Unknown

Dataset Information

Analyzing differences between discursive communities using dialectograms.


ABSTRACT: Word embeddings provide an unsupervised way to understand differences in word usage between discursive communities. A number of 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 uses a focal word. Based on these dialectograms, we provide a new measure of the degree to which words are used differently that overcomes

SUBMITTER: Enggaard T 

PROVIDER: S-EPMC11438868 | biostudies-literature | 2024 Sep

REPOSITORIES: biostudies-literature