There may be an plentiful quantity of unstructured knowledge about historic occasions — information articles, authorities stories, and native bulletins — however extracting this info manually at scale is unattainable. Our methodology analyzes information stories the place flooding is a major topic. We then use the Google Learn Aloud user-agent to isolate major textual content from 80 languages, which is standardized into English through the Cloud Translation API.
Probably the most important step of the extraction course of is finished utilizing the Gemini Massive Language Mannequin (LLM). We engineered a classy immediate that guides Gemini by a strict analytical verification course of:
- Classification: The mannequin distinguishes between stories of precise, ongoing, or previous floods and articles that merely talk about future warnings, coverage conferences, or normal threat modeling.
- Temporal reasoning: Gemini anchors relative references (e.g., “final Tuesday”) towards an article’s publication date to find out exact occasion timing.
- Spatial precision: The system identifies granular areas (neighborhoods and streets) and maps them to standardized spatial polygons utilizing utilizing Google Maps Platform.
The technical validation of Groundsource confirms its reliability for high-stakes analysis. In guide evaluations, we discovered that 60% of extracted occasions had been correct in each location and timing. Crucially, 82% had been correct sufficient to be virtually helpful for real-world evaluation — for instance, by capturing the right administrative district or pinpointing the occasion inside a single day of its reported peak.
The protection supplied by Groundsource represents a massive-scale enlargement over current archives. By remodeling unstructured media into knowledge, now we have generated 2.6 million occasions — a big enhance in comparison with the information present in conventional monitoring methods. Moreover, spatiotemporal matching reveals that Groundsource captured between 85% and 100% of the extreme flood occasions recorded by GDACS between 2020 and 2026, an indication of its effectiveness in figuring out high-impact disasters alongside smaller, localized occasions.
