The aim of EICOS is to provide the methodology, the theoretical and modeling foundations as well as the algorithmic techniques and the necessary software architecture that will facilitate the personalization, integration, and evolution management facilities for information ecosystems that operate over a decentralized infrastructure for a large variety of data types.
The fundamental idea that will provide the means to achieve the above is the concept of dataspace, which involves the structuring of information with semantically rich meta-information, in order to achieve its management in a transparent way, independently of format, structure and origin. A dataspace provides a holistic approach to information management as it allows the data to be registered along with their meta-information. The user (or, system) actions over the dataspace are transparently mapped to actions over the data with a view to the personalization of the content depending of the current user and his context.
A dataspace facilitates the management of data independently of their format, origin and structure. The engine that facilitates the management of the dataspace is called DataSpace Support Platform (DSSP) and provides different APIs to the administrators, programmers and end-users. A DSSP provides transparent querying access to the underlying data; to achieve this, it relies on the tracing of meta-information for these data that allows the mapping of user queries (that are unaware of structural details) to specific low-level queries. An important challenge for the research community –not resolved so far- concerns the (semi)automated capturing of important meta-information for this task and its smooth integration in the dataspace.
