Transparent Semantic Change Detection with Dependency-Based Profiles

TL;DR

This paper introduces a dependency-based profile method for lexical semantic change detection that is effective, interpretable, and outperforms some neural embedding approaches.

Contribution

It presents a novel, transparent approach relying on dependency co-occurrence patterns, offering better interpretability and competitive performance.

Findings

Dependency-based profiles outperform some neural embedding models.

The method provides plausible and interpretable predictions.

It demonstrates effectiveness in semantic change detection.

Abstract

Most modern computational approaches to lexical semantic change detection (LSC) rely on embedding-based distributional word representations with neural networks. Despite the strong performance on LSC benchmarks, they are often opaque. We investigate an alternative method which relies purely on dependency co-occurrence patterns of words. We demonstrate that it is effective for semantic change detection and even outperforms a number of distributional semantic models. We provide an in-depth quantitative and qualitative analysis of the predictions, showing that they are plausible and interpretable.