Assessing Demographic Bias in Named Entity Recognition
This work highlights potential biases in automated knowledge base generation due to systematic exclusion of named entities from certain demographics, which is an incremental analysis of existing methods.
The study assessed demographic bias in Named Entity Recognition (NER) systems for English using synthetically generated corpora, finding that models perform better for specific demographic groups and that debiased embeddings do not resolve this issue, with ELMo showing the least bias.
Named Entity Recognition (NER) is often the first step towards automated Knowledge Base (KB) generation from raw text. In this work, we assess the bias in various Named Entity Recognition (NER) systems for English across different demographic groups with synthetically generated corpora. Our analysis reveals that models perform better at identifying names from specific demographic groups across two datasets. We also identify that debiased embeddings do not help in resolving this issue. Finally, we observe that character-based contextualized word representation models such as ELMo results in the least bias across demographics. Our work can shed light on potential biases in automated KB generation due to systematic exclusion of named entities belonging to certain demographics.