The simplicity of the multidimensional model, however,
stands on some assumptions about the regularity of data which are unnatural in
many applications. In this chapter, we study the implications of relaxing one of the
cores of such assumptions, namely the homogeneity of the structure of OLAP dimensions.
Structurally heterogeneous OLAP data have been reported in the OLAP
literature almost since the origins of the term OLAP itself and have concentrated
significant research work since then (Hurtado, Gutierrez, & Mendelzon, 2005; Huseman,
Lechtenborger, & Vossen, 2000; Jagadish, Lakshmanan, & Srivastava, 1999;
Kimball, 1996; Lehner, Albrecht, & Wedekind, 1998; Malinowski & Zimanyi,
2004; Pedersen, Jensen, & Dyreson, 2001).
Motivation
In the multidimensional data model, dimensions represent the perspectives upon
which data is viewed, and facts represent events that associate points of such dimensions
to measures. For example, a sale of a particular product in a particular store
of a retail chain can be viewed as a fact, which may be represented as a point in a
space whose dimensions are products, stores, and time, and can be associated with
one or more measures such as price or profit.
The phenomenon we study in this chapter is related to OLAP dimensions and, more
precisely, to their structure. The structure of a dimension is modeled as a hierarchy
of categories.
Pages:
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91