Module 6

Relational Database Design

Topics to be covered
 Pitfalls in relational database design
 Functional dependencies
 Armstrong Axioms
 Decomposition
 Desirable properties of decomposition
 Boyce-code normal form
 3rd and 4th normal form
 Mention of other normal forms

12/15/22 Module 6 2
Evaluating relation schemas
 Two levels of relation schemas
 The logical or conceptual view
 How users interpret the relation schemas and
the meaning of their attributes.
 Implementation or storage view
 How the tuples in the base relation are stored
and updated.

12/15/22 Module 6 3
Informal Design Guidelines for
Relational Databases
 Four informal measures of quality for
relation schema design are:
1. Imparting clear semantics to attributes in
Relations
2. Reducing the redundant values in tuples
3. Reducing the null values in tuples
4. Disallowing the possibility of generating
spurious tuples.

12/15/22 Module 6 4
1.Semantics of the Relation
Attributes
 GUIDELINE 1: Informally, each tuple in a relation
should represent one entity or relationship instance.
(Applies to individual relations and their attributes).
 Attributes of different entities (EMPLOYEEs,
DEPARTMENTs, PROJECTs) should not be mixed
in the same relation
 Only foreign keys should be used to refer to other
entities
 Entity and relationship attributes should be kept
apart as much as possible.
 Bottom Line: Design a schema that can be explained
easily relation by relation. The semantics of
attributes should be easy to interpret.
12/15/22 Module 6 5
A Simplified COMPANY
relational database schema

12/15/22 Module 6 6
Two relation schemas suffering
from update anomalies
EMP_DEPT

ENAME SSN BDATE ADDRESS DNUMBER DNAME DMGRSSN

EMP_PROJ

PLOCATI
SSN PNUMBER HOURS ENAME PNAME
ON

12/15/22 Module 6 7
Two relation schemas suffering
from update anomalies
 Although there is nothing wrong logically with
these 2 relations, they are considered poor
designs because they violate guideline 1 by
mixing attributes from distinct real world entities.
 EMP_DEPT mixes attributes of employee and
department and EMP_PROJ mixes attributes of
employees & projects and the WORKS_ON
relationship
 They may be used as views but they cause
problems when used as base relations

12/15/22 Module 6 8
2.Redundant Information in
Tuples and Update Anomalies
 Goal of schema design is to minimize the
storage space used by the base relations.
 Informationis stored redundantly
 Wastes storage

 Causes problems with update anomalies

 Insertion anomalies
 Deletion anomalies
 Modification anomalies

12/15/22 Module 6 9
Two relation schemas suffering
from update anomalies

EMP_DEPT

ENAME SSN BDATE ADDRESS DNUMBER DNAME DMGRSSN

EMP_PROJ

PLOCATI
SSN PNUMBER HOURS ENAME PNAME
ON

12/15/22 Module 6 10
EXAMPLE OF AN INSERT
ANOMALY
 Consider the relation:
 EMP_PROJ(Emp#, Proj#, Ename, Pname,
No_hours)
 Insert Anomaly:
 Cannot insert a project unless an employee is
assigned to it.
 Conversely
 Cannot insert an employee unless an he/she is
assigned to a project.

12/15/22 Module 6 11
EXAMPLE OF AN DELETE
ANOMALY
 Consider the relation:
 EMP_PROJ(Emp#, Proj#, Ename, Pname,
No_hours)
 Delete Anomaly:
 When a project is deleted, it will result in deleting
all the employees who work on that project.
 Alternately, if an employee is the sole employee
on a project, deleting that employee would result
in deleting the corresponding project.

12/15/22 Module 6 12
EXAMPLE OF AN UPDATE
ANOMALY
 Consider the relation:
 EMP_PROJ(Emp#, Proj#, Ename, Pname,
No_hours)
 Update Anomaly:
 Changing the name of project number P1
from “Billing” to “Customer-Accounting”
may cause this update to be made for all
100 employees working on project P1.

