Discovery

In the discovery theme we analyse and research how to find and use linked data on the Web. Two main activities are located in this theme, (i) the voiD vocabulary (learn more in the Infrastructure section), and (ii) research regarding unified discovery mechanisms.

Dataset Description and Discovery

Contact: Jürgen Umbrich and
Michael Hausenblas

Based on the Semantic Sitemap extension we have co-developed the Vocabulary of Interlinked Datasets (voiD). With voiD, linked datasets can be described regarding their topics, interlinking with other datasets or licensing information. Based on these descriptions, the exploration of linked datasets can be performed straight-forward and efficient. We are currently developing a the Linked Datasets Explorer (LDE), a tool allowing to explore and render linked datasets. A first, early prototype is available:

Screen-shot of LDE, the Linked Datasets Explorer, operating on voiD descriptions.

Further, we are working on the Universal Link DIScovery (ULDis) client:

Screen-shot of ULDis, the Universal Link DIScovery client.

Eventually, the semantic indexer Sindice, developed by a team in the DI2 unit can be used to discover structured data such as RDFa, microformats, etc.

Screen-shot of Sindice, the semantic indexer.