View source for 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 [https://voikko.puimula.org/ Voikko] morphological analyzer is used to lemmatize the words into the base format. 

<pre>
sudo apt -y install -y voikko-fi python-libvoikko
pip3 install libvoikko
</pre>

==== References ====

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

See [https://www.tarmo.fi/perusmuotoon/ Tarmo perusmuodoistaja]

=== The word cloud ===

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

<pre>
from wordcloud import WordCloud, STOPWORDS
import matplotlib.pyplot as plt

stopwords = set(STOPWORDS)

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

=== Save in svg format ===

<pre>
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()

</pre>

== The code ==

See the code at [https://gist.github.com/markkuleino/5173152fdd8fa8877c433d300eb71d45 Github].

== Exercises ==