12/15/22 Module 6 13
12/15/22 Module 6 14
Guideline to Redundant Information
in Tuples and Update Anomalies
 GUIDELINE 2:
 Design a schema that does not suffer from the
insertion, deletion and update anomalies.
 If there are any anomalies present, then note
them so that applications can be made to take
them into account.
 In general, it is advisable to use anomaly free
base relations and to specify views that
include the joins for placing together the
attributes frequently referenced in important
queries.

12/15/22 Module 6 15
Problems with Nulls
 If many attributes are grouped together
as a fat relation, it gives rise to many
nulls in the tuples.
 Waste storage
 Problems in understanding the
meaning of the attributes
 Difficult while using Nulls in aggregate
operators like count or sum

12/15/22 Module 6 16
3. Null Values in Tuples
 Interpretations of nulls:
 Attribute not applicable or invalid
 Attribute value unknown (may exist)
 Value known to exist, but unavailable
 GUIDELINE 3:
 Relations should be designed such that their
tuples will have as few NULL values as possible
 Attributes that are NULL frequently could be
placed in separate relations (with the primary key)
 Example:-
 if only 10% of employees have individual offices, it is
better not to include office_number as an attribute in the
employee relation.
 Better create a new relation emp_offices(essn,
office_number)
12/15/22 Module 6 17
Example of Spurious Tuples

12/15/22 Module 6 18
Generation of spurious tuples
 The two relations EMP_PROJ1 and EMP_LOCS as
the base relations of EMP_PROJ, is not a good
schema design.
 Problem is if a Natural Join is performed on the
above two relations it produces more tuples than
original set of tuples in EMP_PROJ.
 These additional tuples that were not in EMP_PROJ
are called spurious tuples because they represent
spurious or wrong information that is not valid.
 This is because the PLOCATION attribute which is
used for joining is neither a primary key, nor a
foreign key in either EMP_LOCS AND EMP_PROJ1.

12/15/22 Module 6 19
Example of Spurious Tuples contd

12/15/22 Module 6 20
4. Spurious Tuples
 Bad designs for a relational database may result
in erroneous results for certain JOIN operations
 The "lossless join" property is used to
guarantee meaningful results for join operations

 GUIDELINE 4:
 Design relation schemas so that they can be
joined with equality conditions on attributes
that are either primary keys or foreign keys in
a way that guarantees that no spurious tuples
are generated.

12/15/22 Module 6 21
Spurious Tuples
 There are two important properties of
decompositions:
 Non-additive or losslessness of the corresponding join
 Preservation of the functional dependencies.

 Note that:
 Property (a) is extremely important and cannot be
sacrificed.
 Property (b) is less stringent and may be sacrificed.

12/15/22 Module 6 22
Summary and Discussion of
Design Guidelines
Problems pointed out:
 Anomalies cause redundant work to be done
during
 Insertion
 Modification
 Deletion
 Waste of storage space due to nulls and difficulty
of performing aggregation operations and joins
due to null values
 Generation of invalid and spurious data during
joins on improperly related base relations.

12/15/22 Module 6 23
Functional dependencies
 Functional dependencies (FDs)
 Isa constraint between two sets of
attributes from the database.
 Assumption
 The entire database is a single universal
relation schema R={A1,A2…An}
 Where A1,A2 … are the attributes.

12/15/22 Module 6 24
Definition
 FDs are:
 used to specify formal measures of the
"goodness" of relational designs
 keys that are used to define normal forms for
relations
 constraints that are derived from the meaning and
interrelationships of the data attributes
 A set of attributes X functionally determines
a set of attributes Y if the value of X
determines a unique value for Y

12/15/22 Module 6 25
Functional Dependencies
 A functional dependency, X -> Y holds if whenever two tuples
have the same value for X, they must have the same value for Y
 For any two tuples t1 and t2 in any relation instance r(R): If
t1[X]=t2[X], then t1[Y]=t2[Y]
 X -> Y in R specifies a constraint on all relation instances r(R)
 This means that the values of the Y component of a tuple in r
depend on, or are determined by, the values of the X
component
 The values of the X component functionally determines the
values of Y component.
 FDs are derived from the real-world constraints on the
attributes
 The main use of FD is to describe R by specifying constraints
on its attributes that must hold at all times.

