Diachronic Usage Relatedness (DURel): A Framework for the Annotation of Lexical Semantic Change

TL;DR

This paper introduces DURel, a framework for annotating how lexical meanings change over time, providing a resource for evaluating models of semantic change with high inter-annotator agreement.

Contribution

It extends existing semantic annotation methods to diachronic contexts, creating a German dataset for lexical semantic change evaluation.

Findings

High inter-annotator agreement in relatedness ratings

A test set of 1,320 use pairs across 22 words

Framework effectively captures semantic change

Abstract

We propose a framework that extends synchronic polysemy annotation to diachronic changes in lexical meaning, to counteract the lack of resources for evaluating computational models of lexical semantic change. Our framework exploits an intuitive notion of semantic relatedness, and distinguishes between innovative and reductive meaning changes with high inter-annotator agreement. The resulting test set for German comprises ratings from five annotators for the relatedness of 1,320 use pairs across 22 target words.