Data Confidentiality on the Semantic Web

This chapter investigates the threat of unwanted Semantic Web inferences. We survey the current efforts to detect and remove unwanted inferences, identify research gaps, and recommend future research directions. We begin with a brief overview of Semantic Web technologies and reasoning methods, followed by a description of the inference problem in traditional databases. In the context of the Semantic Web, we study two types of inferences: (1) entailments defined by the formal semantics of the Resource Description Framework (RDF) and the RDF Schema (RDFS) and (2) inferences supported by semantic languages like the Web Ontology Language (OWL). We compare the Semantic Web inferences to the inferences studied in traditional databases. We show that the inference problem exists on the Semantic Web and that existing security methods do not fully prevent indirect data disclosure via inference channels.