中文摘要：

随着企业／组织中大量遗留系统的产生，遗留系统的业务逻辑提取技术也受到越来越多的关注。对于提取业务逻辑，主域变量是一个很好的入口点。在应用领域中，主域变量对应着重要的业务对象，分析主域变量的状态变化以及主域变量间的关系可以获得大部分的业务逻辑。但现今，国内外对于主域变量识别技术的研究大多只提到它的重要性，并没有提出如何提取主域变量的具体方法。给出了一种基于数据依赖图（DDG）的主域变量识别方法，并将该方法运用到一大型遗留系统的业务逻辑提取中。实践证明该方法大大提高了提取效率，并有效地降低了出错的可能性。

英文摘要：

As a large number of legacy systems in the business enterprise/organization , the legacy system＇ s business rules extraction technology has been more and more attention by public. Regarding the business rules extraction, domain variable is a good entry point. In the application area, domain variables corresponding to important business objects. Most of the business rules can be obtained from the changing of domain variables and the relationship between them. But now, at home and abroad, most of the business rules extraction only mentioned the importance of domain variable and have not certainly proposed how to extract the main domain variables . This paper gave a method to identify domain variables automatically based on DDG, and made use of this method to a large legacy systems. Practice proves that this method raises efficiency greatly, and also reduces the possibility of mistakes.