The formal classifications describing the level of database 
normalization in a data model are called Normal Forms (NF) 
and the process of doing this is Normalization.

First normal form (1NF) lays the groundwork for an 
organised database design:

•	Ensure that each table has a primary key: minimal 
set of attributes which can uniquely identify a record.

•	Eliminate repeating groups (categories of data 
which would seem to be required a different number of times 
on different records) by defining keyed and non-keyed 
attributes appropriately.

•	Atomicity: Each attribute must contain a single 
value, not a set of values.

•	'First normal form' depends on functional 
dependency formula f(x)=y. For every value of x there is 
value for y.

Second normal form (2NF) If a table has a composite key, 
all attributes must be related to the whole key:

•	The database must meet all the requirements of the 
first normal form.

•	The relational schema should not have any partial 
functional dependency i.e. No proper subset of the primary 
key should derive a functional dependency belonging to the 
same schema. For example, consider functional dependencies 
FD:{AB->C, A->D, C->D} here AB is the primary key, as A->D 
this relational schema is not in 2NF.

Third normal form (3NF) requires that data stored in a 
table be dependent only on the primary key, and not on any 
other field in the table.

•	The database must meet all the requirements of the 
first and second normal form.

•	All fields must be directly dependent on the 
primary key field. Any field which is dependent on a non-
key field which is in turn dependent on the Primary Key (ie 
a transitive dependency) is moved out to a separate 
database table.