Disentangled and Robust Representation Learning for Bragging Classification in Social Media

TL;DR

This paper introduces a novel bragging classification approach for social media that uses disentangled representation augmentation and domain-aware adversarial training to improve robustness and performance, addressing data imbalance issues.

Contribution

The paper proposes a disentangle-based representation augmentation combined with a domain-aware adversarial strategy for bragging classification, avoiding external knowledge and noise.

Findings

Achieves state-of-the-art performance on bragging classification datasets.

Effectively handles data imbalance without external knowledge.

Improves robustness of features through domain-aware adversarial training.

Abstract

Researching bragging behavior on social media arouses interest of computational (socio) linguists. However, existing bragging classification datasets suffer from a serious data imbalance issue. Because labeling a data-balance dataset is expensive, most methods introduce external knowledge to improve model learning. Nevertheless, such methods inevitably introduce noise and non-relevance information from external knowledge. To overcome the drawback, we propose a novel bragging classification method with disentangle-based representation augmentation and domain-aware adversarial strategy. Specifically, model learns to disentangle and reconstruct representation and generate augmented features via disentangle-based representation augmentation. Moreover, domain-aware adversarial strategy aims to constrain domain of augmented features to improve their robustness. Experimental results demonstrate that our method achieves state-of-the-art performance compared to other methods.