Browsing Data Science Institute (Workshop Papers) by Author "Umbrich, Jürgen"

Now showing items 1-5 of 5

    • DING! Dataset Ranking using Formal Descriptions 

      Toupikov, Nickolai; Umbrich, Jürgen; Delbru, Renaud; Hausenblas, Michael; Tummarello, Giovanni (2009)
      Considering that thousands if not millions of linked datasets will be published soon, we motivate in this paper the need for an efficient and effective way to rank interlinked datasets based on formal descriptions of their ...

    • Fast and Scalable Pattern Mining for Media-Type Focused Crawling 

      Umbrich, Jürgen; Karnstedt, Marcel; Harth, Andreas (2009)
      Search engines targeting content other than hypertext documents require a crawler that discovers resources identifying files of certain media types. Naive crawling approaches do not guarantee a sufficient supply of new ...

    • SWSE: Answers Before Links! 

      Harth, Andreas; Hogan, Aidan; Delbru, Renaud; Umbrich, Jürgen; O'Riain, Sean; Decker, Stefan (CEUR-WS.org, 2007)
      We present a system that improves on current document- centric Web search engine technology; adopting an entity-centric perspective, we are able to integrate data from both static and live sources into a coherent, interlinked ...

    • SWSE: Objects before documents! 

      Harth, Andreas; Hogan, Aidan; Umbrich, Jürgen; Decker, Stefan (2008)
      Web search engines are immensely useful for locating documents online. However, with more and more structured data being published online, the restriction to the hyperdocument model impairs the usefulness for searching and ...

    • Towards Dataset Dynamics: Change Frequency of Linked Open Data Sources 

      Umbrich, Jürgen; Hausenblas, Michael; Hogan, Aidan; Polleres, Axel; Decker, Stefan (CEUR, 2010)
      Datasets in the LOD cloud are far from being static in their nature and how they are exposed. As resources are added and new links are set, applications consuming the data should be able to deal with these changes. In ...