Efficient Relaxed XML Path Query Matching Based on Extended Dewey Labeling Scheme

Abstract

With the rapid increasing popularity of XML for data representation, there is a lot of interest in relaxed query processing due to the flexible model characteristics of XML data. In this paper, we research the problem of XML path query relaxation. We define two relaxation operations, P-C to A-D edge relaxation and leaf-node-deletion. Two efficient algorithms are proposed to find relaxed answers with the minimal cost of relaxation. Specifically, DynamicStack, based on extended Dewey labeling scheme, calculates relaxed answers with stack data structure to compactly encode the potential maximal matches of path pattern. DynamicStack+ recursively constructs relaxed query pattern with leaf-node-deletion using DynamicStack to get approximate answers. Finally, extensive experiments are performed and results show that our proposed algorithms are effective to find approximate answers.