Database Design &

Schema Refinement

Professor Navneet Goyal

Department of Computer Science & Information Systems
BITS, Pilani
Topics
 Database Design Steps
 Redundancy
 Schema Refinement
 Minimizing Redundancy
 Functional Dependencies (FDs)
 Normalization using FDs
 First Normal Form (1NF)
 Second Normal Form (2NF)
 Third Normal Form (3NF)
 Boyce-Codd Normal Form (BCNF)
© Prof. Navneet Goyal, BITS, Pilani
Database Design Steps
 Requirements Analysis
 Conceptual Modeling (ER Model)
 Logical Modeling (Relational Model)
 Schema Refinement (Normalization)

© Prof. Navneet Goyal, BITS, Pilani

Redundancy
 Same information at many places in
the DB
 Problems:
 Wastage of Space
 Update Anomalies
• Update Anomaly
• Insert Anomaly
• Delete Anomaly
 Normalization is used for
“minimizing” redundancy
© Prof. Navneet Goyal, BITS, Pilani
 Update Anomalies
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
 Insert Anomaly: Cannot insert a project unless
an employee is assigned to it
Inversely - Cannot insert an employee unless he/she is
assigned to a project.

© Prof. Navneet Goyal, BITS, Pilani

 Update Anomalies
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

© Prof. Navneet Goyal, BITS, Pilani

 Solution
Decompose the relation:
EMP_PROJ ( Emp#, Proj#, Ename, Pname, No_hours)
Into the following smaller relations:
EMP (Emp#, Ename)
PROJ (Proj#, Pname)
EMP_PROJ ( Emp#, Proj#, No_hours)
 What happened to update anomalies?
 We need to find out the basis for
decomposing a relation to get rid of
update anomalies

© Prof. Navneet Goyal, BITS, Pilani

 Redundancy
 Integrity constraints, in particular functional
dependencies, can be used to identify
schemas with such problems and to suggest
refinements.
 Main refinement technique: decomposition
(replacing ABCD with, say, AB and BCD, or
ACD and ABD).
 Decomposition should be used judiciously:
 Is there a reason to decompose a relation?
 What problems (if any) does the decomposition
cause?

© Prof. Navneet Goyal, BITS, Pilani

Functional
Dependencies
 Constraints on the set of legal
relations
 Require that the value for a certain
set of attributes determines uniquely
the value for another set of attributes
 A functional dependency is a
generalization of the notion of a key

© Prof. Navneet Goyal, BITS, Pilani

 Functional Dependencies
 A functional dependency X Y holds over relation R if,
for every allowable instance r of R:
 t1  r, t2 r,  (t1) =  X(t2) implies  (t1) =  Y
X Y
(t2)
 i.e., given two tuples in r, if the X values agree, then the Y
values must also agree. (X and Y are sets of attributes.)
 An FD is a statement about all allowable instances of a
relation
 Must be identified based on semantics of application.
 Given some allowable instance r1 of R, we can check if it
violates some FD f, but we cannot tell if f holds over R!
 K is a candidate key for R means that K R
 However, K R does ©notProf.require K to be minimal!
Navneet Goyal, BITS, Pilani
Functional Dependencies
 Let R be a relation schema
  R and   R
 The functional dependency

holds on R if and only if for any legal relations r(R),
whenever any two tuples t1 and t2 of r agree on the
attributes , they also agree on the attributes . That
is,
t1[] = t2 []  t1[ ] = t2 [ ]
 Example: Consider r(A,B ) with the following instance
of r.
1 4
1 5
3 7
 On this instance, A  B does NOT hold, but B  A does
hold.

© Prof. Navneet Goyal, BITS, Pilani

Functional Dependencies
AB A’s B’s

u
If t & u Then
agree they must
here agree
here

© Prof. Navneet Goyal, BITS, Pilani

 Functional Dependencies
 K is a superkey for relation schema R if and only if K  R
 K is a candidate key for R if and only if
 K  R, and
 for no   K,   R
 Functional dependencies allow us to express constraints
that cannot be expressed using superkeys. Consider the
schema:
bor_loan = (customer_id, loan_number, amount )
We expect this functional dependency to hold:
loan_number  amount
but would not expect the following to hold:
amount  customer_name

© Prof. Navneet Goyal, BITS, Pilani

Functional Dependencies
 A functional dependency is trivial if it
is satisfied by all instances of a
relation
 Example:
• customer_name, loan_number 
customer_name
• customer_name  customer_name
 In general,    is trivial if   

