Structural joins based on range labeling schemes are considered as one of the most important topics in studies on XML query processing. When an XML data set is updated, however, the nodes have to be relabeled in order to keep their order relationship. Costly bulk node relabeling should be avoided to allow for continuous processing of queries for dynamic XML trees that are updated often. In this paper, we propose two dynamic node labeling schemes to avoid "gap shortfalls". One is simple local relabeling scheme and the other is more sophisticated in that it uses approximate histograms that keep the statistics of the update operations. These two techniques allow node labels to be managed dynamically and locally. Experiments show that they can avoid bulk relabeling while still permitting update operations. © Springer-Verlag 2004.
CITATION STYLE
Eda, T., Sakurai, Y., Amagasa, T., Yoshikawa, M., Uemura, S., & Honishi, T. (2004). Dynamic range labeling for XML trees. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 3268, 230–239. https://doi.org/10.1007/978-3-540-30192-9_22
Mendeley helps you to discover research relevant for your work.