Research On The Mapping Between XML Mode And Relation Mode Based On Schema

XML is emerging as the standard for representing and exchanging data in World Wide Web. XML is simple and self-describing. In order to organize and manage XML data efficiently, the study of query and store XML data is becoming the hot spot. In all kinds of XML storage management strategies, mapping XML mode to relation mode is a viable and effective way.Up to now, most of the mode mappings just focus on the structure and content of XML documents, but they pay little attention to the Functional dependency, that causes the data redundancy. The Functional dependency plays an important role in data integrity, query performance and data completion. Research of the XML Functional dependency is the foundation for other XML database technologies. And now most of the mapping methods are based on the mode of DTD, but in application, DTD is not a XML document, which is not support the complicate data type and the users define type, it gradually replaced by XML Schema. Therefore, it is necessary to research mapping based on XML Schema.Based on the above, this paper focuses research on mapping XML mode to relation mode based on XML Schema and the improvement of XML data storage schema, which can improve query performance. After study the related work, the XML Schema and XML Function dependence are researched deeply..Pivot node is defined which is the key element of XML Schema., the formalized definition of XML Function dependency is proposed which based on XML Schema. In additional, an inference rules set of Functional dependency is inferred by using Armstrong axiom-system of database system, which is testified. Finally, an approach is studied from the semantic and structural of XML mode, a new method (SFDM) mapping XML to relations for storage is presented. Using this method, according to the pivot node and the Functional dependence, the relation tables are created. SFDM guarantees the integrity of content and structure of XML, keeps full functional dependencies, and reduces much storage redundancy. Furthermore; this paper proves that those relations mapped from XML are all in Second Normal Form (2NF), which improves the query efficiency.To verify the mapping algorithm SFDM, an example which explains the implement of SFDM and an emulation experiment which validates SFDM's feasibility and validity are presented.