# Discovering and Categorising Language Biases in Reddit

arXiv: https://arxiv.org/abs/2008.02754 · HTML: https://mlsanity.com/paper/2008.02754/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Aug 2020.

## Built on by (most cited first, 13 of 13)

- [Ethical and social risks of harm from Language Models](https://mlsanity.com/paper/2112.04359/) (2021)
- [Societal Biases in Language Generation: Progress and Challenges](https://mlsanity.com/paper/2105.04054/) (2021)
- [Dialect prejudice predicts AI decisions about people's character, employability, and criminality](https://mlsanity.com/paper/2403.00742/) (2024)
- [Bias A-head? Analyzing Bias in Transformer-Based Language Model Attention Heads](https://mlsanity.com/paper/2311.10395/) (2023)
- [The Media Bias Taxonomy: A Systematic Literature Review on the Forms and Automated Detection of Media Bias](https://mlsanity.com/paper/2312.16148/) (2023)
- [Using Semantic Similarity and Text Embedding to Measure the Social Media Echo of Strategic Communications](https://mlsanity.com/paper/2303.16694/) (2023)
- [Exploring Bengali Religious Dialect Biases in Large Language Models with Evaluation Perspectives](https://mlsanity.com/paper/2407.18376/) (2024)
- [How Toxic Can You Get? Search-based Toxicity Testing for Large Language Models](https://mlsanity.com/paper/2501.01741/) (2025)
- [Detecting Natural Language Biases with Prompt-based Learning](https://mlsanity.com/paper/2309.05227/) (2023)
- [Neural Contrastive Clustering: Fully Unsupervised Bias Reduction for Sentiment Classification](https://mlsanity.com/paper/2204.10467/) (2022)
- [Applying Standards to Advance Upstream & Downstream Ethics in Large Language Models](https://mlsanity.com/paper/2306.03503/) (2023)
- [A Holistic Indicator of Polarization to Measure Online Sexism](https://mlsanity.com/paper/2404.02205/) (2024)
- [PoPreRo: A New Dataset for Popularity Prediction of Romanian Reddit Posts](https://mlsanity.com/paper/2407.04541/) (2024)

## Abstract (arXiv)

We present a data-driven approach using word embeddings to discover and categorise language biases on the discussion platform Reddit. As spaces for isolated user communities, platforms such as Reddit are increasingly connected to issues of racism, sexism and other forms of discrimination. Hence, there is a need to monitor the language of these groups. One of the most promising AI approaches to trace linguistic biases in large textual datasets involves word embeddings, which transform text into high-dimensional dense vectors and capture semantic relations between words. Yet, previous studies require predefined sets of potential biases to study, e.g., whether gender is more or less associated with particular types of jobs. This makes these approaches unfit to deal with smaller and community-centric datasets such as those on Reddit, which contain smaller vocabularies and slang, as well as biases that may be particular to that community. This paper proposes a data-driven approach to automatically discover language biases encoded in the vocabulary of online discourse communities on Reddit. In our approach, protected attributes are connected to evaluative words found in the data, which are then categorised through a semantic analysis system. We verify the effectiveness of our method by comparing the biases we discover in the Google News dataset with those found in previous literature. We then successfully discover gender bias, religion bias, and ethnic bias in different Reddit communities. We conclude by discussing potential application scenarios and limitations of this data-driven bias discovery method.