License: Creative Commons Attribution 4.0 International license (CC BY 4.0)
When quoting this document, please refer to the following
DOI: 10.4230/LIPIcs.GIScience.2023.28
URN: urn:nbn:de:0030-drops-189233
URL: https://drops.dagstuhl.de/opus/volltexte/2023/18923/
Go to the corresponding LIPIcs Volume Portal


Feng, Yu ; Ding, Linfang ; Xiao, Guohui

GeoQAMap - Geographic Question Answering with Maps Leveraging LLM and Open Knowledge Base (Short Paper)

pdf-format:
LIPIcs-GIScience-2023-28.pdf (1 MB)


Abstract

GeoQA (Geographic Question Answering) is an emerging research field in GIScience, aimed at answering geographic questions in natural language. However, developing systems that seamlessly integrate structured geospatial data with unstructured natural language queries remains challenging. Recent advancements in Large Language Models (LLMs) have facilitated the application of natural language processing in various tasks. To achieve this goal, this study introduces GeoQAMap, a system that first translates natural language questions into SPARQL queries, then retrieves geospatial information from Wikidata, and finally generates interactive maps as visual answers. The system exhibits great potential for integration with other geospatial data sources such as OpenStreetMap and CityGML, enabling complicated geographic question answering involving further spatial operations.

BibTeX - Entry

@InProceedings{feng_et_al:LIPIcs.GIScience.2023.28,
  author =	{Feng, Yu and Ding, Linfang and Xiao, Guohui},
  title =	{{GeoQAMap - Geographic Question Answering with Maps Leveraging LLM and Open Knowledge Base}},
  booktitle =	{12th International Conference on Geographic Information Science (GIScience 2023)},
  pages =	{28:1--28:7},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-288-4},
  ISSN =	{1868-8969},
  year =	{2023},
  volume =	{277},
  editor =	{Beecham, Roger and Long, Jed A. and Smith, Dianna and Zhao, Qunshan and Wise, Sarah},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/opus/volltexte/2023/18923},
  URN =		{urn:nbn:de:0030-drops-189233},
  doi =		{10.4230/LIPIcs.GIScience.2023.28},
  annote =	{Keywords: Geographic Question Answering, Large Language Models, SPARQL, Knowledge Base, Wikidata}
}

Keywords: Geographic Question Answering, Large Language Models, SPARQL, Knowledge Base, Wikidata
Collection: 12th International Conference on Geographic Information Science (GIScience 2023)
Issue Date: 2023
Date of publication: 07.09.2023


DROPS-Home | Fulltext Search | Imprint | Privacy Published by LZI