© Prof. Navneet Goyal, BITS, Pilani

Functional Dependencies
 Consider the relation:
PLOTS (prop#, state, plot#, area, price, Tax_rate)
Information about plots available in India. The constraints
on the relation are:
 Prop# is unique throughout India
 Plot# are unique within a given state
 For a given_state, tax_rate is fixed
 Plots having the same area have the same price, irrespective
of the state in which they are located
 Write all the FDs on the relation
PLOTS

© Prof. Navneet Goyal, BITS, Pilani

 Functional Dependencies
PLOTS

Prop# State Plot# Area Price Tax_rate

FD1 PK

FD2 CK

FD3

FD4

Identify redundancy in PLOTS

Identify update anomalies in PLOTS

© Prof. Navneet Goyal, BITS, Pilani

 Functional Dependencies
PLOTS

Prop# State Plot# Area

FD1 PK

FD2 CK

State Tax_rate Area Price

FD3 FD4

© Prof. Navneet Goyal, BITS, Pilani

Normal Forms
 Normal Forms based on PK
 2 NF
 3 NF
 Normal Forms based on CKs
 Boyce-Codd Normal Form (BCNF)
 Other Normal Forms
 4 NF (Multivalued Dependencies)
 5 NF (Join Dependencies)
 Deal with very rare practical situations

© Prof. Navneet Goyal, BITS, Pilani

 2 NF
 Based on the concept of Full FDs (FFD)
 If A & B are sets of attributes of R, B is said to
be FFD on A if AB, but no proper subset of A
determines B
 No partial dependencies on the PK
 Is PLOTS in 2NF?
 YES
 Single attribute PK
 All relations with single attribute PK are in 2
NF!!
 2 NF applies to relations with composite keys

© Prof. Navneet Goyal, BITS, Pilani

 2 NF
 A relation that is in 1NF & every
non-PK attribute is fully
functionally dependent on the PK,
is said to be in 2 NF
Remove all
1 NF 2 NF
Partial Dependencies

© Prof. Navneet Goyal, BITS, Pilani

 3 NF
 Based on the concept of transitive
dependency
 No non-PK attribute should be
transitively dependent on the PK
 Transitive Dependency
If AB & BC, then A transitively
determines C through B, provided B & C
do not determine A
 Is PLOTS in 3NF?
 NO

© Prof. Navneet Goyal, BITS, Pilani

 3 NF
PLOTS

Prop# State Plot# Area Price Tax_rate

FD1 PK

FD2 CK

FD3

FD4
Prop# transitively determines tax_rate
through state
Prop# transitively determines price
through area
© Prof. Navneet Goyal, BITS, Pilani
 3 NF
 A relation that is in 1NF & 2 NF &
no non-PK attribute is transitively
dependent on the PK, is said to be
in 3 NF
Remove all
2 NF 3 NF
Transitive Dependencies

© Prof. Navneet Goyal, BITS, Pilani

 BCNF
 Based on FDs that take into account all
candidate keys of a relation
 For a relation with only 1 CK, 3NF &
BCNF are equivalent
 A relation is said to be in BCNF if every
determinant is a CK
 Is PLOTS in BCNF?
 NO

© Prof. Navneet Goyal, BITS, Pilani

 Problem 1
 Consider the relation
R(A,B,C) with functional
dependencies ABC and CB.
• Is R in 2NF?
• Is R in 3NF?
• Is R in BCNF?

© Prof. Navneet Goyal, BITS, Pilani

 Problem 2
 For the relation R (A,B,C,D), the
Functional Dependencies are
AB, AC, AD, & BA.
 Find the candidate keys of R
 List transitive dependencies in R
(assume any CK as PK)
 Find the highest current normal
form of R

© Prof. Navneet Goyal, BITS, Pilani

Closure of a set of FDs
 Given a set of FDs F on a relation R, it may
be possible that several other FDs must
also hold for R
 For Example, R=(A,B,C) & FDs, AB & BC
hold in R, then FD AC also holds on R
 For a given value of A, there can be only
one corresponding value of B, & for that
value of B, there can be only one
corresponding value for C
 The closure of F is the set of all FDs that
can be inferred from F, & is denoted by F+

© Prof. Navneet Goyal, BITS, Pilani

Equivalent Set of FDs
 Two sets of FDs, S & T, are equivalent
if the set of relation instances
satisfying S is exactly the same as
the set of relation instances
satisfying T
 S follows from T if every relation
instance that satisfies T also satisfies
all FDs in S
 S & T are equivalent iff S follows
from T, & T follows from S

© Prof. Navneet Goyal, BITS, Pilani

Trivial, Non-trivial &
Completely Non-trivial FDs
AB
 Trivial
If B’s are a subset of the A’s
 Non-trivial
If atleast one of the B’s is not among
A’s
 Completely Non-trivial
If none of the B’s is also one of the
A’s

