We were rewatching some old Friends episodes at home when I took notice that the phrase "oh my god" comes up a lot in that show. As any reasonable person would, I compiled script data and built a website to prove my point.
This webapp lets you check how often phrases like "oh my god" are said in Friends, Seinfeld, and Sex and the City.
I tried to make the data build as self-contained as possible. To that end I have hosted some source files on my personal heliohosting server. There are two unrecoverable aspects to the data:
- The source seinfeld data is no longer available online. Colin sent a copy to me, and thats how i have it.
- The Sex and the City data were messy in their public form (via Kaggle). I cleaned those data up a fair amount and saved the file.
The download-all script contains relevant URLs to download the source data used tobuild the final dataset. After downloading those raw files, the build module contains code to process the files in their raw form in order to prodice a final dataset.
You can download a copy of it here!
I did basically zero work obtaining the source data. Below are shout-outs to those who did that hard work:
- Colin Pollock for the excellent Seinfeld dataset.
- Yusuf Sohoye for the regex to parse through Friends script files.
- I stole the Sex and the City data from Kaggle.
You can download my the full (gzipped) SQLite3 database here!