Saturday, June 8, 2019

Latin WordNet 2.0

Latin WordNet 2.0
The Latin WordNet 2.0 is an on-going collaboration between the University of Exeter and the Linking Latin Project, under the direction of Dr William Michael Short, to create a comprehensive lexico-semantic database of the Latin language. It is intended to model Latin's semantic system as fully and accurately as possible, in a form that is machine-interpretable and machine-actionable, and thus suitable to NLP applications of different kinds, especially in the area of natural language understanding.

This joint project builds on, and extends, the original Latin WordNet developed as part of the Fondazione Bruno Kessler's MultiWordNet project, by including over 70,000 words covering archaic to medieval Latin. A second project, in partnership with the University of Genoa, is creating a dictionary of Latin metaphor (Lexicon Translaticium Latinum) on the basis of the WordNet, to capture rich information about the figurative usage of words, as well as to represent large-scale metaphorical patterns that structure meaning throughout Latin's semantic system. The data of the Latin WordNet is made available under a Creative Commons Attribution-ShareAlike 4.0 International license. and can be found on GitHub.

No comments:

Post a Comment