import newspaper
# Word frequency database
import collections
# numpy library
import numpy as np
# Stuttering participle
import jieba
# Word cloud display library
import wordcloud
# Image processing library
from PIL import Image
# Image gallery
import matplotlib.pyplot as plt
# Access to the article
article = newspaper.Article('https://news.sina.com.cn/o/2019-11-28/doc-iihnzahi3991780.shtml')
# Download article
article.download()
# Analyze the article
article.parse()
# Make a study of the article nlp Handle
article.nlp()
# nlp Processed article splicing
article_words = "".join(article.keywords)
# Precise pattern segmentation ( The default mode )
seg_list_exact = jieba.cut(article_words, cut_all=False)
# Store word segmentation results
object_list = []
# Removed word
rm_words = [' To meet ', ' since ', ' take ']
# Iteration participle object
for word in seg_list_exact:
if word not in rm_words:
object_list.append(word)
# Word frequency statistics
word_counts = collections.Counter(object_list)
# Before acquisition 10 The most frequent words
word_top10 = word_counts.most_common(10)
# Entries and times
for w, c in word_top10:
print(w, c)
# Word frequency display
# Define the word frequency background
mask = np.array(Image.open('bg.jpg'))
wc = wordcloud.WordCloud(
# Set the font format
font_path='C:/Windows/Fonts/simhei.ttf',
# Background map
mask=mask,
# Set the maximum number of words displayed
max_words=100,
# Set font maximum
max_font_size=80
)
# Generating word clouds from dictionaries
wc.generate_from_frequencies(word_counts)
# Create a color scheme from the background image
image_colors = wordcloud.ImageColorGenerator(mask)
# Show word cloud
plt.imshow(wc)
# Turn off the axis
plt.axis('off')
plt.savefig('wc.jpg')
# Display images
plt.show()
