import re 
Import the collections 
The import numpy as np 
The import jieba 
The from PIL import Image 
The import wordcloud 
The import matplotlib. Pylab as PLT 
# to read text 
With the open (" the romance of The Three Kingdoms. TXT ", "r") as fp: 
Words=fp. Read () 
# text preprocessing 
Pattern=re.com from running (u '| | \ t \ n \ | - | : |. (\ | | \ \)? | | "" |! | the | | ') # define the regular expression matching mode 
Words=re. Sub (pattern, ' ', words) 
# text participle 
Swords=jieba. The cut (words, cut_all=False) 
# block words 
Okwords=[] 
# second method nywords=[u ' 'u' is' u 'and' u 'is' u' in ', u 'a' u 'and' u 'I', u 'and' u 'and' u 'yue', u, 'the' u 'in the' u 'and u' no ', u 'I', u 'to' u 'to' u 'and u' is', u 'people' u 'and' u 'see' u 'to' u 'will', u 'a', u 'yi', u, 'then' u 'and', u 'and u', u ',] 
For the word in swords: 
If len (word)==1: 
The continue 
Okwords. Append (word) 
# the second method for word in swords: 
# if word not in nywords: 
# okwords. Append (word) 

# concept-semantic 
Word_counts=collections. Counter (okwords) 
Print (word_counts) 
Word_counts_top10=word_counts. Most_common (10) 
Print (word_counts_top10) 
# word show 
Mask=np. Array (Image. Open (' aixing. PNG ')) 
Love=wordcloud. Wordcloud (
Font_path='C:/Windows/Fonts/simhei. The vera.ttf', 
Background_color="white", 
Mask=mask, 
Max_words=20, 
Max_font_size=100 
) 
Love. Generate_from_frequencies (word_counts) 
Image_colors=wordcloud. ImageColorGenerator (mask) 
Love. Recolor (color_func=image_colors) 
PLT. Imshow (love) 
PLT. Axis (' off ') 
PLT. The show () 

CodePudding user response: