| Summary: | A few years back, William Turkel wrote a series of blog posts called A Naive Bayesian in the Old Bailey, which showed how one could use machine learning to extract interesting documents out of a digital archive. This tutorial is a kind of an update on that blog essay, with roughly the same data but a slightly different version of the machine learner.
The idea is to show why machine learning methods are of interest to historians, as well as to present a step-by-step implementation of a supervised machine learner. This learner is then applied to the Old Bailey digital archive, which contains several centuries’ worth of transcripts of trials held at the Old Bailey in London. We will be using Python for the implementation.
|
|---|
