A Data Exploration of Jeopardy! from 1984 to the Present

Brian Hamilton

About the Project

This project was born out of a lifelong interest in Jeopardy! and trivia. My first attempt at scraping data from J! Archive was for a Discord bot I created called Trebot. This bot allowed users to call for a random question and keep score over a period of time. It was during the making of this that I developed the first script for collecting data from the site and putting it into a database (MongoDB in this case). It only made use of the information on the site about clues for each game.

In the winter of 2018, I developed the idea for using a modified version of this script to collect the rest of the data, including contestant information and more detailed data on the clues. Go to the Methods page of this site to see more information on how I built the dataset, analyzed it, and visualized it.

Plans for the Future

There is still have more that can be analyzed. For example, a topic model can be done on the professions of the contestants. The database already contains the infrastructure for it built in. I also plan on building out an interactive visualization that would allow for users to input their own text to search for in the clues, answers, and categories. It would show every time the text appears throughout the show's history.