Mitigating Gender Stereotypes in Hindi and Marathi

TL;DR

This paper evaluates gender stereotypes in Hindi and Marathi NLP systems, creating a dataset and applying debiasing techniques to reduce bias in word embeddings, addressing a gap in Indic language NLP research.

Contribution

It introduces a novel dataset and bias measurement methods tailored for gendered Indic languages and proposes effective debiasing techniques for Hindi and Marathi.

Findings

Bias reduction achieved in embeddings after applying proposed techniques

Gender bias significantly present in occupation and emotion words

Debiasing methods outperform baseline in bias metrics

Abstract

As the use of natural language processing increases in our day-to-day life, the need to address gender bias inherent in these systems also amplifies. This is because the inherent bias interferes with the semantic structure of the output of these systems while performing tasks like machine translation. While research is being done in English to quantify and mitigate bias, debiasing methods in Indic Languages are either relatively nascent or absent for some Indic languages altogether. Most Indic languages are gendered, i.e., each noun is assigned a gender according to each language's grammar rules. As a consequence, evaluation differs from what is done in English. This paper evaluates the gender stereotypes in Hindi and Marathi languages. The methodologies will differ from the ones in the English language because there are masculine and feminine counterparts in the case of some words. We create a dataset of neutral and gendered occupation words, emotion words and measure bias with the help of Embedding Coherence Test (ECT) and Relative Norm Distance (RND). We also attempt to mitigate this bias from the embeddings. Experiments show that our proposed debiasing techniques reduce gender bias in these languages.