© Prof. Navneet Goyal, BITS, Pilani

Trivial Dependency Rule
The FD A1A2A3…AnB1B2B3…
Bm is equivalent to
A1A2A3…AnC1C2C3…Ck
where the C’s are all those B’s
that are not A’s

© Prof. Navneet Goyal, BITS, Pilani

Closure of a set of FDs
 It is not suff. to consider just the given set
of FDs
 We need to consider all FDs that hold
 Given F, more FDs can be inferred
 Such FDs are said to be logically implied by F
 F+ is the set of all FDs logically implied by F
 We can compute F+using formal defn. of FD
 If F were large, this process would be
lengthy & cumbersome
 Axioms or Rules of Inference provide
simpler technique
 Armstrong;s Axioms

© Prof. Navneet Goyal, BITS, Pilani

Inference Rules for FDs
Armstrong's inference rules:
IR1. (Reflexive) If Y  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 & complete set

of inference rules
Never generates Generate all FDs
any wrong FD that hold

© Prof. Navneet Goyal, BITS, Pilani

 Inference Rules for FDs
Some additional inference rules that are
useful:
IR4: Decomposition: If XYZ, then XY & XZ
IR5: Union: If XY & XZ, then XYZ
IR6: Psuedotransitivity: If XY & WYZ,then WXZ
 Above three inference rules, as well as any
other inference rules, can be deduced from IR1,
IR2, and IR3 (completeness property)
 Prove all the six rules (IR1 – IR6) – Use defn. of
FD & either by direct proof or proof by
contradiction

© Prof. Navneet Goyal, BITS, Pilani

Inference Rules for FDs
IR1. (Reflexive) If Y  X, then X  Y
Proof: Y  X & t1 & t2 Є some instance r of R э
t1[X]=t2[X], then t1[Y]=t2[Y] because Y  X.
IR2. (Augmentation) If X  Y, then XZ  YZ
Proof by contradiction: Assume XY holds but
XZYZ does not. Then there must exist 2 tuples
t1 & t2 э 1. t1[X]=t2[X], 2. t1[Y]=t2[Y]
3. t1[XZ]=t2[XZ] & 4. t1[YZ]≠t2[YZ]
Not possible because from 1 & 3 we deduce
5. t1[Z]=t2[Z], & from 2 & 5 we deuce
6. t1[YZ]=t2[YZ], contradicting 4

© Prof. Navneet Goyal, BITS, Pilani

Example
 R = (A, B, C, G, H, I)
F={ AB
AC
CG  H
CG  I
B  H}
 some members of F+
 AH
• by transitivity from A  B and B  H
 AG  I
• by augmenting A  C with G, to get AG  CG
and then transitivity with CG  I
 CG  HI
• By union rule

© Prof. Navneet Goyal, BITS, Pilani

Procedure for Computing F+
 To compute the closure of a set of functional dependencies F:

F+=F
repeat
for each functional dependency f in F+
apply reflexivity and augmentation rules on f
add the resulting functional dependencies to F +
for each pair of functional dependencies f1and f2 in F +

if f1 and f2 can be combined using transitivity

then add the resulting functional dependency to F +
until F + does not change any further

NOTE: We shall see an alternative procedure for this task later

© Prof. Navneet Goyal, BITS, Pilani

Closure of Attribute Sets
 Set of attributes functionally determined by X
 Closure of a set of attributes X with respect to F
is the set X+ of all attributes that are functionally
determined by X
 Algo. for computing closure: compute F+ & take
all FDs with X on the LHS & take union of the
RHS of all such FDs
 X+ can be calculated by repeatedly applying IR1,
IR2, IR3 using the FDs in F
 Both these approaches become cumbersome if F
is large & consequently F+ is larger

© Prof. Navneet Goyal, BITS, Pilani

Closure of Attribute Sets
 Given a set of attributes  define the closure of  under F
(denoted by +) as the set of attributes that are
functionally determined by  under F

 Algorithm to compute +, the closure of  under F

result := ;
while (changes to result) do
for each    in F do
begin
if   result then result := result  
end
Try to find out why this algorithm works!
Complexity of this algorithm
Can you do any better?