12/15/22 Module 6 26
Lakes of the world
Name Continent Area length
Caspian Sea Asia-Europe 143244 760
Superior NA 31700 350
Victoria Africa 26828 250
Aral Sea Asia 24904 280
Huron NA 23000 206
Michigan NA 22300 307
Tanganyika Africa 12700 420

Continent ->Name Name ->Length

12/15/22 Module 6 27
Graphical representation of
Functional Dependencies

12/15/22 Module 6 28
Examples of FD constraints
 Social security number uniquely determines
employee name
 SSN -> ENAME
 Project number uniquely determines project
name and location
 PNUMBER -> {PNAME, PLOCATION}
 Employee ssn and project number uniquely
determines the hours per week that the
employee works on the project
 {SSN, PNUMBER} -> HOURS

12/15/22 Module 6 29
Examples of FD constraints
 A FD is a property of the attributes in the schema
R, not of a particular legal relation state r of R.
 It must be defined explicitly by someone who
knows the semantics of the attributes of R.
 The constraint must hold on every relation
instance r(R)
 If K is a key of R, then K functionally determines
all attributes in R
 (since we never have two distinct tuples with
t1[K]=t2[K])

12/15/22 Module 6 30
Satisfies algorithm
 Why it is used? To determine whether a relation
r satisfies or does not satisfy a given functional
dependency A  B.
 How it works?
 Sort the tuples of the relation r on the A attributes so
that tuples with equal values under A are next to each
other
 Check that tuples with equal values under attributes A
also have equal values under B
 If it meets the condition 2 then the output of the
algorithm is true else it is false

12/15/22 Module 6 31
Relation state of TEACH
TEACHER -> COURSE
 TEACH
TEXT -> COURSE
TEACHER COURSE TEXT

Teacher Course Text

Smith Data Bartram

Structures
Smith Data Martin
Management
Hall Compilers Hoffmann
Brown
12/15/22
ooad
Module 6
Horowitz 32
Drawbacks of Satifies algorithm
 Using this algorithm is tedious and time
consuming
 So inference axioms are used.

12/15/22 Module 6 33
Inference Rules for Functional
Dependencies
 F is the set of functional dependencies
that are specified on relation schema R.
 Schema designers specifies the most
obvious FDs.
 The other dependencies can be inferred
or deduced from FDs in F.

12/15/22 Module 6 34
Example of Closure
 Department has one manager (DEPT_NO ->
MGR_SSN)
 Manager has a unique phone number
(MGR_SSN->MGR_PHONE) then these two
dependencies together imply that (DEPT_NO-
>MGR_PHONE)
 This defines a concept called as closure that
includes all possible dependencies that can
be inferred from the given set F.
 The set of all dependencies that include F as
well as all dependencies that can be inferred
from F is called the closure of F denoted by
(F+)

12/15/22 Module 6 35
Example
F={SSN  {ENAME, BDATE, ADDRESS, DNUMBER},
DNUMBER {DNAME, DMGRSSN}}
 The inferred functional dependencies are
 SSN  {DNAME, DMGRSSN}
 SSN SSN
 DNUMBER  DNAME
 To determine a systematic way to infer
dependencies, a set of inference rules has to be
discovered that can be used to infer new
dependencies from a given set of dependencies.
This is denoted by F|=X  Y

