Can We Predict the Unpredictable? Leveraging DisasterNet-LLM for Multimodal Disaster Classification
This addresses the need for timely and accurate disaster insights, though it appears incremental as it builds on existing LLM and multimodal techniques.
The paper tackles the problem of integrating multimodal data for disaster management by proposing DisasterNet-LLM, a specialized LLM that achieves higher accuracy of 89.5%, F1 score of 88.0%, AUC of 0.92%, and BERTScore of 0.88% in classification tasks.
Effective disaster management requires timely and accurate insights, yet traditional methods struggle to integrate multimodal data such as images, weather records, and textual reports. To address this, we propose DisasterNet-LLM, a specialized Large Language Model (LLM) designed for comprehensive disaster analysis. By leveraging advanced pretraining, cross-modal attention mechanisms, and adaptive transformers, DisasterNet-LLM excels in disaster classification. Experimental results demonstrate its superiority over state-of-the-art models, achieving higher accuracy of 89.5%, an F1 score of 88.0%, AUC of 0.92%, and BERTScore of 0.88% in multimodal disaster classification tasks.