StereoKG: Data-Driven Knowledge Graph Construction for Cultural Knowledge and Stereotypes
This addresses the need for automated bias analysis tools for NLP researchers and practitioners, offering a scalable alternative to human-compiled lists, though it is incremental as it builds on existing KG and bias detection methods.
The paper tackled the problem of analyzing ethnic or religious bias in NLP models by developing a data-driven pipeline to construct a knowledge graph of cultural knowledge and stereotypes, resulting in a KG covering 5 religious groups and 5 nationalities with 59.2% of non-singleton entries rated as coherent and complete stereotypes, and showing that training on this KG improves cultural awareness and classification performance on hate speech detection.
Analyzing ethnic or religious bias is important for improving fairness, accountability, and transparency of natural language processing models. However, many techniques rely on human-compiled lists of bias terms, which are expensive to create and are limited in coverage. In this study, we present a fully data-driven pipeline for generating a knowledge graph (KG) of cultural knowledge and stereotypes. Our resulting KG covers 5 religious groups and 5 nationalities and can easily be extended to include more entities. Our human evaluation shows that the majority (59.2%) of non-singleton entries are coherent and complete stereotypes. We further show that performing intermediate masked language model training on the verbalized KG leads to a higher level of cultural awareness in the model and has the potential to increase classification performance on knowledge-crucial samples on a related task, i.e., hate speech detection.