结巴分词使用

# -*- coding:utf-8 -*-
import re #正则
import jieba
from wordcloud import WordCloud
import jieba.analyse
from collections import Counter
import matplotlib.pyplot as plt #图片包
import pandas as pd
#加载自定义词典
jieba.load_userdict("E:/Z/Z&R语言/京东评论/zidingyi.txt")
#加载停用词词典
stopwords = {}.fromkeys([ line.rstrip() for line in open('E:/Z/Z&R语言/京东评论/tingyong.txt',encoding="utf-8") ])
with open("E:/Z/Z&R语言/京东评论/jdphonepinglun-utf8.txt",encoding="utf-8") as f:
    str = f.read()
    shangpingid = re.compile(r"(?<=\{\"_id\":\").*?(?=\",)", re.S)
    pinglun = re.compile(r"(?<=:\[).*?(?=\])", re.S)
    idlist = re.findall(shangpingid, str)
    pinglunlist = re.findall(pinglun, str)
    # print("id数量",len(idlist))
    # print("评论数量",len(pinglunlist))
'''关键词和相应的权重
    # 第一个参数：待提取关键词的文本
    # 第二个参数：返回关键词的数量，重要性从高到低排序
    # 第三个参数：是否同时返回每个关键词的权重
    # 第四个参数：词性过滤，为空表示不过滤，若提供则仅返回符合词性要求的关键词
    keywords = jieba.analyse.extract_tags(pinglunlist[1], topK=20, withWeight=True, allowPOS=())
    # 访问提取结果
    for item in keywords:
        # 分别为关键词和相应的权重
        print(item[0], item[1])
'''
'''分词
        seg_list = jieba.cut(j, cut_all=True)
        print("Full Mode: " + "/ ".join(seg_list))  # 全模式
        seg_list = jieba.cut(j, cut_all=False)
        print(",".join(seg_list))  # 精确模式 默认是精确模式
        seg_list = jieba.cut_for_search(j)  # 搜索引擎模式
        print(",".join(seg_list))
'''
#标签云绘制
def Cloud(cut,name=""):

    counts = []
    for word in cut:
        if len(word) == 1 or word in stopwords:
            continue
        else:
            counts.append(word)
    c = Counter(counts)
    #写csv文件
    words = []
    pingshu = []
    for item in c.most_common(2000):
        words.append(item[0])
        pingshu.append(item[1])
    dataframe = pd.DataFrame({'词': words, '出现次数': pingshu})
    dataframe.to_csv(name+"的词频数.csv", index=False, sep=',', encoding="utf_8_sig")
    #绘制标签云
    wordCloud = WordCloud(
        font_path='data/simhei.ttf',  # 设置字体为中文字体
        background_color='black',  # 设置词云背景颜色
    )
    wordCloud.fit_words(c)  # 配置词云
    wordCloud.to_file(name+"标签云.jpg")
    print(name)
    # plt.imshow(wordCloud)
    # plt.axis("off")
    # plt.show()

#总评论
cutzong = jieba.lcut(str, cut_all=False)  # jieba.lcut 可以直接输出列表
Cloud(cutzong,"总体")
#各商品id
i=0
try:
    for j in pinglunlist:
        cutData = jieba.lcut(j, cut_all=False)  # jieba.lcut 可以直接输出列表
        if len(cutData) >= 20000:
            Cloud(cutData,idlist[i])
            print(" ".join(cutData))
            i=i+1
        else:
            i=i+1
except:
    pass