Domain Classification-based Source-specific Term Penalization for Domain Adaptation in Hate-speech Detection
This addresses domain adaptation for hate-speech detection, an incremental improvement over prior methods that rely on manually curated term lists.
The paper tackles the problem of poor out-of-domain performance in hate-speech detection by proposing a domain adaptation approach that automatically penalizes source-specific terms using a domain classifier and feature-attribution scores, resulting in consistent improvements in cross-domain evaluation.
State-of-the-art approaches for hate-speech detection usually exhibit poor performance in out-of-domain settings. This occurs, typically, due to classifiers overemphasizing source-specific information that negatively impacts its domain invariance. Prior work has attempted to penalize terms related to hate-speech from manually curated lists using feature attribution methods, which quantify the importance assigned to input terms by the classifier when making a prediction. We, instead, propose a domain adaptation approach that automatically extracts and penalizes source-specific terms using a domain classifier, which learns to differentiate between domains, and feature-attribution scores for hate-speech classes, yielding consistent improvements in cross-domain evaluation.