Efficient processing of tree distance axes in XML databases

Abstract

In recent years, XML has quickly adopted as the industry standard for storing and exchanging data, especially in the web. Along with its wide adoption and great flexibility, XML database are consumed by not only the database experts but also by many types of users, some of them are not technical people with database specialization. Therefore, there is a need of producing a more user-friendly, semantic query language to exploit XML database. On the other hand, along with the soaring popularity of XML, XML repositories has expanded rapidly over the years and the XML query language must ensure its scalability and efficiency on huge databases. Unfortunately, XPath, the W3C recommended query language XML query languages, is not so

flexible, user-friendly and efficient. In this report, we will propose some additional axes to the current XPath system to make XPath more flexible, user-friendly and easy to use. Those axes are called non-directional axes that do not depend on the XML hierarchical structure but the natural, user-centric, semantic relationships. Moreover, we also show that we can incorporate those new axes into current XPath system efficiently and without much significant overhead.