Skip to content

Latest commit

 

History

History
11 lines (7 loc) · 833 Bytes

README.md

File metadata and controls

11 lines (7 loc) · 833 Bytes

Computing Melville

"Computing Melville" is a small-scale annotation campaign carried out using Transkribus. The aim of the project is to develop a Machine Learning model able to perform an automated transcription task on a collection of Herman Melville's handwritten manuscripts regarding his last novella, "Billy Budd", left unfinished in 1891, and "Rip Van Winkle's Lilac", an experimental combination of prose and poetry.

The project was developed for the final exam of the course "Semantic Digital Libraries" held by Giovanni Colavizza @ University of Bologna, MA "DHDK"

The website is available at the following link.

Contributors