Research On Semantic-Driven XML Normalization

Currently, XML has become standard for data representation and exchange on the Web by W3C. And XML data is increasing greatly by Web exchanging and processing. It requires a lot more to XML schema. A central goal of database design is to ensure that there are no unintended redundancies: redundant information not only takes up unnecessary storages, inflates data transfer cost, but also can lead to update anomalies. The harm of XML data anomalies is further higher than relational data as web is open. This paper firstly studies semantic constrains in XML data deeply and then designs the algorithm to normalize these XML data in order to obtain a good XML schema. Since we reduce the data redundancy and keep the data consistance, it has important theoretic siginificance and practical value to the research on XML schema normalization.In this paper, XML data normalization is investigated from semantic angle based on exsiting DTD and XML Schema specification. The contributions of this paper are as follows:①The formalized definition of XML document and the semantic contrains of XML Schema are researched. And the concepts of XML function dependency,indirect function dependency and transitive function dependency are proposed based on path and tree tuple. The logical inplication and cover definition of XML Schema function dependency are also presented in this papar.②The different level XML normal form are defined based on the definition of formalized XML function dependency, in order to ensure the standard of XML normalization, eliminate semantic ambiguity, minimize data redundancy and keep data consistency and maintanancy.③Based on above model one new XAP algorithm is designed to discover function dependency and detect data redundancy for XML. This can help detect redundancy effectively in existed XML document. It is also given to analyze its complexity.