Aggressive Language Detection with Joint Text Normalization via Adversarial Multi-task Learning

TL;DR

This paper introduces an adversarial multi-task learning framework that jointly performs text normalization and aggressive language detection, significantly improving detection accuracy on multiple datasets by addressing social media text irregularities.

Contribution

It proposes a novel joint learning approach with adversarial training for ALD and TN, enhancing detection performance over existing methods.

Findings

Outperforms all baselines on four datasets

Joint learning improves ALD accuracy

Adversarial training effectively separates task-specific features

Abstract

Aggressive language detection (ALD), detecting the abusive and offensive language in texts, is one of the crucial applications in NLP community. Most existing works treat ALD as regular classification with neural models, while ignoring the inherent conflicts of social media text that they are quite unnormalized and irregular. In this work, we target improving the ALD by jointly performing text normalization (TN), via an adversarial multi-task learning framework. The private encoders for ALD and TN focus on the task-specific features retrieving, respectively, and the shared encoder learns the underlying common features over two tasks. During adversarial training, a task discriminator distinguishes the separate learning of ALD or TN. Experimental results on four ALD datasets show that our model outperforms all baselines under differing settings by large margins, demonstrating the necessity of joint learning the TN with ALD. Further analysis is conducted for a better understanding of our method.