Faceted fusion of RDF data

Abstract

With the rapid growth of resource description framework (RDF) data, it shows a steady trend of data decentralization and fragmentation. This trend results in two consequences. First, data decentralization and fragmentation easily leads to a narrow understanding of specific topic for users, because of difficulty in obtaining fully-faceted RDF data on the topic. Second and more importantly, users need to exert considerable effort to in searching for the RDF data of interest. In this paper, we propose a novel approach called Faceted Fusion of RDF data (FF) to solve these challenges. This method aggregates distributed RDF data on a specific topic according to different facets, based on two topological properties of a topic-specific RDF Graph (TRG). FF first constructs TRGs from RDF datasets and then discovers a set of facets by leveraging these two topological properties. We conduct three experiments over six RDF datasets to evaluate our approach. The experimental results indicate that FF outperforms three other approaches based on data distance, structure distance and structure/attribute respectively.