Search CALO Publications:

Author and Title
Michalowski, Martin, José Luis Ambite, Snehal Thakkar, Rattapoom Tuchinda, Craig A. Knoblock, and Steve Minton. Retrieving and Semantically Integrating Heterogeneous Data from the Web. IEEE Intelligent Systems, 19(3), 2004.

Abstract
The Semantic Web promises seamless integration of heterogeneous data from distributed sources, letting agents (human users or automated programs) perform sophisticated and detailed analyses of this data. An agent would send a query, expressed in terms of its preferred ontology (schema), to a system that would then find and integrate the relevant data from multiple sources and return it using the agent’s ontology. Before achieving this vision, however, we must address several challenges. We need technologies to integrate data described in different ontologies, for example, as well as different types of data, such as images or structured data. In addition, a Semantic Web-based system must recognize when different objects at different sites denote the same real-world entity. Other challenges include efficiently querying distributed information and converting legacy data in traditional databases and Web sites (HTML) into more semantic representations such as RDF.

Building Finder is a running application that showcases our approach to meeting these challenges. The application integrates satellite imagery, geospatial data, and structured and semi-structured data from various online data sources using Semantic Web technologies. Users can query an integrated view of these sources and request Building Finder to accurately superimpose buildings and streets obtained from various sources on satellite imagery. The data sources integrated by Building Finder are heterogeneous not only in terms of the data, but also in terms of how the application accesses the sources.

Download