- Combi-Operator — Database Support for Data Mining Applications

https://doi.org/10.1016/B978-012722442-8/50045-8Get rights and content

Publisher Summary

This chapter identifies the data intensive subproblem of aggregating high-dimensional data in all possible low-dimensional projections, which occurs in several established data mining techniques. It explores that existing OLAP SQL-extensions are insufficient for high-dimensional data and proposes a new SQL-operator, which seamlessly fits into the set of existing OLAP group by operators. The main drawbacks of the existing operators are (1) very large query size and (2) suboptimal performance. This chapter proposes efficient implementations for the operator, which take the limited resources of main memory into account. It demonstrates on a number of real and synthetic data sets that for the identified subproblem, the new implementations yield a large speedup over existing methods built in commercially available database systems.

References (0)

Cited by (16)

View all citing articles on Scopus
View full text