Building ontologies for disaster management: seismic risk domain

The evolution of technologies and Internet generates an increasing need to share information on the Web. In particular, while spatial data infrastructures on one hand increased geographical data availability, on the other produced semantic interoperability barriers among systems. Ontology development may represent an improvement of semantic interoperability. The increase of economical and social costs, deriving from seismic events, leads to a prevention strategy to reduce damages. The need of having a systematic approach to the definitions of seismic risk has led to the attempt of constructing an ontology. This paper aims at building an ontology in a specific domain of risk prevention and disaster management. Specifically, an ontology has been built in the field of seismic risk following three phases: finding main concepts of domain (Seismic Risk) associating the definition included in our glossary; hierarchically organizing concepts finding fathers and sons of concepts with relationship IS_A; studying other types of relationships associated to concepts. In a context where definition abound, with a very articulate vocabulary, achieving consensus is not an easy task. In analyzing territorial phenomena domains, preconsensus ontologies have been preferred, collecting previously all existing definitions, and then seeking consensus among all actors. On this purpose the ontology has been implemented using two tools, Towntology and Protégé.