Structured data is increasingly created, transmitted, and archived across the Web. One of the most important movements in this context, the Linked Data movement, promulgates the rapid online publication of large-scale structured data following three basic principles: i) the creation of a Unique Resource Identifier (URI) for each resource published, ii) the publication of structured data at the URI corresponding to the the resource, and iii) the inclusion of links relating each resource to similar online resources.
Publication of Linked Data in various formalisms is currently booming on the Web, fostered by important governmental agencies, social networks, and scientists from various domains. Unrelated communities of users are increasingly interested in querying, manipulating, and arbitrary combining Linked Data in the large to create new applications. Unfortunately, the current data infrastructures at their disposal only provide very limited and rather inefficient support to their needs.
The present proposal focuses on designing new distributed data management infrastructures to promote distributed, community-based manipulations of such heterogenous and interlinked data in the large. Two related research problems are tackled. The first problem revolves around the design and implementation of an efficient storage platform to represent, query, and distribute very large amounts of Linked Data. The second problem focuses on new abstraction mechanisms to integrate Linked Data sets a posteriori once they have already been published. The rest of this document defines the main research challenges to be tackled in this context, and outlines a research agenda to realize this vision. It follows the structure suggested in the Work Programme to describe a Career Reintegration Grant proposal.
Call for proposal
See other projects for this call