 |
Home > Section 7 > Subsection 7 (Normalisation - part 1) |    |
Learning outcomes:
1. Be aware that Normalisation is primarilly
concerned with Functional Dependency
2. Be aware that Normalisation usually involves
splitting tables
3. Be aware that there are five normal forms
indicating how structured the data is
Normalisation has its origins in mathematical set theory. It was first developed by Edger Codd, the same person who devised the Relational database, in the early 1970s. Normalisation is a process that consists of several clearly defined stages and requires an understanding of several other concepts the most important of which is functional dependency. Functional dependency forms the basis of the normalisation process beyond the first stage.
Data is said to be in first, second, third fourth or fifth normal form indicating how structured it is. Data in first normal form is said to be less normalised than that in second normal form which, in turn, is less normalised than that in third normal form etc. Data can, by default, either be in one of these 'normal forms' or be made to conform to one of them by applying a number of strategies which usually involves splitting a table up into several smaller ones.
Practically normalisation usually means the splitting up of fields from one table into several smaller ones.
The process of splitting up a table into two or more tables is technically called Projection or Projecting out.
Before discussing normalisation it is necessary to introduce the concept of composite keys.
For: Clinicians | NHS managers | Non healthcare workers
Written by Robin Beaumont & Chris Noden e-mail:robin@ieg-net.co.uk or chrisnoden@hotmail.com