基于聚类分析的模糊Schema匹配算法的研究

The Disquisition of Clustering Based Opaque Schema Matching

Schema匹配是集成不同信息源数据的关键问题,本文提出一种适用于Web信息集成的Schema匹配算法-基于聚类分析的模糊Schema匹配算法,将Schema匹配问题转换为聚类分析问题,主要通过元素的数据-实例分析来实现,同时算法中也综合了元素名字匹配的结果,以提高聚类的成功率。

Schema matching is the key problem of integrating heterogeneous sources. The paper develops a schema matching method- clustering based on Opaque Schema Matching, which is applicable to integrate heterogeneous sources. The method switches the schema-matching problem to clustering question. It is mainly realized with the analysis to the element data - example, at the same time in the algorithm it also synthesizes the element name match result so as to enhances the success ratio of clustering.