The electrical
products have brands, but the musical do not. Similarly, not all the products
are sold on shelves of physical stores, since the retailer has an e-commerce site
Figure 1. A homogeneous product dimension; (a) hierarchy schema; (b) hierarchy
domain
Note: To each category node in (a) correspond a set of element nodes in (b). This dimension is homogeneous; that
is, each element node has the same structure; their ancestors induce the same subgraph.
(A)
Product
Brand
Category
Department
All
(B)
b1 b2
d1
all
c1 c2
p1 p2 p3 p4
0 Hurtado & Gutierrez
Copyright ?© 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of
Idea Group Inc. is prohibited.
where some products are offered. Figure 3 shows the different structures mixed in
this heterogeneous version of a product dimension. The example could be turned
much more complex if we consider different categories and attributes associated to
electric products such as speakers, video systems, and so on.
In a relational database setting, an OLAP dimension can be viewed as a set of tuples,
whose attributes are the categories of the hierarchy schema. In particular, if the
Figure 2. A heterogeneous product dimension; (a) hierarchy schema; (b) hierarchy
domain
Note: This dimension is heterogeneous; that is, there are element nodes with different structures.
Pages:
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94