Quick Look at Normalization

Dr. Bassam Hammo

Normalization
1NF
2NF
3NF
BCNF
Decomposition
Anomalies (Review)
Anomalies are inconsistencies in data that occur due to
unnecessary redundancy.
Update anomaly
Some copies of a data item are updated, but others are
not.
Insertion anomaly
Can’t insert “real” data without also inserting unrelated
or “made up” data
Deletion anomaly
Can’t delete some data without also deleting other,
unrelated data
 Data Normalization

Primarily a tool to validate and improve a logical

design so that it satisfies certain constraints that avoid
unnecessary duplication of data.
The process of decomposing relations with anomalies
to produce smaller, well-structured relations.
Anomaly Examples
OID ODate CID CName
1 8/10/2004 2 ABC Inc
2 8/10/2004 2 ABC Inc
3 8/11/2004 3 XZY Co
4 8/11/2004 4 QWE Inc
5 8/12/2004 4 QWE Inc
Add a customer: Must also add an order

Update the name of Customer #2: Must update multiple rows

Delete Order #3: Must also delete info about Customer #3

 Functional Dependencies and Keys
Functional Dependency: The value of one attribute (the
determinant) determines the value of another attribute.
Candidate Key: A possible key.
 Each non-key field is functionally dependent on every
candidate key.
No attribute in the key can be deleted without destroying the
property of unique identification.
Functional Dependencies

EMPLOYEE
Emp_ID Last_Name First_Name M_I

Alternate way to show dependencies

Emp_ID  Last_Name, First_Name, M_I
Steps in Normalization
(doesn’t meet the
definition of a Entity
Remove multivalued &
relation) composite attributes.
First Normal Meet definition of relation
Form
Remove partial
dependencies
Second
Normal Form
Remove transitive
dependencies
Third Normal
Form
Remove any remaining
functional dependencies
Boyce-Codd
Normal Form
 First Normal Form (1NF)
If a table of data meets the definition of a relation, it is in first
normal form.
Every relation has a unique name.
Every attribute value is atomic (single-valued).
Every row is unique.
Attributes in tables have unique names.
The order of the columns is irrelevant.
The order of the rows is irrelevant.
 Second Normal Form (2NF)

1NF and no partial functional dependencies.

Partial functional dependency: when one or more non-
key attributes are functionally dependent on part of the
primary key.
Every non-key attribute must be defined by the entire
key, not just by part of the key.
If a relation has a single attribute as its key, then it is
automatically in 2NF.
 Second Normal Form (2NF)
ACTIVITY Key: Student_ID, Activity
Student_ID Activity Fee Activity  Fee
A relation
that is not
in 2NF
Fee is determined by Activity
Student_ID Activity Fee
222-22-2020 Swimming 30
232-22-2111 Golf 100
222-22-2020 Golf 100
255-24-2332 Hiking 50
 Second Normal Form (2NF)
Divide the STUDENT_ACTIVITY
relation into Student_ID Activity Key: Student_ID and Activity
two relations
that now
meet 2NF ACTIVITY_COST Key: Activity
Activity Fee Activity  Fee

Student_ID Activity Activity Fee

222-22-2020 Swimming Swimming 30
232-22-2111 Golf Golf 100
222-22-2020 Golf Hiking 50
255-24-2332 Hiking
Third Normal Form (3NF)
2NF and no transitive dependencies
Transitive dependency: a functional dependency
between two or more non-key attributes.
Third Normal Form (3NF)

HOUSING Key: Student_ID

Student_ID Building Fee Building  Fee
A relation Student_ID  Building Fee
with a
transitive
dependency
Student_ID Building Fee
222-22-2020 Dabney 1200
232-22-2111 Liles 1000
222-22-5554 The Range 1100
255-24-2332 Dabney 1200
330-25-7789 The Range 1100
 Third Normal Form (3NF)
STUDENT_HOUSING Key: Student_ID
Divide the Student_ID Building Student_ID  Building
relation into
two relations BUILDING_COST
that now Key: Building
Building Fee
meet 3NF Building  Fee

S
2
tu
2 2
d e n t_
-2 2 -2 0
ID
2 0
B u ild in g
D a b n e y Building Fee
2
2
3 2
2 2
-2 2 -2 1
-2 2 -5 5
1 1
5 4
L ile s
T h e R a n g e
Dabney 1200
2 5 5 -2 4 -2 3 3 2 D a b n e y Liles 1000
3 3 0 -2 5 -7 7 8 9 T h e R a n g e
The Range 1100
Boyce-Codd Normal Form (BCNF)
3NF and every determinant is a candidate key.
 Boyce-Codd Normal Form (BCNF)
STUDENT_ADVISOR
Primary Key: Student_ID, Major
Student_ID Advisor Major
Candidate Key: Student_ID, Advisor
Advisor  Major
A relation
where a
determinant is Student_ID Major Advisor
not a 222-22-2020 MIS Leigh
candidate key 232-22-2111 Management Gowan
222-22-2020 MIS Roberts
222-22-2111 Marketing Reynolds
255-24-2332 Marketing Reynolds
Note: Students can have a double major and have an advisor
for each major. An advisor works only with students in their
assigned area.
Boyce-Codd Normal Form (BCNF)
STUDENT_ADVISOR
Divide the Key: Student_ID, Advisor
Student_ID Advisor
relation into two
relations that
ADVISOR_MAJOR Key: Advisor
meet BCNF
Advisor Major Advisor  Major

Student_ID Advisor Advisor Major

222-22-2020 Leigh Leigh MIS
232-22-2111 Gowan Gowan Management
222-22-2020 Roberts Roberts MIS
222-22-2111 Reynolds
Reynolds Marketing
255-24-2332 Reynolds