1、数据是使用scrapy-redis爬取的，存放在redis里面，爬取的是最近大热电影《海王》

2、使用了jieba中文分词解析库

3、使用了停用词stopwords，过滤掉一些无意义的词

4、使用matplotlib+wordcloud绘图展示

from redis import Redis

import json

import jieba

from wordcloud import WordCloud

import matplotlib.pyplot as plt

# 加载停用词

# stopwords = set(map(lambda x: x.rstrip('\n'), open('chineseStopWords.txt').readlines()))

stopwords = set()

with open('chineseStopWords.txt') as f:

    for line in f.readlines():

        stopwords.add(line.rstrip('\n'))

    stopwords.add(' ')

    # print(stopwords)

    # print(len(stopwords))

# 读取影评

db = Redis(host='localhost')

items = db.lrange('review:items', 0, -1)

# print(items)

# print(len(items))

# 统计每个word出现的次数

# 过滤掉停用词

# 记录总数，用于计算词频

words = {}

total = 0

for item in items:

    data = json.loads(item)['review']

    # print(data)

    # print('------------')

    for word in jieba.cut(data):

        if word not in stopwords:

            words[word] = words.get(word, 0) + 1

            total += 1

print(sorted(words.items(), key=lambda x: x[1], reverse=True))

# print(len(words))

# print(total)

# 词频

freq = {k: v / total for k, v in words.items()}

print(sorted(freq.items(), key=lambda x: x[1], reverse=True))

# 词云

wordcloud = WordCloud(font_path='simhei.ttf',

                      width=500,

                      height=300,

                      scale=10,

                      max_words=200,

                      max_font_size=40).fit_words(frequencies=freq)  # Create a word_cloud from words and frequencies

plt.imshow(wordcloud, interpolation="bilinear")

plt.axis('off')

plt.show()

绘图结果：

参考：

https://github.com/amueller/word_cloud

http://amueller.github.io/word_cloud/