# Unsupervised Discovery of Gendered Language through Latent-Variable Modeling

## TL;DR

This paper introduces a generative latent-variable model to quantify and analyze gendered language differences, revealing significant stereotypical distinctions in descriptions of men and women.

## Contribution

The paper presents a novel unsupervised model that jointly captures adjective choice and sentiment conditioned on gender, enabling quantitative analysis of gendered language.

## Key findings

- Significant gender differences in adjective use were identified.
- Positive descriptions of women often relate to their bodies.
- Differences align with common gender stereotypes.

## Abstract

Studying the ways in which language is gendered has long been an area of interest in sociolinguistics. Studies have explored, for example, the speech of male and female characters in film and the language used to describe male and female politicians. In this paper, we aim not to merely study this phenomenon qualitatively, but instead to quantify the degree to which the language used to describe men and women is different and, moreover, different in a positive or negative way. To that end, we introduce a generative latent-variable model that jointly represents adjective (or verb) choice, with its sentiment, given the natural gender of a head (or dependent) noun. We find that there are significant differences between descriptions of male and female nouns and that these differences align with common gender stereotypes: Positive adjectives used to describe women are more often related to their bodies than adjectives used to describe men.