On space constrained set selection problems

Abstract

Space constrained optimization problems arise in a variety of applications, ranging from databases to ubiquitous computing. Typically, these problems involve selecting a set of items of interest, subject to a space constraint.We show that in many important applications, one faces variants of this basic problem, in which the individual items are sets themselves, and each set is associated with a benefit value. Since there are no known approximation algorithms for these problems, we explore the use of greedy and randomized techniques. We present a detailed performance and theoretical evaluation of the algorithms, highlighting the efficiency of the proposed solutions.

A Multidimensional Data Model with Subcategories for Expressing and Repairing Summarizability

“…Most existing MD models make the restrictive assumption of having strict and homogeneous dimensions with a single bottom-level attribute and as a result avoid having to deal with this problem (because with this highly restrictive assumption every cube view can be derived from any other pre-computed cube view). Some efforts have been made to tackle this problem [33,63] but because of the lack of support in the modeling layer, the resulting solutions are not efficient.…”

“…In [63], the problem of cube view materialization in data warehouses is treated as a variant of the basic problem of space constrained optimization. The authors explore the use of greedy and randomized techniques for this problem.…”

“…⊆ Extensive research has been done in the area of cube view selection, i.e. deciding which views to materialize [63,5,57,68]. The subcategory hierarchy of a compliant EHM dimension contains all the information required to determine a complete set of pre-computed cube views, i.e.…”