Abstract: With the development of sematic Web, the need of ontology is also growing. However, nowadays most of the data which is stored in relational databases can not be used by sematic Web application. Therefore it is a problem to be solved how to transform the data in relational database to the data that can be used by sematic Web. In consider of these issues, an automatic method of constructing OWL ontologies from relational database is proposed in this paper. The proposed method divides ontology construction process into two parts：semantic finding and ontology mapping. And a graph called M-Graph is defined. With the help of the generating and analysis of the M-Graph combined with some mapping rules, the OWL ontology is automatically constructed. Experiment proves that the proposed method can constructe OWL ontology automatically, and the result ontology is more extensive and accurate.

摘要： 随着语义网的发展，对本体的需要也越来越大。但是目前大多数的数据被存储在关系数据库中，这些数据不能被语义网应用程序所访问。因此如何将关系数据中的数据转化为可以被语义网应用的数据，是一个需要解决的问题。针对上述问题，提出了一种从数据库自动地构建OWL本体的方法。该方法把本体构建过程分为语义信息的发现和本体映射两个部分。并且定义了一种图M-Graph，通过M-Graph的生成与分析，结合映射规则自动地构建OWL本体。实验验证，该方法可自动地由关系数据库构建OWL本体，并且可以得到相对丰富和准确的本体。