In this paper, we propose s-OLAP, a framework for supporting approximate range query evaluation on data cubes that meaningfully makes use of two innovative perspectives of OLAP research, namely dimensionality reduction and probabilistic synopses. The application scenario of s-OLAP is a networked and heterogeneous very large Data Warehousing environment where applying traditional algorithms for processing OLAP queries is too much expensive and not convenient because of the size of data cubes, and the computational cost needed to access and process multidimensional data. s-OLAP relies on intelligent data representation and processing techniques, among which: (i) the amenity of exploiting the Karhunen-Loeve Transform (KLT) for obtaining dimensionality reduction of data cubes, and (ii) the definition of a probabilistic framework that allows us to provide a rigorous theoretical basis for ensuring probabilistic guarantees over the degree of approximation of the retrieved answers, which is a critical point in the context of approximate query answering techniques in OLAP. © 2009 Springer Berlin Heidelberg.
CITATION STYLE
Cuzzocrea, A. (2009). S-OLAP: Approximate olap query evaluation on very large data warehouses via dimensionality reduction and probabilistic synopses. In Lecture Notes in Business Information Processing (Vol. 24 LNBIP, pp. 248–262). Springer Verlag. https://doi.org/10.1007/978-3-642-01347-8_21
Mendeley helps you to discover research relevant for your work.