,但平时的很多文本数据处理任务都交给了Python.这些年来,接触和使用了很多Python工具包,特别是在文本处 ...

  2. [Python] 机器学习库资料汇总

    声明:以下内容转载自平行宇宙. Python在科学计算领域,有两个重要的扩展模块:Numpy和Scipy.其中Numpy是一个用python实现的科学计算包.包括: 一个强大的N维数组对象Array: ...

  3. [resource]Python机器学习库

    reference: http://qxde01.blog.163.com/blog/static/67335744201368101922991/ Python在科学计算领域,有两个重要的扩展模块: ...

  4. [转]Python机器学习工具箱

    原文在这里  Python在科学计算领域,有两个重要的扩展模块:Numpy和Scipy.其中Numpy是一个用python实现的科学计算包.包括: 一个强大的N维数组对象Array: 比较成熟的(广播 ...

  5. 【转】常见的python机器学习工具包比较

    http://algosolo.com/ 分析对比了常见的python机器学习工具包,包括: scikit-learn mlpy Modular toolkit for Data Processing ...

  6. python机器学习《回归 一》

    唠嗑唠嗑 依旧是每一次随便讲两句生活小事.表示最近有点懒,可能是快要考试的原因,外加这两天都有笔试和各种面试,让心情变得没那么安静的敲代码,没那么安静的学习算法.搞得第一次和技术总监聊天的时候都不太懂 ...

  7. 2016年GitHub排名前20的Python机器学习开源项目(转)

    当今时代,开源是创新和技术快速发展的核心.本文来自 KDnuggets 的年度盘点,介绍了 2016 年排名前 20 的 Python 机器学习开源项目,在介绍的同时也会做一些有趣的分析以及谈一谈它们 ...

  8. Python 结巴分词

    今天的任务是对txt文本进行分词,有幸了解到"结巴"中文分词,其愿景是做最好的Python中文分词组件.有兴趣的朋友请点这里. jieba支持三种分词模式: *精确模式,试图将句子 ...

  9. Python机器学习包

    常用Python机器学习包 Numpy:用于科学计算的包 Pandas:提供高性能,易于使用的数据结构和数据分析工具 Scipy:用于数学,科学工程的软件 StatsModels:用于探索数据.估计统 ...

随机推荐

  1. [Unity热更新]tolua# & LuaFramework(一):基础

    一.tolua# c#调用lua:LuaState[变量名/函数名] 1.LuaState a.执行lua代码段 DoString(string) DoFile(.lua文件名) Require(.l ...

  2. Ubuntu 14.04 Vim编辑文件的一般操作

    vim编辑文件的一般操作 1. vim #在命令行中输入vim,进入vim编辑器 2. i #按一下i键,下端显示 --INSERT-- #插入命令,在vim中可能任意字符都有作用 3. Esc #退 ...

  3. Android无线测试之—UiAutomator UiSelector API介绍之七

    对象搜索—索引与实例 一.索引与实例说明: 1)index:在同一级中的编号,在兄弟类中组件的编号,index从0开始 2)instance:同一个布局中同一类组件的编号,instance从0开始 二 ...

  4. 【BZOJ4810】[Ynoi2017]由乃的玉米田 bitset+莫队

    [BZOJ4810][Ynoi2017]由乃的玉米田 Description 由乃在自己的农田边散步,她突然发现田里的一排玉米非常的不美.这排玉米一共有N株,它们的高度参差不齐.由乃认为玉米田不美,所 ...

  5. Coursera课程《Machine Learning》学习笔记(week2)

    1 特征 1-1 什么是特征? 我的理解就是,用于描述某个样本点,以哪几个指标来评定,这些个指标就是特征.比方说对于一只鸟,我们评定的指标就可以是:(a)鸟的翅膀大还是小?(b)鸟喙长还是短?(c)鸟 ...

  6. SQLServer中exists和except用法

    一.exists 1.1 说明 EXISTS(包括 NOT EXISTS)子句的返回值是一个BOOL值.EXISTS内部有一个子查询语句(SELECT ... FROM...),我将其称为EXIST的 ...

  7. JavaScript遍历IP段内所有IP

    思路:将两个IP转换为数字进行比较,小的那个慢慢加一,直到变成大的那个IP所转换的数字,将这其中的数字再转换为IP地址即为IP段内所有的IP. //IP转数字 function ip2int(ip) ...

  8. mysql练手

    1.根据图创建下列表格 没有外键的表先创建,创建顺序为teacher,class,course,student CREATE TABLE class ( cid INT NOT NULL auto_i ...

  9. Vue.js刷新当前页面

    Vue.js的路由跳转很强大,类似ajax的局部刷新,路由跳转时候页面是不刷新的,刷新当前页面的功能不建议用,但是有的时候确实需要刷新当前页面来实现某些需求,这个时候,我们有三 种方法可以实现. 第一 ...

  10. B-Tree vs LSM-tree

    什么是B-树 一.已排序文件的查找时间 对一个有N笔记录的已排序表进行二叉查找,可以在O(log2N)比较级完成.如果表有1,000,000笔记录,那么定位其中一笔记录,将在20 ( log21,00 ...