One of the easiest things to analyze about your emails is the most frequently used words. We can create a word cloud to see the most frequently used words. Let's first remove the archived emails:

from wordcloud import WordCloud 

df_no_arxiv = dfs[dfs['from'] != '[email protected]']
text = ' '.join(map(str, sent['subject'].values))

Next, let's plot the word cloud:

stopwords = ['Re', 'Fwd', '3A_']
wrd = WordCloud(width=700, height=480, margin=0, collocations=False)
for sw in stopwords:
    wrd.stopwords.add(sw)
wordcloud = wrd.generate(text)

plt.figure(figsize=(25,15))
plt.imshow(wordcloud, interpolation='bilinear')
plt.axis("off")
plt.margins(x=0, y=0)

I added some extra stop words to filter out from the graph. The output for me is as follows:

This tells me what I mostly communicate about. From the analysis of emails from 2011 to 2019, the most frequently used words are new, site, project, Data, WordPress, and website. This is really good, right? What is presented in this chapter is just a starting point. You can take this further in several other directions.