TY - CHAP

T1 - GANDR - Georelating Dataset, Metrics, and Evaluation

AU - Moltzen, Kai

AU - Usbeck, Ricardo

PY - 2025/12/19

Y1 - 2025/12/19

N2 - Georelating has been introduced to learn geospatial representations of events from textual reports, which requires the interpretation of spatial relations. To foster the development and evaluation of Georelating systems, we construct the silver-standard Georelating Annotated Natural Disaster Reports dataset GANDR and benchmark our LLM agent architecture as a baseline (areal F1 = 0.609, fuzzy cell match score = 0.833) for this new task.GANDR comprises synthetic disaster reports referencing 1,000 US and 1,000 EU cities, annotated with Discrete Global Grid System (DGGS) cells for efficient geospatial integration. We propose a set of five complementary metrics capitalizing on the DGGS annotations for efficient and comprehensive evaluation.Analysis reveals the potential of reasoning LLMs integrated with geographical knowledge bases to address variation across spatial relations such as (inter-)cardinal directions. We highlight the estimation of the impact area's size as a key challenge of Georelating.

AB - Georelating has been introduced to learn geospatial representations of events from textual reports, which requires the interpretation of spatial relations. To foster the development and evaluation of Georelating systems, we construct the silver-standard Georelating Annotated Natural Disaster Reports dataset GANDR and benchmark our LLM agent architecture as a baseline (areal F1 = 0.609, fuzzy cell match score = 0.833) for this new task.GANDR comprises synthetic disaster reports referencing 1,000 US and 1,000 EU cities, annotated with Discrete Global Grid System (DGGS) cells for efficient geospatial integration. We propose a set of five complementary metrics capitalizing on the DGGS annotations for efficient and comprehensive evaluation.Analysis reveals the potential of reasoning LLMs integrated with geographical knowledge bases to address variation across spatial relations such as (inter-)cardinal directions. We highlight the estimation of the impact area's size as a key challenge of Georelating.

KW - georelating

KW - LLMs

KW - spatial reasoning

KW - DGGS

KW - dataset

KW - metrics

KW - Informatics

U2 - 10.1145/3764912.3770819

DO - 10.1145/3764912.3770819

M3 - Article in conference proceedings

SN - 9798400721793

T3 - GeoAI '25

SP - 61

EP - 71

BT - GeoAI '25: Proceedings of the 8th ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery

A2 - Newsam, Shawn

A2 - Yang, Lexie

PB - Association for Computing Machinery

CY - New York, NY, USA

ER -