© Prof. Navneet Goyal, BITS, Pilani

Example of Attribute Set
Closure
 R = (A, B, C, G, H, I)
 F = {A  B, A  C, CG  H, CG  I, B  H}
 (AG)+
1. result = AG
2. result = ABCG (A  C and A  B)
3. result = ABCGH (CG  H and CG  AGBC)
4. result = ABCGHI (CG  I and CG  AGBCH)
 Is AG a candidate key?
1. Is AG a super key?
1. Does AG  R? == Is (AG)+  R
2. Is any subset of AG a superkey?
1. Does A  R? == Is (A)+  R
2. Does G  R? == Is (G)+  R
© Prof. Navneet Goyal, BITS, Pilani
Uses of Attribute Closure
There are several uses of the attribute closure algorithm:
 Testing for superkey:
 To test if  is a superkey, we compute +, and check if +
contains all attributes of R.
 Testing functional dependencies
 To check if a functional dependency    holds (or, in
other words, is in F+), just check if   +.
 That is, we compute + by using attribute closure, and
then check if it contains .
 Is a simple and cheap test, and very useful
 Computing closure of F
 For each   R, we find the closure +, and for each S  +,
we output a functional dependency   S.

© Prof. Navneet Goyal, BITS, Pilani

 Canonical Cover
 Sets of functional dependencies may have
redundant dependencies that can be inferred from
the others
 For example: A  C is redundant in: {A  B, B  C}
 Parts of a functional dependency may be redundant
• E.g.: on RHS: {A  B, B  C, A  CD} can be simplified
to
{A  B, B  C, A  D}
• E.g.: on LHS: {A  B, B  C, AC  D} can be simplified
to
{A  B, B  C, A  D}
 Intuitively, a canonical cover of F is a “minimal” set
of functional dependencies equivalent to F, having
no redundant dependencies or redundant parts of
dependencies

© Prof. Navneet Goyal, BITS, Pilani

Equivalence of Sets of FDs
 Two sets of FDs F and G are equivalent if:
- every FD in F can be inferred from G, &
- every FD in G can be inferred from F
 Hence, F and G are equivalent if F+=G+
Definition: F covers G if every FD in G can be
inferred from F (i.e., if G+F+)
 F and G are equivalent if F covers G and G
covers F
 There is an algorithm for checking
equivalence of sets of FDs

© Prof. Navneet Goyal, BITS, Pilani

Extraneous Attributes
 Consider a set F of functional dependencies and the
functional dependency    in F.
 Attribute A is extraneous in  if A  
and F logically implies (F – {  })  {( – A)  }.
 Attribute A is extraneous in  if A  
and the set of functional dependencies
(F – {  })  { ( – A)} logically implies F.
 Note: implication in the opposite direction is trivial in
each of the cases above, since a “stronger” functional
dependency always implies a weaker one
 Example: Given F = {A  C, AB  C }
 B is extraneous in AB  C because {A  C, AB  C}
logically implies A  C (I.e. the result of dropping B from
AB  C).
 Example: Given F = {A  C, AB  CD}
 C is extraneous in AB  CD since AB  C can be inferred
even after deleting C

© Prof. Navneet Goyal, BITS, Pilani

 Testing if an Attribute is
Extraneous
 Consider a set F of functional dependencies
and the functional dependency    in F.
 To test if attribute A   is extraneous in 
1. compute ({} – A)+ using the dependencies in F
2. check that ({} – A)+ contains A; if it does, A is
extraneous
 To test if attribute A   is extraneous in 
1. compute + using only the dependencies in
F’ = (F – {  })  { ( – A)},
2. check that + contains A; if it does, A is
extraneous

© Prof. Navneet Goyal, BITS, Pilani

 Canonical Cover
 A canonical cover for F is a set of dependencies Fc such that
 F logically implies all dependencies in Fc, and
 Fc logically implies all dependencies in F, and
 No functional dependency in Fc contains an extraneous
attribute, and
 Each left side of functional dependency in Fc is unique.
 To compute a canonical cover for F:
repeat
Use the union rule to replace any dependencies in F
1  1 and 1  2 with 1  1 2
Find a functional dependency    with an
extraneous attribute either in  or in 
If an extraneous attribute is found, delete it from  

