Analysis of Socially Unacceptable Discourse with Zero-shot Learning
This work addresses the need for robust tools to analyze SUD and promote responsible online communication, but it is incremental as it applies existing zero-shot methods to a specific domain.
The paper tackled the problem of detecting and characterizing Socially Unacceptable Discourse (SUD) online by applying entailment-based zero-shot text classification with pre-trained transformer models, showing good generalization to unseen data and potential for generating labeled datasets for extremist narratives.
Socially Unacceptable Discourse (SUD) analysis is crucial for maintaining online positive environments. We investigate the effectiveness of Entailment-based zero-shot text classification (unsupervised method) for SUD detection and characterization by leveraging pre-trained transformer models and prompting techniques. The results demonstrate good generalization capabilities of these models to unseen data and highlight the promising nature of this approach for generating labeled datasets for the analysis and characterization of extremist narratives. The findings of this research contribute to the development of robust tools for studying SUD and promoting responsible communication online.