Details

Autor(en) / Beteiligte

• Seidler, Katja
• Peukert, Eric
• Hackenbroich, Gregor
• Lehner, Wolfgang

Titel

Approximate Query Answering and Result Refinement on XML Data

Ist Teil von

• Scientific and Statistical Database Management, p.78-86

Ort / Verlag

Berlin, Heidelberg: Springer Berlin Heidelberg

Link zum Volltext

Quelle

Alma/SFX Local Collection

Beschreibungen/Notizen

• Today, many economic decisions are based on the fast analysis of XML data. Yet, the time to process analytical XML queries is typically high. Although current XML techniques focus on the optimization of query processing, none of these support early approximate feedback as possible in relational Online Aggregation systems. In this paper, we introduce a system that provides fast estimates to XML aggregation queries. While processing, these estimates and the assigned confidence bounds are constantly improving. In our evaluation, we show that without significantly increasing the overall execution time our system returns accurate guesses of the final answer long before traditional systems are able to produce output.