Complaint Identification in Social Media with Transformer Networks

Abstract

Complaining is a speech act extensively used by humans to communicate a negative inconsistency between reality and expectations. Previous work on automatically identifying complaints in social media has focused on using feature-based and task-specific neural network models. Adapting state-of-the-art pre-trained neural language models and their combinations with other linguistic information from topics or sentiment for complaint prediction has yet to be explored. In this paper, we evaluate a battery of neural models underpinned by transformer networks which we subsequently combine with linguistic information. Experiments on a publicly available data set of complaints demonstrate that our models outperform previous state-of-the-art methods by a large margin achieving a macro F1 up to 87.

arXiv commentAccepted at COLING 2020

References in corpus (5)

• Regularizing and Optimizing LSTM Language Models
• Integrating Multimodal Information in Large Pretrained Transformers
• Identifying leading indicators of product recalls from online reviews using positive unlabeled learning and domain adaptation
• Humor Detection: A Transformer Gets the Last Laugh
• Effective Use of Transformer Networks for Entity Tracking