DeCrisisMB: Debiased Semi-Supervised Learning for Crisis Tweet Classification via Memory Bank

TL;DR

DeCrisisMB introduces a memory bank-based debiasing technique for semi-supervised crisis tweet classification, significantly improving performance and generalization in emergency response scenarios.

Contribution

The paper proposes a novel memory bank approach to debias semi-supervised learning for crisis tweet classification, addressing class imbalance and improving accuracy.

Findings

Outperforms existing debiasing methods in experiments

Enhances generalization in out-of-distribution data

Achieves superior results in crisis tweet classification

Abstract

During crisis events, people often use social media platforms such as Twitter to disseminate information about the situation, warnings, advice, and support. Emergency relief organizations leverage such information to acquire timely crisis circumstances and expedite rescue operations. While existing works utilize such information to build models for crisis event analysis, fully-supervised approaches require annotating vast amounts of data and are impractical due to limited response time. On the other hand, semi-supervised models can be biased, performing moderately well for certain classes while performing extremely poorly for others, resulting in substantially negative effects on disaster monitoring and rescue. In this paper, we first study two recent debiasing methods on semi-supervised crisis tweet classification. Then we propose a simple but effective debiasing method, DeCrisisMB, that utilizes a Memory Bank to store and perform equal sampling for generated pseudo-labels from each class at each training iteration. Extensive experiments are conducted to compare different debiasing methods' performance and generalization ability in both in-distribution and out-of-distribution settings. The results demonstrate the superior performance of our proposed method. Our code is available at https://github.com/HenryPengZou/DeCrisisMB.