A dataflow platform for in-silico experiments based on linked data

A big part of the work carried out by scientists nowadays involves data manipulation: access and integration of different local and online datasets, filtering for relevant information, aggregating and visualising the data to look for or defend hypotheses. The Linked Data Initiative is pushing dataset maintainers to publish data in a highly reusable way through a set of open standards, such as RDF and SPARQL. The adoption of these technologies in the scientific community is still marginal, partly for the limits of the available tools for consuming and manipulating data. We present a concrete pipeline language and a working prototype (including a visual editor and a pipeline engine) by which users can build and share applications consuming and visualising linked data. © 2014 Springer International Publishing.