Interactivity, Scalability and Resource Control for Efficient KDD Support in DBMS

The conflict between resource consumption and query performance in the data mining context often has no satisfactory solution. This not only stands in sharp contrast to the need of the analysts for interactive response times, but also makes the seamless integration of data mining operators into common multiuser database systems a difficult and (so far) not very prosperous task. We believe that an efficient solution to the problems of database support for KDD has to affect the whole query processing from the data access on disk up to the complex data mining operators. The basic idea of our framework is to provide resource efficiency and interactivity through precise control over the order in which data is processed from the index structure through the whole query tree. It consists of an index that is basically an extension of the UB-Tree and allows to translate efficient data access patterns into various data orderings. Our KDD-algebra exploits these orderings to allow the control of re...

• Type: Conference
• Year: 2004
• Where: CINQ
• Authors: Matthias Gimbel, Michael Klein, Peter C. Lockemann
• Keywords: CINQ 2004, Data Mining, Data Mining Operators, Database, Interactive Response Times