Hate Speech Detection in Limited Data Contexts using Synthetic Data Generation

TL;DR

This paper introduces a synthetic data generation approach to improve hate speech detection in low-resource languages, enabling effective model training with limited data by transferring hate targets from high-resource language examples.

Contribution

The paper presents three novel methods for synthesizing hate speech data in low-resource languages using high-resource language examples, enhancing hate speech detection where data is scarce.

Findings

Synthetic data improves model performance in low-resource languages.

Models trained on synthetic data can outperform those trained on limited real data.

The approach enables hate speech detection in languages with minimal existing data.

Abstract

A growing body of work has focused on text classification methods for detecting the increasing amount of hate speech posted online. This progress has been limited to only a select number of highly-resourced languages causing detection systems to either under-perform or not exist in limited data contexts. This is majorly caused by a lack of training data which is expensive to collect and curate in these settings. In this work, we propose a data augmentation approach that addresses the problem of lack of data for online hate speech detection in limited data contexts using synthetic data generation techniques. Given a handful of hate speech examples in a high-resource language such as English, we present three methods to synthesize new examples of hate speech data in a target language that retains the hate sentiment in the original examples but transfers the hate targets. We apply our approach to generate training data for hate speech classification tasks in Hindi and Vietnamese. Our findings show that a model trained on synthetic data performs comparably to, and in some cases outperforms, a model trained only on the samples available in the target domain. This method can be adopted to bootstrap hate speech detection models from scratch in limited data contexts. As the growth of social media within these contexts continues to outstrip response efforts, this work furthers our capacities for detection, understanding, and response to hate speech.