09 — Text as data

A lot has been written.

Word.

Slack channel: #09-text-as-data

Text in often unstructured — but if of course contains information. This week we work with text and methods that tease of the information into a structured, computable shape.

Lecture slides

Download as pdf

Code

Check the course repository for the application.

Further recommended resources

• https://smithio.medium.com/scraping-airbnb-website-with-python-beautiful-soup-and-selenium-8ec86e327b6c
• https://web.stanford.edu/~gentzkow/research/text-as-data.pdf
• https://www.journals.uchicago.edu/doi/full/10.1086/688176
• https://www.nber.org/system/files/working_papers/w26577/w26577.pdf