12/15/22 Module 6 36
Inference Rules for FDs
 Given a set of FDs F, we can infer additional FDs that hold
whenever the FDs in F hold
 Armstrong's inference rules:
 IR1. (Reflexive) If Y subset-of X, then X -> Y
 IR2. (Augmentation) If X -> Y, then XZ -> YZ
 (Notation: XZ stands for X U Z)
 IR3. (Transitive) If X -> Y and Y -> Z, then X -> Z

 IR1, IR2, IR3 form a sound and complete set of inference rules
 By sound we mean,any dependency that we can infer
from F by using IR1 through IR3 satisfies the
dependencies in F(or) [if axioms are correctly applied they
cannot derive false dependencies]
 By complete we mean that by using IR1 through IR3
repeatedly to infer dependencies until no more
dependencies can be inferred results in complete set of
all possible dependencies that can be inferred from F
12/15/22 Module 6 37
 Inference Rules for FDs
 Some additional inference rules that are useful:
 Decomposition: If X -> YZ, then X -> Y and X -> Z
 Union or additive: If X -> Y and X -> Z, then X -> YZ
 Psuedotransitive: If X -> Y and WY -> Z, then WX -> Z

 The last three inference rules, as well as any

other inference rules, can be deduced from IR1,
IR2, and IR3 (completeness property)

12/15/22 Module 6 38
Examples
1. Given the set F={AB,CX, BXZ}
derive ACZ using the inference axioms
2. Given F={AB, CD} with C subset of
B show that F|=AD

12/15/22 Module 6 39
Redundant functional
dependencies
 Given a set F of FDs, a FD AB of F is said to
be redundant with respect to the FDs of F iff
AB can be derived from the set of FDs F-
{AB}
 Redundant FDs are extra and unnecessary and
can be safely removed from the set F.
 Eliminating redundant FDs allows us to minimize
the set of FDs.

12/15/22 Module 6 40
 Equivalence of Sets of Functional
Dependencies
For a given set F of FDs the set F+ may contain a large number of FDs.
It is desirable to find sets that contain smaller number of FDs than F and
still generate all the FDs of F+. Sets of FDs that satisfy this condition ar
said to be equivalent sets.
 A set of FD F is said to cover another set of FDs E if
every FD in E is also in F+. If every dependency in E can
be inferred from F; alternatively E is covered by F
 Two sets of FDs E and F are equivalent if E+ = F+.
Hence, equivalence means that every FD in E can be
inferred from F, and every FD in F can be inferred from E;
E is equivalent to F if both the conditions E covers F and
F covers E hold.

12/15/22 Module 6 41
Minimal Functional
Dependencies (minimal cover )
 is useful in eliminating unnecessary
functional dependencies.
 Also called as Irreducibe Set of F
 F is transformed such that each FD in it
that has more than one attribute in the
RHS is reduced to a set of FDs that have
only one attribute on the RHS.

12/15/22 Module 6 42
Minimal cover
(a) every RHS of each dependency is a
single attribute;
(b) for no X -> A in F is the set F - {X -> A}
equivalent to F; no
redundancies
(c) for no X -> A in F and proper subset Z of
X is F - {X -> A} U {Z -> A} equivalent to
F. no dependencies may be replaced by
a dependency that involves a subset
of the left hand side.

12/15/22 Module 6 43
Extraneous Attributes
 Further reduction of the size of the FDs of F by
removing either extraneous left attributes with
respect to F or extraneous right attributes with
respect to F.
 F be a set of FDs over schema R and let
A1A2B1B2.
 A1 is extraneous iff
 FΞF-{A1A2B1B2}U{A2B1B2}

12/15/22 Module 6 44
CANONICAL COVER (FC)
1. Every FD of FC is simple. RHS has one
attribute
2. FC is left-reduced
3. FC is nonredudant

12/15/22 Module 6 46
Problem
Given a set F of FDs find a cononical cover
for F
FC = {XZ, XYWP, XYZWQ, XZR}
1. FC= {XZ, XYW, XYP, XYZ,
XYW, XYQ, XZR}
2. FC = {XZ, XYW, XYP, XYQ,
XZR}

