A knowledge fusion model for Web information

Abstract

The Web was designed as an information space by Tim Berners-Lee, with the goal not only that it should be available for human reading, but also that machine would be able to participate in and help users to communicate with each other. Web information integration can only free them from the tedious tasks of finding the relevant data sources. But the most existing systems do not really integrate information. Deeper information integration, or knowledge fusion (KF), involves not only delivering the information available via the links to user, but also analyzing, and merging the information results coming from information sources by solving the result consistencies, removing duplicates, etc. In the paper, we give a formal model for knowledge fusion for XML data and the KF-based information access architecture.