Dataspaces: Coexistence with Heterogeneity

Alon Halevy, David Maier, Michael Franklin

Dataspaces are loosely related collections of heterogeneous information. Examples of dataspaces include the enterprise, the desktop, digital libraries, government agencies, and the smart home. The goal of Dataspace Support Systems (DSSPs) is to provide several services over such collections, including querying and browsing, discovery of content and relationships between disparate items, and robust and efficient access to the information. The key aspect that distinguishes DSSPs from Information Integration systems is that they do not require reconciling semantic heterogeneity before any of these services can be provided. The goal is to offer some services immediately, and invest the effort required for semantic integration only where necessary. I will describe the dataspaces and their associated support systems, and discuss some recent work that gets us closer to realizing such systems. I will also describe where techniques from Knowledge Representation and Reasoning will be critical for the success of such systems.

Subjects: 11. Knowledge Representation; 10. Knowledge Acquisition

Submitted: Mar 6, 2006

This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.