SIM-NET: A View-Based Semantic Similarity Model for Ad Hoc Networks of Geospatial Databases

Semantic similarity is a fundamental notion in GIScience for achieving semantic interoperability among geospatial data. Until now, several semantic similarity models have been proposed; however, few of these models address the issues related to the assessment of semantic similarity in ad hoc networks. Also, several models are based on a definition of concepts where features are independent, an assumption that reduces the richness of the geospatial concept representation. This article presents the conceptual basis for Sim-Net, a novel semantic similarity model for ad hoc networks based on Description Logics (DL). Sim-Net is based on the multi-view paradigm. This paradigm is used to include inferential knowledge in semantic similarity, that is, the knowledge about implicit dependencies between features of concepts. In Sim-Net, assessing semantic similarity relies on the notions of Semantic Reference Systems and Formal Concept Analysis (FCA), which are combined to establish a common semantic reference frame for ontologies of the ad hoc network called the view lattice. The Sim-Net semantic similarity measure distinguishes concepts that belong to different or similar domains and takes into account the neighbours of a concept in the network. An application example is used to show the positive impact of the properties of Sim-Net.