Updates to LexiconMining.jl

Version 0.3.0 of LexiconMining.jl is a ground-up rewrite using ChatGPT to extract data from Lewis-Short’s Latin Dictionary.
morphology
lexica
LLMs
latin
update
Author

Neel Smith

Published

January 17, 2025

Release notes: version 0.3.0

Version 0.3.0 (now available on the main branch of the github repository) is a complete reworking of the previous package. It uses ChatGPT to extract morphological information from Lewis-Short’s Latin Dictionary, parses the ChatGPT output into an object model, and from those objects generates data usable with the Tabulae system for building Latin morphological parsers.

State of implementation in version 0.3.0:

  • complete summaries of Lewis-Short from ChatGPT available
  • initial parsing of ChatGPT summaries for nouns, verbs and adjectives implemented
  • generation of Tabulae data for nouns implemented
More information about Tabulae

See documentation for the Tabulae.jl package.

Installation

LexiconMining.jl is an unpublished package. You can install it directory from its github repository.

  • in package mode from the Julia REPL ] add https://github.com/neelsmith/LexiconMining.jl.git
  • from a Julia terminal session: using Pkg; Pkg.add(url="https://github.com/neelsmith/LexiconMining.jl.git")

Documentation

  • package documentation is available on quarto.pub

Source

Progress

  • check this milestone in the github repository to follow progress in implementing generation of Tabulae data sets