Evelina project.

So as part of my wider attempts to automate archival transcription, i’ve become interested in automatic identification of authors. I can see two main uses for this in Burney studies: firstly, to figure out who-wrote-what in re-drafting of correspondence late in Burney’s life by her wider circle, and also settling once and for all which books by Mrs Meeke were written by Bessie Allen, Burney’s ne’er-do-well stepsister.

As a first step, I’ve been following this project. AFAICT, if everything runs smoothly, the only issue will be cleaning up the data before feeding it in if we want to apply it to authors beyond Kaggle’s data set.