Python word cloud from a html page

Introduction

Analyze html tables using word clouds.

Theory

Fetching the table

Data scraping is easiest using Pandas. BeautifulSoup is an other good option.

Linguistic analyzation

The Finnish language is used, thus Voikko morphological analyzer is used to lemmatize the words into the base format.

sudo apt -y install -y voikko-fi python-libvoikko
pip3 install libvoikko

References

https://data.solita.fi/finnish-stemming-and-lemmatization-in-python/

See Tarmo perusmuodoistaja

The word cloud

WordCloud library is easy to use. It can create svg graphics, but fonts might get mixed.

#The word cloud
from wordcloud import WordCloud, STOPWORDS
import matplotlib.pyplot as plt

stopwords = set(STOPWORDS)


#.generate_from_frequencies()

wordcloud = WordCloud(width = 800, height = 800,
                background_color ='white',
                stopwords = stopwords,
                collocations=False, 
                min_font_size = 10).generate( ' '.join(map(str, bwords)))

Save in svg format

name = title + "_" + AS.replace(" ","") 

wordcloud_svg = wordcloud.to_svg(embed_font=True)
f = open(name + ".svg","w+")
f.write(wordcloud_svg )
f.close()

# plot the WordCloud image
plt.figure(figsize = (8, 8), facecolor = None)
plt.imshow(wordcloud)
plt.axis("off")
plt.tight_layout(pad = 0)

#plt.rcParams['svg.fonttype'] = 'non

plt.rcParams["savefig.format"] = 'png'

print( name )
plt.savefig( name )

plt.show()

The code

See the code at Github.