WP2: DMI Architecture Research

Research Aims

The ADMIRE infrastructure will provide a core set of services on two levels: high-level services with a description relating to specific user domains to allow abstract workflows to be formed, and low-level services with a description to allow automatic discovery of all of the processes that together can realise the abstract process.

The high level description uses the domain ontology from WP1, while the low level description uses the platform ontology from WP1. To realise the actual processes, WP1 will provide the operational ontology, which will be used to map between the two levels. The generic nature of DMI services, together with their semantic description, will allow an architecture that is flexible and can be easily tailored for use in a specific domain.

ADMIRE WP2 has the following objectives:

  1. Develop a definition of the ADMIRE architecture to support DMI
  2. Validate the architecture through a series of prototypes
  3. Provide the design and definitions to work packages WP3, WP4, and WP5
  4. Publish and encourage wider adoption of the architecture

...making data mining easier

Key Reports

  • Deliverable report D2.1, the ADMIRE Project, Aug 2008. ADMIRE - Architecture. Ally Hume, Liangxiu Han, Jano van Hemert, and Malcolm Atkinson.
  • Research prototype D2.2 DMI Using a Pipeline, Feb 2009. Ally Hume, Liangxiu Han, Jano van Hemert, and Malcolm Atkinson.
  • Deliverable report D2.4, the ADMIRE Project, Aug 2009. ADMIRE - Internal report on goals, scope and draft definition of AA3. A. Hume, Liangxiu Han, J.I. van Hemert, and M.P. Atkinson.

...making data mining easier