Seeing Hate Differently: Hate Subspace Modeling for Culture-Aware Hate Speech Detection
This work addresses hate speech detection for diverse cultural groups, but it is incremental as it builds on existing methods with specific enhancements.
The paper tackles the problem of biased training labels and varying interpretations of hate speech across cultural backgrounds by proposing a culture-aware framework that constructs individuals' hate subspaces, resulting in an average performance improvement of 1.05% over state-of-the-art methods.
Hate speech detection has been extensively studied, yet existing methods often overlook a real-world complexity: training labels are biased, and interpretations of what is considered hate vary across individuals with different cultural backgrounds. We first analyze these challenges, including data sparsity, cultural entanglement, and ambiguous labeling. To address them, we propose a culture-aware framework that constructs individuals' hate subspaces. To alleviate data sparsity, we model combinations of cultural attributes. For cultural entanglement and ambiguous labels, we use label propagation to capture distinctive features of each combination. Finally, individual hate subspaces, which in turn can further enhance classification performance. Experiments show our method outperforms state-of-the-art by 1.05\% on average across all metrics.