Sarcasm Detection in a Disaster Context
This work addresses the problem of understanding sarcasm in disaster contexts for natural language processing applications, but it is incremental as it applies existing methods to a new dataset.
The paper tackles sarcasm detection in disaster-related tweets by introducing HurricaneSARC, a dataset of 15,000 annotated tweets, and achieves up to 0.70 F1 score using pre-trained language models.
During natural disasters, people often use social media platforms such as Twitter to ask for help, to provide information about the disaster situation, or to express contempt about the unfolding event or public policies and guidelines. This contempt is in some cases expressed as sarcasm or irony. Understanding this form of speech in a disaster-centric context is essential to improving natural language understanding of disaster-related tweets. In this paper, we introduce HurricaneSARC, a dataset of 15,000 tweets annotated for intended sarcasm, and provide a comprehensive investigation of sarcasm detection using pre-trained language models. Our best model is able to obtain as much as 0.70 F1 on our dataset. We also demonstrate that the performance on HurricaneSARC can be improved by leveraging intermediate task transfer learning. We release our data and code at https://github.com/tsosea2/HurricaneSarc.