Multilingual Offensive Language Identification with Cross-lingual Embeddings
This work addresses the challenge of detecting offensive content in non-English social media for companies and governments, though it is incremental as it builds on existing cross-lingual methods.
The paper tackled the problem of offensive language identification in low-resource languages by applying cross-lingual embeddings and transfer learning from English data, achieving F1 macro scores of 0.8415 for Bengali, 0.8568 for Hindi, and 0.7513 for Spanish.
Offensive content is pervasive in social media and a reason for concern to companies and government organizations. Several studies have been recently published investigating methods to detect the various forms of such content (e.g. hate speech, cyberbulling, and cyberaggression). The clear majority of these studies deal with English partially because most annotated datasets available contain English data. In this paper, we take advantage of English data available by applying cross-lingual contextual word embeddings and transfer learning to make predictions in languages with less resources. We project predictions on comparable data in Bengali, Hindi, and Spanish and we report results of 0.8415 F1 macro for Bengali, 0.8568 F1 macro for Hindi, and 0.7513 F1 macro for Spanish. Finally, we show that our approach compares favorably to the best systems submitted to recent shared tasks on these three languages, confirming the robustness of cross-lingual contextual embeddings and transfer learning for this task.