Databases Reference
In-Depth Information
Chapter 13
A Comparison of
Three Systems
Modeling
Methodologies
Michael P. Kushner, II-Yeol Song, and
Kyu-Young Whang
O
.
BJECT
MODELING
METHODOLOGIES
ARE
BECOMING
INCREASINGLY
POPULAR
There are many different methodologies available to model a problem. A
question that arises for many analysts is how do these techniques differ.
This chapter compares three object methodologies: entity-relationship (ER)
modeling, Shlaer and Mellor's Information Modeling, and Rumbaugh's
Object Modeling Technique (OMT). Two scenarios are used to illustrate
the methodology's syntax and to aid in the analysis of the techniques.
DESCRIPTION OF THE METHODOLOGIES
Entity-Relationship Modeling
The Entity Relationship model was first proposed by Peter Chen in 1976
and has been expanded to include the concepts of participation con-
straints, generalization/specialization, and aggregation. The Entity Rela-
tionship model is primarily used to design databases. It emphasizes that
the data model closely map to the real world scenario being modeled.
There are many variations of notations for an ER diagram. In this article
Elmasri and Navathe's definitions and notations are used except for the
notation for total/partial participation. The basic components of this
model are the entity, its attributes, and the relationships between entities.
An entity is a thing that independently exists in the real world; it is usually
a noun. It is modeled as a rectangle with the entity name inside the rectan-
gle (see Exhibit 1). Entities also require more than one attribute and more
than one instance to be a meaningful entity type in an application domain.
Search WWH ::
Custom Search