Unsupervised Discovery of Implicit Gender Bias
This work addresses the challenge of detecting subtle social biases in various domains without relying on subjective human judgments, which is incremental as it builds on existing methods for bias detection.
The paper tackled the problem of identifying implicit gender bias in text, particularly against women, by developing an unsupervised model that surfaces biased comments, revealing that biased comments towards female politicians involve mixed criticisms while those towards other female public figures focus on appearance and sexualization.
Despite their prevalence in society, social biases are difficult to identify, primarily because human judgements in this domain can be unreliable. We take an unsupervised approach to identifying gender bias against women at a comment level and present a model that can surface text likely to contain bias. Our main challenge is forcing the model to focus on signs of implicit bias, rather than other artifacts in the data. Thus, our methodology involves reducing the influence of confounds through propensity matching and adversarial learning. Our analysis shows how biased comments directed towards female politicians contain mixed criticisms, while comments directed towards other female public figures focus on appearance and sexualization. Ultimately, our work offers a way to capture subtle biases in various domains without relying on subjective human judgements.