WCSE

An Automatic Approach to Build Geographical Knowledge Base from Geographical Data Sources for Earthquake Emergency Response

Abstract— Geographical knowledge base plays crucial role in earthquake emergency. However, how to automatically build a such knowledge base is still a challenge for practical use. In this research, we propose a novel method to extract geographical knowledge from geographical data sources for earthquake emergency. We first design a flexible hierarchical concept model, then we propose an extracting algorithm to extract knowledge from multiple data sources. To handle the place name disambiguation problem, we suggest a Hausdorff distance based method called MRBH-dist. We represent the extracted knowledge base in RDF format and use it via Protégé. This work has already applied successfully to National Earthquake Society Service Project (NESSP) in China.

Index Terms— earthquake emergency management, knowledge base, ontology; place name disambiguation

Qin-Yong Li, Jian-Gong Song, Jiang-Hua Lv, Shi-Long Ma
Beihang University, CHINA

[Download]

Cite: Qin-Yong Li, Jian-Gong Song, Jiang-Hua Lv, Shi-Long Ma, "An Automatic Approach to Build Geographical Knowledge Base from Geographical Data Sources for Earthquake Emergency Response," Proceedings of 2017 the 7th International Workshop on Computer Science and Engineering, pp. 654-658, Beijing, 25-27 June, 2017.