A mixed approach for automated spatial ontology alignment.

The appearance of the semantic web has caused an ever-increasing use of ontologies as formal tools to represent knowledge explicitly in various applications. Since data integration is an essential part of information systems, ontology alignment is becoming more important in the web world. Moreover, applications on the web such as agents need to fuse information automatically and on-the-fly. This paper presents a methodology to automate spatial ontology alignment. In this paper, linguistic scales and extensional information of ontologies as demonstrators of concept semantics along with the structural information of ontologies are used to discover semantic correspondences in a new way. A procedure to compose results from extensional and intensional approaches is presented and error sources and fusion issues are highlighted. The methodology is tested on a case study, which is related to two ontologies containing road data. The results of the ontology alignment indicate a precision of 0.81, and a recall of 0.7. Such achievements present the high efficiency of the proposed method. [ABSTRACT FROM AUTHOR]