SINBAD: Towards Structure-Independent Querying of Common Neighbors in XML Databases

XML query languages use directional path expressions to locate data in an XML data collection. They are tightly coupled to the structure of a data collection, and can fail when evaluated on the same data in a different structure. This paper extends XPath expressions with a new structure-independent, nondirectional axis called the neighborhood axis. Given a pair of context nodes, the neighborhood axis returns those nodes that are common neighbors of the context nodes in any direction. Such axis finds its usefulness in structure-independent query formulation as well as supporting relevant results computation in designindependent XML keyword search. We propose an algorithm called SINBAD that exploits the novel notion of node locality and small size of XML structure tree to efficiently determine the common neighbors of the context nodes. Our empirical study demonstrates that SINBAD, built on top of an existing path materializationbased relational storage scheme, has promising query performance.