Dialectograms: Machine Learning Differences between Discursive Communities

TL;DR

This paper introduces dialectograms, an unsupervised visualization method that maps how different communities use words distinctly, revealing nuanced differences in discourse beyond simple word usage variations.

Contribution

It presents a novel visualization technique and a new measure for analyzing differences in word usage across communities using full embedding spaces.

Findings

Revealed affective polarization in political discourse

Identified differences in perceptions of political actions

Mapped community-specific word usage patterns

Abstract

Word embeddings provide an unsupervised way to understand differences in word usage between discursive communities. A number of recent papers have focused on identifying words that are used differently by two or more communities. But word embeddings are complex, high-dimensional spaces and a focus on identifying differences only captures a fraction of their richness. Here, we take a step towards leveraging the richness of the full embedding space, by using word embeddings to map out how words are used differently. Specifically, we describe the construction of dialectograms, an unsupervised way to visually explore the characteristic ways in which each community use a focal word. Based on these dialectograms, we provide a new measure of the degree to which words are used differently that overcomes the tendency for existing measures to pick out low frequent or polysemous words. We apply our methods to explore the discourses of two US political subreddits and show how our methods identify stark affective polarisation of politicians and political entities, differences in the assessment of proper political action as well as disagreement about whether certain issues require political intervention at all.