Developing a Conceptual Model › Normalizing Your Data Model

Normalizing Your Data Model

Normalization is the process of removing data redundancy and duplication from the entities and attributes of a model.

This process involves:

• Regrouping attributes
• Splitting entities
• Reassigning primary keys

Before starting the process you need to understand data relationships and familiarize yourself with definitions of terms on key dependencies.

In order to understand the relationships among data items, you must determine which attributes of an entity are dependent on the entity's other attributes. Each entity must have a unique key by which it can be uniquely identified. The key can be a single attribute or group of attributes. A key must have two properties:

• In each instance of an entity, the value of the key must uniquely identify that instance.
• If the key is composed of more than one attribute, each of the attributes must be essential to the unique identification of the entity.