Databases Reference
In-Depth Information
Table 14.2
An Interpretation of Measurements
Metrics
tends to 0 when
¼
tends to 0,5 when
¼
tends to 1 when
¼
RvsE
No relationships or very
few relationships
2,5 relationships per
entity
Very many relationships per
entity
DA
No derived attributes
Half of attributes are
derived
All attributes except one are
derived
CA
No composite
attributes
Half of attributes are
composite
All attributes are composite
RR
No redundant
relationships
Half of relationships are
redundant
All relationships are redundant
(impossible in practice)
M:NRel
No M:N relationships
Half of relationships are
M:N
All relationships are M:N
IS_ARel
Each subtype has about
one parent
All IS_A hierarchies are
binary trees
Very bushy tree: a complex
hierarchy with multiple
inheritance
Table 14.3
Values of the Metrics for the Example in Figure 14.5
Metrics
Values
RvsE
0.5357
DA
0.0740
CA
0.1071
RR
0
M:NRel
0.0666
IS_ARel
0.2975
The Kiviat diagram shown in Figure 14.6 is a graphical representation
of the values of the metrics shown in Table 14.3 This diagram is useful
because it allows designers to evaluate the overall complexity of an E/R
schema at a glance. It also serves to compare different conceptual schemas
and then to improve their quality.