12/15/22 Module 6 47
Normal Forms Based on Primary
Keys
1. Normalization of Relations
2. Practical Use of Normal Forms
3. Definitions of Keys and Attributes
participating in Keys
4. First Normal Form
5. Second Normal Form
6. Third Normal Form

12/15/22 Module 6 48
Normalization of Relations

 2NF, 3NF, BCNF based on keys and FDs of a

relation schema (relational design by analysis)
 4NF based on keys, multi-valued dependencies :
MVDs; 5NF based on keys, join dependencies :
JDs (relational design by synthesis)
 Additional properties may be needed to ensure a
good relational design: lossless join and
dependency preservation

12/15/22 Module 6 49
Normalization of Relations

 Proposed by Codd
 Normalization:analysing the given relation based on their FDs and
primary keys to achieve the desirable properties of
 Minimizing redundancies
 Minimizing anomalies
 Provides the database designer with
 Formal framework for analyzing relation schemas based on keys
and FD
 Series of normal form tests

Normal form refers to the highest normal form condition

that it meets and indicates the degree to which it can be
normalized
12/15/22 Module 6 50
Normalization of Relations

 Lossless join or nonadditive join property: it

guarantees that the spurious tuple generation
problem does not occur with respect to the
relation schemas created after decomposition.
 Dependency preservation property: it ensures
that each functional dependency is represented
in some individual relation resulting after
decomposition.

12/15/22 Module 6 51
Practical Use of Normal Forms
 Normalization is carried out in practice so that
the resulting designs are of high quality and
meet the desirable properties
 The practical utility of these normal forms
becomes questionable when the constraints on
which they are based are hard to understand
or to detect
 The database designers need not normalize to
the highest possible normal form. (usually up to
3NF, BCNF or 4NF)
 Denormalization: the process of storing the join
of higher normal form relations as a base
relation—which is in a lower normal form

12/15/22 Module 6 52
Definitions of Keys and Attributes
Participating in Keys
 A superkey of a relation schema R = {A1, A2, ....,
An} is a set of attributes S subset-of R with the
property that no two tuples t1 and t2 in any legal
relation state r of R will have t1[S] = t2[S]

 A key K is a superkey with the additional property

that removal of any attribute from K will cause K
not to be a superkey any more.

12/15/22 Module 6 53
Definitions of Keys and Attributes
Participating in Keys
 If a relation schema has more than one key,
each is called a candidate key. One of the
candidate keys is arbitrarily designated to be the
primary key, and the others are called
secondary keys.
 A Prime attribute must be a member of some
candidate key
 A Nonprime attribute is not a prime attribute—
that is, it is not a member of any candidate key.

12/15/22 Module 6 54
First Normal Form

 Disallows composite attributes, multivalued

attributes, and nested relations; attributes
whose values for an individual tuple are non-
atomic
 Hence 1NF disallows relations within relations or
relations as attribute values within tuples
 Considered to be part of the definition of relation

12/15/22 Module 6 55
Normalization into 1NF

12/15/22 Module 6 56
Normalization into 1NF
 To achieve 1NF there are 3 techniques:
1. Remove the attribute that violates 1NF and place it in a
separate relation along with the primary key
2. Expand the key so that there will be a separate tuple in
the original relation. It has disadvantage of introducing
redundancy.
3. If max no. of values is known for an attribute than
replace each attribute with that many no. of atomic
attributes. It has disadvantage of introducing NULL
values.
 1st solution is considered the best because it does
not suffer from redundancy and it is completely
general having no limit placed on a max no. of
values.

