Abstract

Schema matching is a primary problem in the hot research fields such as data space, semantic web and so on. The existing methods extract the elements' own information, structure information and data instance of every element, and then choose the pair of elements with most similar semantic as matching elements. But the difference between elements in elements' own information and structure information hinder the extraction of semantic and then influence the accuracy of matching result. Through analyzing the reason of structure information difference, some kinds of structure information difference are summarized and the corresponding identification and solving algorithms are proposed in this paper. Extensive simulation experiments are conducted and the results show that the accuracy of matching result is increased by adding the identification and solving algorithms before matching.

Paper Details