Abstract
The proliferation of online news sources enables rapid dissemination of disaster information, yet much of this content remains unstructured and difficult to analyze automatically. This paper presents a framework for transforming unstructured news reports into structured, actionable disaster intelligence. The approach combines continuous web scraping of news feeds, rule-based content filtering, and large language model (LLM)-based natural language processing for automated classification and information extraction. Key details including disaster type, geographic location, casualties, extent of damage, and official response are aggregated into a structured dataset and visualized on interactive spatio-temporal maps, supporting situational awareness and timely crisis management decisions.
| Original language | English |
|---|---|
| Pages (from-to) | 1503-1510 |
| Number of pages | 8 |
| Journal | Procedia Computer Science |
| Volume | 284 |
| DOIs | |
| Publication status | Published - 2026 |
| Event | 8th Information Systems International Conference, ISICO 2025 - Hybrid, Bali, Indonesia Duration: 4 Aug 2025 → 6 Aug 2025 |
Keywords
- Disaster management
- Information extraction
- Large language models
- Spatio-temporal analysis
- Web scraping
Fingerprint
Dive into the research topics of 'Leveraging large language models and web scraping for automated disaster information extraction from online news'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver