Grammatical Profiling for Semantic Change Detection

TL;DR

This paper introduces grammatical profiling as an alternative to distributional semantics for detecting semantic change, demonstrating its effectiveness and interpretability through comprehensive analysis.

Contribution

It presents a novel grammatical profiling method that captures morphosyntactic changes for semantic change detection, outperforming some existing distributional approaches.

Findings

Grammatical profiling can effectively detect semantic change.

The method provides plausible and interpretable predictions.

It outperforms some distributional semantic methods.

Abstract

Semantics, morphology and syntax are strongly interdependent. However, the majority of computational methods for semantic change detection use distributional word representations which encode mostly semantics. We investigate an alternative method, grammatical profiling, based entirely on changes in the morphosyntactic behaviour of words. We demonstrate that it can be used for semantic change detection and even outperforms some distributional semantic methods. We present an in-depth qualitative and quantitative analysis of the predictions made by our grammatical profiling system, showing that they are plausible and interpretable.