文本挖掘与NLP笔记——代码向：分词

分词：jieba.cut

words = jieba.cut("我来到北京大学",cut_all=True)
print('全模式：'+'/'.join([w for w in words])) #全模式

words = jieba.cut("我来到北京大学",cut_all=False)
print('精确模式：'+'/'.join([w for w in words])) #精确模式,默认

words = jieba.cut_for_search("小明毕业于北京大学，后在美国哈佛大学深造")
print('/'.join([w for w in words])) #搜索引擎模式,在精确模式的基础上，对长词在此划分

全模式：我/来到/北京/北京大学/大学

精确模式：我/来到/北京大学

请练习添加自定义词典

词性：jieba.posseg

import jieba.posseg as pg

for word, flag in pg.cut("你想去学校填写学生寒暑假住校申请表吗？"):
    print('%s %s' % (word, flag))

'你/学校/填写/学生/寒暑假/住校/申请表'

分词引入停用词

import jieba
import pandas as pd
import numpy as np

paths = '中英文停用词.xlsx'
dfs = pd.read_excel(paths,dtype=str)

stopwords = ['想','去','吗','？']

words = jieba.cut("你想去学校填写学生寒暑假住校申请表吗？")
'/'.join([w for w in words if (w not in stopwords)])#此处’/'表示换行

'你/学校/填写/学生/寒暑假/住校/申请表'

txt转dataframe函数

import random
import jieba.posseg as pg
import pandas as pd
import numpy as np

def generatorInfo(file_name):
    # 读取文本文件
    with open(file_name, encoding='utf-8') as file:
        line_list = [k.strip() for k in file.readlines()]
        data = []
        for k in random.sample(line_list,1000):
            t = k.split(maxsplit=1)
            #data_label_list.append(t[0])
            #data_content_list.append(t[1])
            data.append([t[0],' '.join([w for w,flag in pg.cut(t[1]) if (w not in dfs['stopwords']) and (w !=' ') and (len(w)>=2)])])
    return data

file_name = 'cnews.train.txt'
df = pd.DataFrame(np.array(generatorInfo(file_name)),columns=['类别','分词'])
path = '训练集分词结果(随机选取1000个样本).xlsx'
df.to_excel(path,index=False)
df

词云图：wordcloud

%pylab inline
import matplotlib.pyplot as plt
from wordcloud import WordCloud

text = ' '.join(list(df['分词']))
wcloud = WordCloud(
    font_path='simsun.ttc', #字体路径
    background_color='white', #指定背景颜色
    max_words=500,   #词云显示最大词数
    max_font_size=150,  #指定最大字号
    #mask = mask #背景图片
) 

wcloud = wcloud.generate(text)  #生成词云
plt.imshow(wcloud)
plt.axis('off')
plt.show()

提取关键词：jieba.analyse.extract_tags

import jieba.analyse
import pandas as pd
import numpy as np

path = '训练集分词结果(随机选取1000个样本).xlsx'
df = pd.read_excel(path,dtype=str)
s = ' '.join(list(df['分词']))
for w,x in jieba.analyse.extract_tags(s,withWeight=True):
    print('%s %s' % (w,x))

请练习基于TextRank算法抽取关键词

import jieba.analyse
import pandas as pd
import numpy as np

path = '训练集分词结果(随机选取1000个样本).xlsx'
df = pd.read_excel(path,dtype=str)
tag = list(set(list(df['类别'])))

for t in tag:
    s = ' '.join(list(df[df['类别']==t]['分词']))
    print(t)
    for w,x in jieba.analyse.extract_tags(s,withWeight=True):
        print('%s %s' % (x,w))

构建词向量

构建词向量简单的有两种分别是TfidfTransformer和 CountVectorizer

#CountVectorizer会将文本中的词语转换为词频矩阵
from sklearn.feature_extraction.text import CountVectorizer
path = '训练集分词结果(随机选取1000个样本).xlsx'
df = pd.read_excel(path,dtype=str)
corpus = df['分词']
#vectorizer = CountVectorizer(max_features=5000)
vectorizer = CountVectorizer()
X = vectorizer.fit_transform(corpus)
print(X)

from sklearn.feature_extraction.text import TfidfTransformer
import datetime

starttime = datetime.datetime.now()
transformer = TfidfTransformer()
tfidf = transformer.fit_transform(X)
word = vectorizer.get_feature_names()
weight = tfidf.toarray()
print(weight)

