Federated Content-Based Retreival Cluster Purpose: The purpose of this cluster is to exploit possible federation architecture based on a
self-describing framework. In this self-describing framework, the data schema, methods, and services are captured by the grand schema
of the system. This grand schema form the basis for resource description of the specific repository. The resources of the entire cluster may be stored centralized, distributed or a hybrid of the two approaches.
The data schema captures the entities as well as the relationships between these entities used in the repository (i.e. the entity relationship graph). The methods
are those software products that can be shared by the federation. As an example, texture feature extrations are possible candidates in this area. The services
are those mechanisms used as entry points to utilize the repository. For example, catalog service allows a remote client to examine the catalog of the repository. Search service, in which the retrieval of data based on either metadata or content description of the data, is also provided.
Participants:
IBM/Johns Hopkins University/Raytheon |
|
