BOS at LSCDiscovery: Lexical Substitution for Interpretable Lexical Semantic Change Detection

TL;DR

This paper presents an interpretable lexical semantic change detection method for Spanish that uses lexical substitutes to identify sense gains and losses over time, achieving high accuracy in shared task evaluations.

Contribution

The approach introduces a novel use of lexical substitutes for detecting semantic change, enhancing interpretability and providing detailed insights into sense evolution.

Findings

Achieved second best in sense loss and gain detection

Provides interpretable insights into semantic change

Effective in Spanish lexical semantic change detection

Abstract

We propose a solution for the LSCDiscovery shared task on Lexical Semantic Change Detection in Spanish. Our approach is based on generating lexical substitutes that describe old and new senses of a given word. This approach achieves the second best result in sense loss and sense gain detection subtasks. By observing those substitutes that are specific for only one time period, one can understand which senses were obtained or lost. This allows providing more detailed information about semantic change to the user and makes our method interpretable.