Article: Deep learning-based public crisis event identification for multimodal data contexts Journal: International Journal of Information and Communication Technology (IJICT) 2026 Vol.27 No.59 pp.98 - 116 Abstract: Public crisis events surface across text, sensors, imagery, and logs, yet single-source detectors miss early weak cues. To address fragmented evidence, this study presents a multimodal bidirectional transformer for crisis recognition. First, source-aware tokens preserve time, space, provenance, and quality while synchronisation gates align asynchronous streams. Then, cross-source attention separates corroboration from dissent and memory tokens retain long-range hints. Finally, self-supervised pretraining and calibrated classification deliver auditable alerts. On composite streams, the method reaches AUPRC 0.612, AUROC 0.915, F1 0.672, ECE 0.038, and average lead time 31.6 minutes, exceeding the best baseline by 7.9 AUPRC points, 3.2 AUROC points, 6.9 F1 points, and 8.5 minutes. These gains provide earlier, more reliable, and well-calibrated alerts for public response. Inderscience Publishers - linking academia, business and industry through research

Title: Deep learning-based public crisis event identification for multimodal data contexts

Authors: Wei Gao

Addresses: Basic Courses Department, Zhejiang Police College, Hangzhou 310053, China

Abstract: Public crisis events surface across text, sensors, imagery, and logs, yet single-source detectors miss early weak cues. To address fragmented evidence, this study presents a multimodal bidirectional transformer for crisis recognition. First, source-aware tokens preserve time, space, provenance, and quality while synchronisation gates align asynchronous streams. Then, cross-source attention separates corroboration from dissent and memory tokens retain long-range hints. Finally, self-supervised pretraining and calibrated classification deliver auditable alerts. On composite streams, the method reaches AUPRC 0.612, AUROC 0.915, F1 0.672, ECE 0.038, and average lead time 31.6 minutes, exceeding the best baseline by 7.9 AUPRC points, 3.2 AUROC points, 6.9 F1 points, and 8.5 minutes. These gains provide earlier, more reliable, and well-calibrated alerts for public response.

Keywords: event identification; multimodal data; bidirectional transformer; spatiotemporal alignment.

DOI: 10.1504/IJICT.2026.153930

International Journal of Information and Communication Technology, 2026 Vol.27 No.59, pp.98 - 116

Received: 16 Feb 2026
Accepted: 24 Mar 2026
Published online: 08 Jun 2026 *

Title: Deep learning-based public crisis event identification for multimodal data contexts

Keep up-to-date