Federated Multi-task Learning for Complaint Identification from Social Media Data

Abstract

Complaining is a speech act that is often used by consumers to signify a breach of expectation, i.e., an expression of displeasure on a consumer's behalf towards an organization, product, or event. Complaint identification has been previously analyzed based on extensive feature engineering in centralized settings, disregarding the non-identically independently distributed (non-IID), security, and privacy-preserving characteristics of complaints that can hamper data accumulation, distribution, and learning. In this work, we propose a Bidirectional Encoder Representations from Transformers (BERT) based multi-task framework that aims to learn two closely related tasks,viz. complaint identification (primary task) and sentiment classification (auxiliary tasks) concurrently under federated-learning settings. Extensive evaluation on two real-world datasets shows that our proposed framework surpasses the baselines and state-of-the-art framework results by a significant margin.