词语分类：人工vsKmeans

from sklearn.cluster import KMeans

starttime = datetime.datetime.now()

path = '训练集分词结果(随机选取1000个样本).xlsx'
df = pd.read_excel(path,dtype=str)
corpus = df['分词']

kmeans=KMeans(n_clusters=10)   #n_clusters:number of cluster
kmeans.fit(weight)
res = [list(df['类别']),list(kmeans.labels_)]
df_res = pd.DataFrame(np.array(res).T,columns=['人工分类','Kmeans分类'])
path_res = 'Kmeans自动分类结果.xlsx'
df_res.to_excel(path_res,index=False)
df_res

path = 'Kmeans自动分类结果.xlsx'
df = pd.read_excel(path,dtype=str)

df['计数'] = [1 for m in range(len(df['人工分类']))]
df1 = pd.pivot_table(df, index=['人工分类'], columns=['Kmeans分类'], values=['计数'], aggfunc=np.sum, fill_value=0)
co = ['人工分类']
co.extend(list(df1['计数'].columns))
df1 = df1.reset_index()
df2 = pd.DataFrame((np.array(df1)),columns=co)

path_res = '人工与Kmeans分类结果对照.xlsx'
df2.to_excel(path_res,index=False)

df2

import random

def is_contain_chinese(check_str):
    for ch in check_str:
        if u'\u4e00' <= ch <= u'\u9fff':
            return 1
    return 0

def generatorInfo(file_name):
    """
    batch_size：生成数据的batch size
    seq_length：输入文字序列长度
    num_classes：文本的类别数
    file_name：读取文件的路径
    """

    # 读取文本文件
    with open(file_name, encoding='utf-8') as file:
        line_list = [k.strip() for k in file.readlines()]
        #data_label_list = []   # 创建数据标签文件
        #data_content_list = []   # 创建数据文本文件
        data = []
        for k in random.sample(line_list,1000):
            t = k.split(maxsplit=1)
            #data_label_list.append(t[0])
            #data_content_list.append(t[1])
            data.append([t[0],' '.join([w for w,flag in jieba.posseg.cut(t[1]) if (w not in dfs['stopwords']) and (w !=' ') and (flag not in ["nr","ns","nt","nz","m","f","ul","l","r","t"]) and (len(w)>=2)])])

    return data

#导入中文停用词表
paths = '中英文停用词.xlsx'
dfs = pd.read_excel(paths,dtype=str)

file_name = 'cnews.train.txt'
df = pd.DataFrame(np.array(generatorInfo(file_name)),columns=['类别','分词'])
df

汇总

import random
import jieba
import pandas as pd
import numpy as np
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.cluster import KMeans
from sklearn.feature_extraction.text import TfidfTransformer

def is_contain_chinese(check_str):
    for ch in check_str:
        if u'\u4e00' <= ch <= u'\u9fff':
            return 1
    return 0

def generatorInfo(file_name):
    """
    batch_size：生成数据的batch size
    seq_length：输入文字序列长度
    num_classes：文本的类别数
    file_name：读取文件的路径
    """

    # 读取文本文件
    with open(file_name, encoding='utf-8') as file:
        line_list = [k.strip() for k in file.readlines()]
        #data_label_list = []   # 创建数据标签文件
        #data_content_list = []   # 创建数据文本文件
        data = []
        for k in random.sample(line_list,1000):
            t = k.split(maxsplit=1)
            #data_label_list.append(t[0])
            #data_content_list.append(t[1])
            data.append([t[0],' '.join([w for w,flag in jieba.posseg.cut(t[1]) if (w not in dfs['stopwords']) and (w !=' ') and (flag not in ["nr","ns","nt","nz","m","f","ul","l","r","t"]) and (len(w)>=2)])])

    return data

#导入中文停用词表
paths = '中英文停用词.xlsx'
dfs = pd.read_excel(paths,dtype=str)

file_name = 'cnews.train.txt'
df = pd.DataFrame(np.array(generatorInfo(file_name)),columns=['类别','分词'])

#统计词频
corpus = df['分词'] #语料中的单词以空格隔开
#vectorizer = CountVectorizer(max_features=5000)
vectorizer = CountVectorizer()
X = vectorizer.fit_transform(corpus)

#文本向量化
transformer = TfidfTransformer()
tfidf = transformer.fit_transform(X)
word = vectorizer.get_feature_names()
weight = tfidf.toarray()

kmeans=KMeans(n_clusters=10)   #n_clusters:number of cluster
kmeans.fit(weight)

res = [list(df['类别']),list(kmeans.labels_)]
df_res = pd.DataFrame(np.array(res).T,columns=['人工分类','Kmeans分类'])

df_res['计数'] = [1 for m in range(len(df_res['人工分类']))]
df1 = pd.pivot_table(df_res, index=['人工分类'], columns=['Kmeans分类'], values=['计数'], aggfunc=np.sum, fill_value=0)
co = ['人工分类']
co.extend(list(df1['计数'].columns))
df1 = df1.reset_index()
df2 = pd.DataFrame((np.array(df1)),columns=co)
df2

df['Kmeans分类'] = df_res['Kmeans分类']
df