Statistical Analysis of the owl:sameAs Network for Aligning Concepts in the Linking Open Data Cloud

Abstract

Abstract. The massively distributed publication of linked data has brought to the attention of scientific community the limitations of classic methods for achieving data integration and the opportunities of pushing the boundaries of the field by experimenting this collective enterprise that is the linking open data cloud. While reusing existing ontologies is the choice of preference, the exploitation of ontology alignments still is a required step for easing the burden of integrating heterogeneous data sets. Alignments, even between the most used vocabularies, is still poorly supported in systems nowadays whereas links between instances are the most widely used means for bridging the gap between different data sets. We provide in this paper an account of our statistical and qualitative analysis of the network of instance level equivalences in the Linking Open Data Cloud (i.e. the sameAs network) in order to automatically compute alignments at the conceptual level. Moreover, we explore the effect of ontological information when adopting classical Jaccard methods to the ontology alignment task. Automating such task will allow in fact to achieve a clearer conceptual description of the data at the cloud level, while improving the level of integration between datasets.

sameAs.cc: The Closure of 500M owl:sameAs Statements

“…Existing algorithms for assessing whether or not two IRIs denote the same resource are currently evaluated on relatively small datasets [5]. The availability of a large dataset of real-world identity links can help quantify the utility of existing alignment algorithms such as [4].…”

Alignment and dataset identification of linked data in Semantic Web

“…), which are created using instance alignment techniques, but similar relationships between concepts and properties are not inherent. Concept alignment in LOD has been investigated to some extent in the recent past and showed significant progress both in precision and coverage, but property‐level alignment is yet to achieve considerable attention, coverage, and results other than a recent effort by Gunaratna et al…”

“…The idea of utilizing instance data for concept alignment has also been considered in the recent past and shown to be effective . Parundekar et al proposed that identifying equivalent instances belonging to concepts leads to an alignment between those concepts.…”

“…Moreover, the system is able to find one‐to‐one and composite (a set of classes making a concept/class) concept coverings. Along this line of work, Correndo et al incorporated a statistical approach utilizing owl:sameAs links between instances with the Jaccard coefficient measurement to measure the overlap of instances in aligning concepts. Nikolov and Motta also utilized owl:sameAs links to infer mappings between ontology concepts in the LOD.…”

“…Graph‐based ontology analysis approach is another kind of approach; it groups concepts and properties on several ontologies to build an upper ontology for querying underlying data. Furthermore, instance alignments are used for both querying and concept alignments . In this sense, the three types of alignments we briefly discussed are tightly coupled with interesting applications as well as among themselves.…”