12.4 Summary
In this chapter, we looked at text. We started with strings and the characters in them and moved on to regular expressions to perform sophisticated matching and substitution. While character patterns are interesting, they do not tell us much about linguistic content. For that, we saw many examples of natural language processing (NLP) using the spacy Python package.
Understanding text is complicated, even for people sometimes. Use these tools carefully, and do not assume your results represent the absolute truth. For example, compare
“I love your new apartment!”
and
“I love that my boss bought a €150,000 sports car
and I bought a 20-year old rusty van.”
There is “love” in each, but it is easy to miss the sarcasm in the second.
