Multi-criteria geographic information retrieval model based on geospatial semantic integration

Abstract

The geospatial semantic web development requires search mechanisms to overcome the syntactic comparisons and perform semantic analysis in order to retrieve information conceptually similar to the searched one by the user. This would allow reducing the risk of return empty results no match found when there is no exact correspondence between the query and the information available in data repositories. In this work, we describe an information retrieval model to integrate a semantic criterion with geospatial criteria in a no homogenous vector space. The criteria represent the dimensions of this space; these dimensions are weighted in function of the user's preferences or his/her profile. The integration is based on a mathematical expression to evaluate the relevance of each result. We present a system that implements the geospatial semantic approach proposed to retrieve information from the domain of cultural tourism, specifically museums. The results show the advantages of integrating geospatial and semantic criteria taking into account user profiles to offer more customized (personalized) service. © 2011 Springer-Verlag.