12/15/22 Module 6 57
12/15/22 Module 6 58
Normalization nested relations into 1NF

12/15/22 Module 6 59
Additional problems from
schaum series
 Pg 178, 5.1

12/15/22 Module 6 60
Second Normal Form
 Uses the concepts of FDs, primary key
Definitions:
 Prime attribute - attribute that is member of the
primary key K
 Full functional dependency - a FD Y -> Z
where removal of any attribute from Y means the
FD does not hold any more
Examples: - {SSN, PNUMBER} -> HOURS is a full FD
since neither SSN -> HOURS nor PNUMBER -> HOURS hold
- {SSN, PNUMBER} -> ENAME is not a full FD (it is called a
partial dependency ) since SSN -> ENAME also holds

12/15/22 Module 6 61
Second Normal Form

 A relation schema R is in second normal

form (2NF) if every non-prime attribute A
in R is fully functionally dependent on the
primary key

 R can be decomposed into 2NF relations

via the process of 2NF normalization

12/15/22 Module 6 62
Normalizing into 2NF

12/15/22 Module 6 63
Conversion to 2NF
AA A

BB D
Convert to
CC

12/15/22 Module 6 64
Additional problem on 2nd
normal form
Prog_task(prog_ID, Prog_Pack_ID,
prog_Pac_name, Tot-Hours-wor)
Prog_Pack_IDProg_Pac_name
1. What is the highest normal form?
2. Transform into next highest form?

12/15/22 Module 6 65
Third Normal Form

Definition:
 Transitive functional dependency - a FD X -> Z
that can be derived from two FDs X -> Y and Y -> Z
Examples:
- SSN -> DMGRSSN is a transitive FD since
SSN -> DNUMBER and DNUMBER -> DMGRSSN hold
- SSN -> ENAME is non-transitive since there is no set
of attributes X where SSN -> X and X -> ENAME

12/15/22 Module 6 66
Third Normal Form
 A relation schema R is in third normal form (3NF) if it is
in 2NF and no non-prime attribute A in R is transitively
dependent on the primary key
 R can be decomposed into 3NF relations via the process
of 3NF normalization
NOTE:
In X -> Y and Y -> Z, with X as the primary key, we
consider this a problem only if Y is not a candidate key.
When Y is a candidate key, there is no problem with the
transitive dependency .
E.g., Consider EMP (SSN, Emp#, Salary ).
Here, SSN -> Emp# -> Salary and Emp# is a candidate
key.

12/15/22 Module 6 67
Normalization into 3NF

12/15/22 Module 6 68
Normalizing into 2NF and 3NF.

12/15/22 Module 6 69
SUMMARY

12/15/22 Module 6 70
Normalize the following relation

12/15/22 Module 6 71
Normalization into 2NF

12/15/22 Module 6 73
Normalization into 3NF

12/15/22 Module 6 75
Additional problems
 Pg 186,5.13

12/15/22 Module 6 76
Boyce-Codd normal form

12/15/22 Module 6 78
BCNF (Boyce-Codd Normal Form)
 A relation schema R is in Boyce-Codd Normal
Form (BCNF) if whenever an FD X -> A holds
in R, then X is a superkey of R
 Each normal form is strictly stronger than
the previous one
 Every 2NF relation is in 1NF
 Every 3NF relation is in 2NF
 Every BCNF relation is in 3NF
 There exist relations that are in 3NF but not
in BCNF
 The goal is to have each relation in BCNF (or
3NF)
12/15/22 Module 6 79
How is BCNF different from 3NF?

 For a FD XA the 3NF allows this

dependency in a relation if ‘A’ is a primary-
key attribute and ‘X’ is not a candidate key
 To test whether a relation is in BCNF ‘X’
must be a candidate key.
 So relation in BCNF will definitely be in
3NF but not the other way around.

12/15/22 Module 6 80
A relation TEACH that is in 3NF
but not in BCNF

12/15/22 Module 6 81
Achieving the BCNF by
Decomposition
 Two FDs exist in the relation TEACH:
 fd1: { student, course} -> instructor
 fd2: instructor -> course
 {student, course} is a candidate key for this relation
 So this relation is in 3NF but not in BCNF
 A relation NOT in BCNF should be decomposed so
as to meet the non-additive (lossless) join property,
while possibly forgoing the preservation of all
functional dependencies in the decomposed
relations.

