Webinar on Cross-Linguistic Colexifications

The Chair for Multilingual Computational Linguistics at the University of Passau organizes an open webinar on Cross-Linguistic Colexifications, with two talks that offer basic overviews on current methodology that can be used to assemble and analyze cross-linguistic colexification data.

The talks by Johann-Mattis List (University of Passau) and Thanasis Georgakopoulos (University of Thessaloniki) will take place via Zoom on November 19, 2024 from 16:15 to 17:45 CET.

Participation is open for everybody. To register and obtain a Zoom link, we ask you kindly to send an email to mcl-admin@uni-passau.de.

Abstracts

Introduction to Cross-Linguistic Colexifications (Johann-Mattis List)

Johann-Mattis List will give a basic introduction to cross-linguistic colexifications. Colexification is a cover term used to denote cases of polysemy and homophony in individual languages. While the term itself is a mere terminological convention, giving up on the distinction if cases where the same sound sequence can code for different senses in one and the same language are due to the polysemy of one word or due to the homophony of multiple words, has helped us to design computer-assisted approaches that allow us to aggregate large-scale colexification data from multilingual wordlists and to analyze them in various ways. The webinar will introduce some basic ideas regarding cross-linguistic colexifications and then show how we use computer-assisted methods to aggregate data for the Database of Cross-Linguistic Colexifications (https://clics.clld.org).

Introduction to Semantic Maps (Thanasis Georgakopoulos)

Thanasis Georgakopoulos will give a brief overview of the semantic maps model, a powerful, theory-neutral, methodological tool for representing meaning connections cross-linguistically. In the early years of the model, semantic maps were mainly hand-crafted, but eventually efficient algorithms were developed that can automatically generate the maps directly from linguistic data. In the webinar, we will retrace the development of these different approaches to map creation and move from theory to practice. Applying the method to data from a sample of languages, we will see how one can construct such graphs both manually and automatically.