Multilingual Twitter Corpus and Baselines for Evaluating Demographic Bias in Hate Speech Recognition
This work addresses the problem of demographic bias evaluation in hate speech recognition for researchers and practitioners, though it is incremental as it builds on existing fairness evaluation methods with new data.
The authors tackled the lack of real-world multilingual data for evaluating demographic bias in hate speech detection by assembling and publishing a Twitter corpus with inferred author demographics across five languages, and they measured the performance and fairness of four baseline classifiers on this dataset.
Existing research on fairness evaluation of document classification models mainly uses synthetic monolingual data without ground truth for author demographic attributes. In this work, we assemble and publish a multilingual Twitter corpus for the task of hate speech detection with inferred four author demographic factors: age, country, gender and race/ethnicity. The corpus covers five languages: English, Italian, Polish, Portuguese and Spanish. We evaluate the inferred demographic labels with a crowdsourcing platform, Figure Eight. To examine factors that can cause biases, we take an empirical analysis of demographic predictability on the English corpus. We measure the performance of four popular document classifiers and evaluate the fairness and bias of the baseline classifiers on the author-level demographic attributes.