3.2. Data Modeling & the Relational Database Model

Organizing Data in a Database


In today’s businesses, there are so many data elements. Thus it is critical to keep data organized,
so, that it can be used effectively.

A database should be designed to store all data relevant to the business. It should provide quick
access to the data and easy modification. It must reflect the business processes of the
organization.

When building a database, an organization must consider the following questions:

• Content: What data should be collected and at what cost? • Access: What data should be provided to which users and when? • Logical structure: How should data be arranged so that it makes sense to a given user? • Physical organization: where should data be physically located?

3.2.1. Data Modeling

Data model is a tool that the database designers use to show the logical relationships among data.
When data modeling done at a level of entire organization it is known as enterprise data modeling.



E-R diagram is an example of a data model. It uses graphical symbols to show the organization of
and relationships between data. The original ER diagram used boxes to indicate data items or
entities and diamonds to show relationships between them. However ER diagrams have evolved
and many alternate diagrammatic techniques have emerged.
ER diagram is independent of a database model. It allows end-users to visualise the data
requirements that has been captured and how it could be accessed.

3.2.2. The Relational Database Model

The relational model describes data using a standard tabular format. In a database structured
according to the relational model, all data elements are placed in two dimensional tables, called
relations which are the logical equivalent of files. The tables in relational databases organize data
in rows and columns, simplifying data access and manipulation. It is easier for managers to
understand the relational model than other database models.

In the relational model, each row of a table represents a data entity, with the columns of the table

representing attributes. Each attribute can take on only certain values. The allowable values for
these attributes are called the domain. The domain for a particular attribute indicates what values

can be placed in each of the columns of the relational table.

Example: Domain for the attribute gender- Male or female, which are formed using characters.
With the wide acceptance of object-oriented concepts the need to manage data-objects raised.
Object-oriented database model and Object-relational database models emerged as a result, The
relational database model was quite adequate to deal with traditional business applications. Thus
the object-oriented database model was required only for special purposes. Modern relational
database management systems provide some object-oriented features thus they are based on
object-relational database model.

3.2.2.1 Manipulating Data





Once data has been placed into a relational database, users can make inquiries & analyze data.
To manipulate relational databases a set of relational operators have been defined. Basic data
manipulations using relational operators include selecting, projecting & joining. Selecting involves
eliminating rows according to certain criteria.
Projecting involves eliminating columns in a table.
Joining involves combining two or more tables.

Selecting Operation



Projection Operation



Join Operation



As long as the table share at least one common data attribute, the tables in a relational database
can be linked to provide useful information and reports. One of the primary advantages of a
relational database is that it allows tables to be linked. It is easier to control, more flexible, more
intuitive than other approaches because it organizes data in tables. The ability to link relational
tables also allows users to relate data in new ways without having to redefine complex
relationships. Because of the advantages of the relational model, many companies use it for large
corporate databases, such as those for marketing and accounting. The relational model can be
used with personal computers and mainframe systems.

3.2.2.2. Data Cleanup

The characteristics of valuable data include that the data is accurate, complete, economical,
flexible, relevant, simple, timely, verifiable, accessible and secure. The purpose of data cleanup is
to develop data with these characteristics. A database can contain errors. For example, a survey
of a thousand electric-utility companies found that customer databases were only 45.6 percent
accurate. The errors were caused by inaccurate data entry.
Formalized approaches, such as database normalization are often used to clean up problems with
data.

3.2.3. Brief Comparison of Database Models

Each of the database models is used by a variety of organizations and each has advantages &
disadvantages. The primary advantage of the hierarchical model is processing efficiency. A
hierarchical database system can take less time to manipulate than other database models, as thedata relationships are less complex. This is best when the data forms a natural hierarchy.
Hierarchical models are difficult to change and the databases can be difficult to install.
Network models are more flexible than hierarchical model in data organizing aspect. Network
models are difficult to develop and use because of the complexity of the data relationships.
The relational database model is the most widely used. It is easier to control, more flexible and
more initiative than the others because it organizes data in tables. The ability to link relational
tables also allows users to relate data in new ways without having to redefine complex
relationships. Many companies use it for large corporate databases such as marketing and
accounting.