Improving Interpretability of Lexical Semantic Change with Neurobiological Features

TL;DR

This paper introduces a novel approach that maps contextualized word embeddings to neurobiological features, enhancing interpretability of lexical semantic change and outperforming previous methods in estimating change degrees.

Contribution

The paper proposes a neurobiological feature space mapping for interpretability of lexical semantic change, enabling systematic understanding and discovery of overlooked change types.

Findings

The method achieves superior performance in estimating LSC compared to previous approaches.

It uncovers new types of semantic change overlooked by prior studies.

The approach effectively identifies words with specific semantic change patterns.

Abstract

Lexical Semantic Change (LSC) is the phenomenon in which the meaning of a word change over time. Most studies on LSC focus on improving the performance of estimating the degree of LSC, however, it is often difficult to interpret how the meaning of a word change. Enhancing the interpretability of LSC is a significant challenge as it could lead to novel insights in this field. To tackle this challenge, we propose a method to map the semantic space of contextualized embeddings of words obtained by a pre-trained language model to a neurobiological feature space. In the neurobiological feature space, each dimension corresponds to a primitive feature of words, and its value represents the intensity of that feature. This enables humans to interpret LSC systematically. When employed for the estimation of the degree of LSC, our method demonstrates superior performance in comparison to the majority of the previous methods. In addition, given the high interpretability of the proposed method, several analyses on LSC are carried out. The results demonstrate that our method not only discovers interesting types of LSC that have been overlooked in previous studies but also effectively searches for words with specific types of LSC.