Approximate retrieval of XML data with ApproXPath
ADC '08 Proceedings of the nineteenth conference on Australasian database - Volume 75
| Hi-index | 0.00 |
Metadata plays an important role in describing and proscribing data in both traditional and XML applications. This dissertation presents a framework to represent the metadata and maintain its consistency with the data in querying and restructuring XML data. The data model extension is called MetaDOM. MetaDOM separates data and metadata into different scopes and supports multiple levels of metadata (i.e. meta-metadata). The query language extension is called MetaXQuery. MetaXQuery is an extensible framework in which the special semantics of different kinds of metadata are specified as "plug-in" components. Each component is a set of simple, low-level operations. We first present the theoretical foundations of the framework. The framework shows how to utilize each metadata-specific component to retrieve, certify, sanitize, filter, group, and restructure data with metadata. We then show how to convert MetaXQuery expressions into low-level algebraic operators in MetaTAX and how to extend a native XML DBMS, namely eXist, to support the framework. The implementation judiciously reuses eXist's indexes and query evaluation engine to attain high efficiency. Finally, we setup a benchmark platform as well as a data/metadata generator to test the performance of our system for further optimization and development.