Linked Data Informativeness

By leveraging Semantic Web technologies, Linked Open Data provides an extensive amount of structured information in a wide variety of domains. Principles of Liked Data facilitate access and re-use of semantic information, both for human and machine consumption. However, information overload due to the availability of a large amount of semantic data, as well as the need for automatic interpretation and analysis of the Web of Data require systematic approaches to manage the quality of the published information. Identifying the value of information provided by Linked Data enables a general understanding about the significance of semantic resources. This can lead to better information filtering functionalities in Semantic Web-based applications. The aim of this paper is to propose a novel approach, derived from information theory, to measure the informativeness in the context of the Web of Data. We experiment with the metric and present its applications in a variety of areas, including Linked Data quality analysis, faceted browsing, and ranking.