Graduate Student Posters 2007

Author: Abir Qasem

In order to use the Web as a true global knowledge base, we need to be able to integrate its massive and diverse data. Although information integration techniques have seen success in the database area, they are difficult to apply in a web environment that is open, decentralized and unstructured. The Semantic Web extends the existing web with structure, and a mechanism to specify formal and shareable semantics. Therefore the Semantic Web may be a better infrastructure to apply information integration techniques. In our work, we adapt a fast algorithm (PDMS) from the information integration community, define a subset of OWL (a Semantic Web language) that is compatible with the PDMS algorithm and extend the PDMS algorithm to take advantage of additional information provided by the Semantic Web. Our solution is effective, since it queries all the sources that have relevant information, including the ones that are not immediately apparent from the query. It is also efficient, because it will ignore sources that do not have relevant information, and thus reduce the reasoning time to get an answer.