Towards automated knowledge-based mapping between individual conceptualisations to empower personalisation of Geospatial Semantic Web

Abstract

Geospatial domain is characterised by vagueness, especially in the semantic disambiguation of the concepts in the domain, which makes defining universally accepted geo- ontology an onerous task. This is compounded by the lack of appropriate methods and techniques where the individual semantic conceptualisations can be captured and compared to each other. With multiple user conceptualisations, efforts towards a reliable Geospatial Semantic Web, therefore, require personalisation where user diversity can be incorporated. The work presented in this paper is part of our ongoing research on applying commonsense reasoning to elicit and maintain models that represent users' conceptualisations. Such user models will enable taking into account the users' perspective of the real world and will empower personalisation algorithms for the Semantic Web. Intelligent information processing over the Semantic Web can be achieved if different conceptualisations can be integrated in a semantic environment and mismatches between different conceptualisations can be outlined. In this paper, a formal approach for detecting mismatches between a user's and an expert's conceptual model is outlined. The formalisation is used as the basis to develop algorithms to compare models defined in OWL. The algorithms are illustrated in a geographical domain using concepts from the SPACE ontology developed as part of the SWEET suite of ontologies for the Semantic Web by NASA, and are evaluated by comparing test cases of possible user misconceptions.

More information

Identifiers

Catalogue record

Export record