Inferring Geographical Ontologies from Multiple Resources for Geographical Information Retrieval

Many documents that can be found in the World Wide Web include some kind of geographical information, often in an implicit way. The use of resources like gazetteers and geographical ontologies can improve the results in Geographical Information Retrieval. Unfortunately, the construction of such ontologies is a long and laborious process; therefore, building an ontology in a semi-automatical way exploiting multiple sources is an interesting and useful task. Our work is focused on the integration of data from gazetteers (GNS and GNIS), the WordNet general domain ontology, and Wikipedia, the free encyclopedia. The result of our effort is an ontology implemented as a set of Prolog clauses, that can be easily expanded with both new data and relationships. Categories and Subject Descriptors H.4 [Information Systems Applications]: Miscellaneous General Terms Geographical Information Retrieval, Ontologies, Wikipedia, WordNet, gazetteers

• Type: Conference
• Year: 2006
• Where: GIR
• Authors: Davide Buscaldi, Paolo Rosso, Piedachu Peris
• Keywords: Geographical Information, Geographical Information Retrieval, Geographical Ontologies, GIR 2006, Information Retrieval