Measuring diachronic sense change: new models and Monte Carlo methods for Bayesian inference

TL;DR

This paper develops new Bayesian models and Monte Carlo methods to measure and analyze how word senses change over time, demonstrated on ancient Greek texts with automatic sense annotation.

Contribution

It introduces a simplified generative sense change model and more efficient MCMC inference methods, enabling uncertainty quantification in diachronic sense analysis.

Findings

First to quantify uncertainty in sense prevalence over time

Achieves better efficiency in Bayesian inference methods

Provides credible sets aligning with expert annotations

Abstract

In a bag-of-words model, the senses of a word with multiple meanings, e.g. "bank" (used either in a river-bank or an institution sense), are represented as probability distributions over context words, and sense prevalence is represented as a probability distribution over senses. Both of these may change with time. Modelling and measuring this kind of sense change is challenging due to the typically high-dimensional parameter space and sparse datasets. A recently published corpus of ancient Greek texts contains expert-annotated sense labels for selected target words. Automatic sense-annotation for the word "kosmos" (meaning decoration, order or world) has been used as a test case in recent work with related generative models and Monte Carlo methods. We adapt an existing generative sense change model to develop a simpler model for the main effects of sense and time, and give MCMC methods for Bayesian inference on all these models that are more efficient than existing methods. We carry out automatic sense-annotation of snippets containing "kosmos" using our model, and measure the time-evolution of its three senses and their prevalence. As far as we are aware, ours is the first analysis of this data, within the class of generative models we consider, that quantifies uncertainty and returns credible sets for evolving sense prevalence in good agreement with those given by expert annotation.