The current Web of Data contains a large amount of interlinked data. However, there is still a limited understanding about the quality of the links connecting entities of different and distributed data sets. Our goal is to provide a collection of indicators that help assess existing interlinking. In this paper, we present a framework for the intrinsic evaluation of RDF links, based on core principles of Web data integration and foundations of Information Retrieval. We measure the extent to which links facilitate the discovery of an extended description of entities, and the discovery of other entities in other data sets. We also measure the use of different vocabularies. We analysed links extracted from a set of data sets from the Linked Data Crawl 2014 using these measures.
CITATION STYLE
Sarasua, C., Staab, S., & Thimm, M. (2017). Methods for intrinsic evaluation of links in the Web of Data. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10249 LNCS, pp. 68–84). Springer Verlag. https://doi.org/10.1007/978-3-319-58068-5_5
Mendeley helps you to discover research relevant for your work.