EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Multihead Average Pseudo-Margin Learning for Disaster Tweet Classification

Iustin Sîrbu, Robert-Adrian Popovici, Traian Rebedea, Ștefan Trăușan-Matu

Information · 24.05.2025 · 16 (6) · 434

DOI: 10.3390/info16060434 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Information, online erschienen am 24.05.2025. Verfasst von Iustin Sîrbu und 3 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

During natural disasters, social media platforms, such as X (formerly Twitter), become a valuable source of real-time information, with eyewitnesses and affected individuals posting messages about the produced damage and the victims. Although this information can be used to streamline the intervention process of local authorities and to achieve a better distribution of available resources, manually annotating these messages is often infeasible due to time and cost constraints. To address this challenge, we explore the use of semi-supervised learning, a technique that leverages both labeled and unlabeled data, to enhance neural models for disaster tweet classification. Specifically, we investigate state-of-the-art semi-supervised learning models and focus on co-training, a less-explored approach in recent years. Moreover, we propose a novel hybrid co-training architecture, Multihead Average Pseudo-Margin, which obtains state-of-the-art results on several classification tasks. Our approach extends the advantages of the voting mechanism from Multihead Co-Training by using the Average Pseudo-Margin (APM) score to improve the quality of the pseudo-labels and self-adaptive confidence thresholds for improving imbalanced classification. Our method achieves up to 7.98% accuracy improvement in low-data scenarios and 2.84% improvement when using the entire labeled dataset, reaching 89.55% accuracy on the Humanitarian task and 91.23% on the Informative task. These results demonstrate the potential of our approach in addressing the critical need for automated disaster tweet classification. We made our code publicly available for future research.

Zitieren

APA

Sîrbu, I., Popovici, R., Rebedea, T., & Trăușan-Matu, Ș. (2025). Multihead Average Pseudo-Margin Learning for Disaster Tweet Classification. Information, 16(6), 434. https://doi.org/10.3390/info16060434

Vancouver

Sîrbu I, Popovici R, Rebedea T, Trăușan-Matu Ș. Multihead Average Pseudo-Margin Learning for Disaster Tweet Classification. Information. 2025;16(6):434. doi:10.3390/info16060434

BibTeX
@article{Srbu2025,
  author = {Sîrbu, Iustin and Popovici, Robert-Adrian and Rebedea, Traian and Trăușan-Matu, Ștefan},
  title = {Multihead Average Pseudo-Margin Learning for Disaster Tweet Classification},
  journal = {Information},
  year = {2025},
  volume = {16},
  number = {6},
  pages = {434},
  doi = {10.3390/info16060434}
}
Kostenlos starten →