Normalizing XML schemas through relational databases
James Jian Qiang Lu, Sonali Renjen · 2005
The theory of normal forms is one of the most widely studied subjects in databases. It provides the basis for designing anomaly-free relational schemas (see [3] for an introduction). Recently, Wu et. al. [6] and Arenas and Libkin [1] extended the theory to XML and semistructured data by appropriately generalizing the notions of normal forms. They then developed algorithms for checking and transforming DTDs with respect to these definitions.