Intelligent Multimedia Indexing and Retrieval through Multi-source Information Extraction and Merging

Thierry Declerck, H. Saggion, H. Cunningham, J. Kuper, F. de Jong, D. Reidsma, Y. Wilks, P. Wittenburg

In: Proceeedings of IJCAI. International Joint Conference on Artificial Intelligence (IJCAI) Seiten 409-414 2003.


In this paper we present the merging component of the MUMIS project1 which combines several xml-documents into one resulting xml-document. The domain of application of the MUMIS project is soccer, and each xml-document contains information about soccer games. The input documents for the merging component contain incomplete, often erroneous, and mutually inconsistent information on the same game. By applying rules which express semantic knowledge on soccer, the merging component is able to correct errors and to solve inconsistencies between the input documents. This process results in a more complete xml-document on the same game as the input documents.

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence