We consider the problem of accurately estimating the number of approximate XML answers for a given query, and propose an efficient method that (1) accurately computes selectivity estimates for each relaxed XML query, using a natural generalization of the correlated subpath tree (CST) summary structure, and (2) carefully combines these estimates by analyzing the nature of overlap between the different relaxed twig queries.
|Original language||English (US)|
|Number of pages||12|
|Journal||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|State||Published - Dec 1 2004|
All Science Journal Classification (ASJC) codes
- Theoretical Computer Science
- Computer Science(all)