BanglaMM-Disaster: A Multimodal Transformer-Based Deep Learning Framework for Multiclass Disaster Classification in Bangla

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Islam, Ariful, Hossen, Md Rifat, Arif, Md. Mahmudul, Noman, Abdullah Al, Rahman, Md Arifur
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911288207081472
author Islam, Ariful
Hossen, Md Rifat
Arif, Md. Mahmudul
Noman, Abdullah Al
Rahman, Md Arifur
author_facet Islam, Ariful
Hossen, Md Rifat
Arif, Md. Mahmudul
Noman, Abdullah Al
Rahman, Md Arifur
contents Natural disasters remain a major challenge for Bangladesh, so real-time monitoring and quick response systems are essential. In this study, we present BanglaMM-Disaster, an end-to-end deep learning-based multimodal framework for disaster classification in Bangla, using both textual and visual data from social media. We constructed a new dataset of 5,037 Bangla social media posts, each consisting of a caption and a corresponding image, annotated into one of nine disaster-related categories. The proposed model integrates transformer-based text encoders, including BanglaBERT, mBERT, and XLM-RoBERTa, with CNN backbones such as ResNet50, DenseNet169, and MobileNetV2, to process the two modalities. Using early fusion, the best model achieves 83.76% accuracy. This surpasses the best text-only baseline by 3.84% and the image-only baseline by 16.91%. Our analysis also shows reduced misclassification across all classes, with noticeable improvements for ambiguous examples. This work fills a key gap in Bangla multimodal disaster analysis and demonstrates the benefits of combining multiple data types for real-time disaster response in low-resource settings.
format Preprint
id arxiv_https___arxiv_org_abs_2511_21364
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle BanglaMM-Disaster: A Multimodal Transformer-Based Deep Learning Framework for Multiclass Disaster Classification in Bangla
Islam, Ariful
Hossen, Md Rifat
Arif, Md. Mahmudul
Noman, Abdullah Al
Rahman, Md Arifur
Machine Learning
Computer Vision and Pattern Recognition
Natural disasters remain a major challenge for Bangladesh, so real-time monitoring and quick response systems are essential. In this study, we present BanglaMM-Disaster, an end-to-end deep learning-based multimodal framework for disaster classification in Bangla, using both textual and visual data from social media. We constructed a new dataset of 5,037 Bangla social media posts, each consisting of a caption and a corresponding image, annotated into one of nine disaster-related categories. The proposed model integrates transformer-based text encoders, including BanglaBERT, mBERT, and XLM-RoBERTa, with CNN backbones such as ResNet50, DenseNet169, and MobileNetV2, to process the two modalities. Using early fusion, the best model achieves 83.76% accuracy. This surpasses the best text-only baseline by 3.84% and the image-only baseline by 16.91%. Our analysis also shows reduced misclassification across all classes, with noticeable improvements for ambiguous examples. This work fills a key gap in Bangla multimodal disaster analysis and demonstrates the benefits of combining multiple data types for real-time disaster response in low-resource settings.
title BanglaMM-Disaster: A Multimodal Transformer-Based Deep Learning Framework for Multiclass Disaster Classification in Bangla
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.21364