12/15/22 Module 6 82
Achieving the BCNF by
Decomposition
 Three possible decompositions for relation TEACH
 {student, instructor} and {student, course}
 {course, instructor } and {course, student}
 {instructor, course } and {instructor, student}
 All three decompositions will lose fd1.
 We have to settle for sacrificing the functional dependency
preservation. But we cannot sacrifice the non-additivity
property after decomposition.
 Out of the above three, only the 3rd decomposition
will not generate spurious tuples after join.(and
hence has the non-additivity property).

12/15/22 Module 6 83
Lossless or lossy
decompositions
 When we decompose a relation we need to
make sure that we can recover the original
relation from the new relations that have
replaced it.
 If we can recover the original relation then
the decomposition is lossless else it is lossy.
 Example 5.11 pg 162

12/15/22 Module 6 86
Testing for lossless joins

 Lossless join algorithm

 Example 5:12

12/15/22 Module 6 87
Fourth Normal Form (4NF)
 Multi-valued dependency (MVD)
 Represents a dependency between attributes (for
example, A,B and C) in a relation, such that for each
value of A there is a set of values for B and a set of
value for C. However, the set of values for B and C
are independent of each other.
 A multi-valued dependency can be further defined as
being trivial or nontrivial.
 A MVD A->-> B in relation R is defined as being
trivial if
 B is a subset of A or
 AUB=R
 A MVD is defined as being nontrivial if neither of the
above two conditions is satisfied.

12/15/22 Module 6 89
Fourth Normal Form (4NF)
 Fourth normal form (4NF)
A relation that is in Boyce-Codd normal form
and contains no nontrivial multi-valued
dependencies.
 It is used for removing multivalued
dependency.
 In 4NF no table should contain two or
more one-to-many or many-to-many
relationships that are not directly related to
the key.
12/15/22 Module 6 90
Multivalued Dependencies and
Fourth Normal Form
 The EMP relation with two MVDs: ENAME —>> PNAME and
ENAME —>> DNAME.
 Decomposing the EMP relation into two 4NF relations
EMP_PROJECTS and EMP_DEPENDENTS

12/15/22 Module 6 91
12/15/22 Module 6 92
Fifth Normal Form (5NF)
 Join dependency
 Describes a type of dependency. For example, for
a relation R with subsets of the attributes of R
denoted as A, B, …, Z, a relation R satisfies a join
dependency if, and only if, every legal value of R
is equal to the join of its projections on A, B, …, Z.

 Lossless-join dependency
 A property of decomposition, which ensures that
no spurious tuples are generated when
relations are reunited through a natural join
operation.

12/15/22 Module 6 93
Fifth Normal Form (5NF)
 Definition:
􀁺 A relation schema R is in fifth normal form (5NF) (or
Project-Join Normal Form (PJNF)) with respect to a
set F of functional, multivalued, and join dependencies
if, for every nontrivial join dependency JD(R1, R2, ...,
Rn) in F+ (that is, implied by F), every Ri is a superkey
of R.
 In other words, A relation that has no join
dependency.

12/15/22 Module 6 94
Relation SUPPLY with Join Dependency
and conversion to Fifth Normal Form
(c) The relation SUPPLY with no MVDs is in 4NF but not in 5NF if it has
the JD(R1, R2, R3).
(d) Decomposing the relation SUPPLY into the 5NF relations R1, R2, and
R3.

12/15/22 Module 6 95
Fifth Normal Form (5NF)
 Join dependency is a very peculiar
semantic constraint that is very difficult to
detect in practical databases with
hundreds of attributes
 Hence 5NF is rarely used in practice

12/15/22 Module 6 96