Abstract: The elaboration of semantic matching between heterogeneous sources is a fundamental step in the design of data sharing applications.A new schema matching approach，context-based complex schema matching is introduced.It takes into account data sources’ context to complete the complex concepts，and effectively reduces large spaces of possible matches.First，it extracts the complex concepts from the data sources’ informational context and adds the complex concepts to the data sources，then computes the data sources’ concept matching with matching module.Experiment shows this approach can not only avoid searching large scale of possible matches，but improve the recall and precision of schema matching.

Key words: informational context, complex concept, complex matching, context analysis

摘要： 异构数据源间的模式匹配技术是数据共享应用设计的基本步骤。提出了一种新型的模式匹配方法——基于数据源信息语境的复杂模式匹配方法，它采用考虑数据源的信息语境来计算复杂匹配，能够有效地解决复杂匹配时产生的候选匹配查询空间过大的问题。该方法首先从数据源的信息语境中抽取复杂概念，并将抽取的复杂概念添加到原有数据源中，然后通过模式匹配模块计算数据源间的概念匹配。实验表明，该方法不仅能大大降低候选匹配查询空间的规模，而且能够提高模式匹配的查全率和查准率。

关键词: 信息语境, 复杂概念, 复杂匹配, 语境分析