Jun 17, 2022 | 17:00—18:00
Colexification is a linguistic phenomenon that occurs when two different concepts are expressed using only one word in a language.
In this talk, I show that an interdisciplinary approach which adds computer science and psychology to classical linguistics can expand the horizon of what we can do with colexification. In particular, I explore the structure of meanings in human language through the analysis of colexification networks, developing methods to automatically improve affective science resources and text analysis methods. Moreover, I show that models based on colexification networks outperform state-of-the-art approaches.
These results can be seen as a first step towards the construction of a new generation of theoretically-informed, explainable, and transparent methods to understand natural language.