NORMALIZATION

1. To avoid the duplication of value using the normalization theory. We can design a database that minimizes the redundancy, reduced integrity problem and easy to maintain.

 2. Normalization theory describes the arrangement of relation at attribute as form.

Need for Normalization:-

·         Improve database design.

·         Ensure minimum redundancy of data.

·         Reduce need to reorganize data when design is modified.

·         Removes anomalies of database activity.

Normalization using functional dependency:

                A relational database design using functional dependency. We can define several normal form that represent good database design.

Desirable properties of decomposition:-

A concept of considering lending schema

Lending schema= {Branch Name, Branch City, Asset, Customer Name,  loan number, Amount}

We can decompose into the following three relation:

1. Branch Name = {Branch Name, Branch City, Assets}

2. Loan Schema = {Branch Name, Loan Number, Amount}

3. Browser Schema = {Customer Name, Loan Number}

This decomposition has several desirable properties.

Lossless joining decomposition:-

                When decomposition a relation into a number of smaller relation it’s crucial that the decomposition lossless.

                Let R be a relation schema and F be a set of functional dependency on R. Let r1,r2,…rn form a decomposition of R. The following functional dependence on

                                R1 ⋃ R2 -> R1

                                R1 ⋂ R2 ->R2

Dependency preservation relation:

                There is another goal in the relational database design dependency preservation  when a update is made to database system should be able to check that update will not create an illegal relation.

Example: If he have to check update efficiency we should design relation database schema that allow update valuation without computation of joining.

Repetition of information:

                In the above lending schema it was necessary to respect the city and assets for a branch for each loan. The decomposition separate branch and loan data into listing relation their by eliminating redundancy clearly a lack of redundancy exited our decomposition is desirable.