Abstract
This paper proposes a self-adaptive scope allocation scheme for labeling dynamic XML documents. It is general, light-weight and can be built upon existing data retrieval mechanisms. Bayesian inference is used to compute the actual scope allocated for labeling a certain node based on both the prior information and the actual document. Through extensive experiments, we show that the proposed Bayesian allocation model can practically and significantly improve the performance of the conventional fixed scope allocation models.
Original language | English |
---|---|
Title of host publication | Database and Expert Systems Applications |
Subtitle of host publication | 15th International Conference, DEXA 2004, Zaragoza, Spain, August 30-September 3, 2004. Proceedings |
Editors | Fernando Galindo, Makoto Takizawa, Roland Traunmuller |
Publisher | Springer |
Pages | 811-821 |
Number of pages | 10 |
ISBN (Electronic) | 978-3-540-30075-5 |
ISBN (Print) | 978-3-540-22936-0 |
DOIs | |
Publication status | Published - 2004 |
Event | 15th International Conference on Database and Expert Systems Applications, DEXA 2004 - Zaragoza, Spain Duration: 30 Aug 2004 → 3 Sept 2004 Conference number: 15 |
Publication series
Name | Lecture Notes in Computer Science |
---|---|
Volume | 3180 |
Conference
Conference | 15th International Conference on Database and Expert Systems Applications, DEXA 2004 |
---|---|
Abbreviated title | DEXA |
Country/Territory | Spain |
City | Zaragoza |
Period | 30/08/04 → 3/09/04 |