until F does not change
 Note: Union rule may become applicable after some
extraneous attributes have been deleted, so it has to be re-
applied © Prof. Navneet Goyal, BITS, Pilani
 Computing Canonical Cover
 R = (A, B, C)
F = {A  BC, B  C, A  B, AB  C}
 Combine A  BC and A  B into A  BC
 Set is now {A  BC, B  C, AB  C}
 A is extraneous in AB  C
 Check if the result of deleting A from AB  C is implied
by the other dependencies
• Yes: in fact, B  C is already present!
 Set is now {A  BC, B  C}
 C is extraneous in A  BC
 Check if A  C is logically implied by A  B and the
other dependencies
• Yes: using transitivity on A  B and B  C.
• Can use attribute closure of A in more complex cases
 The canonical cover is: A  B, B  C
© Prof. Navneet Goyal, BITS, Pilani
Problems with
Decompositions
There are three potential problems to
consider:
 Some queries become more expensive
• e.g., What is the price of prop# 1?
 Given instances of the decomposed relations, we
may not be able to reconstruct the corresponding
instance of the original relation!
• Fortunately, not in the PLOTS example
• How we could say this?
 Checking some dependencies may require joining
the instances of the decomposed relations.
• Fortunately, not in the PLOTS example
• How we could say this?
Tradeoff: Must consider these issues vs. redundancy
© Prof. Navneet Goyal, BITS, Pilani
 Lossy Decomposition
A B C A B Note that we
1 2 3 1 2 can never
get anythng
4 5 6 4 5 less than the
7 2 8 7 2 original
relation
A B C B C
1 2 3 2 3 Since we
4 5 6 5 6
don’t know
which tuples
7 2 8 2 8 are spurious
and which
Spurious 1 2 8 are genuine,
Tuples 7 2 3 JOIN we have
indeed lost
information
© Prof. Navneet Goyal, BITS, Pilani
Lossy Decomposition
S# Status City
S3 30 Paris
S5 30 Athens

S# Status S# City
1
S3 30 S3 Paris
S5 30 S5 Athens

2 S# Status Status City

S3 30 30 Paris
S5 30 30 Athens

© Prof. Navneet Goyal, BITS, Pilani

 Lossless Decomposition
 Observe that S satisfies the FDs:
 S#  Status & S#  City
 It can not be a coincidence that S is equal to
the join of its projections on {S#, Status} &
{S#, City}
 Heaths’ Theorem:
Let R{A,B,C} be a relation, where A, B, & C
are sets of attributes. If R satisfies AB &
AC, then R is equal to the join of its
projections on {A,B} & {A,C}
 Observe that in 2 the FD, S#  City is lost

© Prof. Navneet Goyal, BITS, Pilani

 Lossless Decomposition

 The decomposition of R into R1, R2, …Rn is

lossless if for any instance r of R
r = R1 (r ) R2 (r ) …… Rn (r )
 We can replace R by R1 & R2, knowing that
the instance of R can be recovered from the
instances of R1 & R2
 We can use FDs to show that decompositions
are lossless

© Prof. Navneet Goyal, BITS, Pilani

Lossless Decomposition

Theorem
A decomposition of R into R1 and R2 is
lossless join wrt FDs F, if and only if at
least one of the following dependencies is
in F+:
• R1  R2  R1
• R1  R2  R2
In other words, R1  R2 forms a superkey
of either R1 or R2

© Prof. Navneet Goyal, BITS, Pilani

Dependency Preservation
 Let Fi be the set of dependencies
in F + that include only attributes
in Ri.
• A decomposition is dependency
preserving, if
(F1  F2  …  Fn )+ = F +
• If it is not, then checking updates for
violation of functional dependencies
may require computing joins, which is
expensive.

© Prof. Navneet Goyal, BITS, Pilani

 Testing for Dependency
Preservation
 To check if a dependency    is preserved in a decomposition of
R into R1, R2, …, Rn we apply the following test (with attribute
closure done with respect to F)
 result = 
while (changes to result) do
for each Ri in the decomposition
t = (result  Ri)+  Ri
result = result  t
 If result contains all attributes in , then the functional dependency
   is preserved.
 We apply the test on all dependencies in F to check if a
decomposition is dependency preserving
 This procedure takes polynomial time, instead of the exponential
time required to compute F+ and (F1  F2  …  Fn)+
© Prof. Navneet Goyal, BITS, Pilani
 Example
 R = (A, B, C )
F = {A  B, B  C}
Key = {A}
 R is not in BCNF
 Decomposition R1 = (A, B), R2 = (B, C)
 R1 and R2 in BCNF
 Lossless-join decomposition
 Dependency preserving

© Prof. Navneet Goyal, BITS, Pilani

 4 NF
 BCNF removes any anomalies due to FDs
 Further research has led to the
identification of another type of
dependency called Multi-valued
Dependency (MVD)
 Proposed by R Fagin* in 1977
 MVDs can also cause data redundancy
 MVDs are a generalization of FDs
* R Fagin: “Multi-valued Dependencies & a new normal form for
relational databases,” ACM TODS2, No. 3 (Sept. 1977)
© Prof. Navneet Goyal, BITS, Pilani
 4 NF
 Consider the following relation HCTX:

 In relational databases, repeating groups are not allowed

Course Teacher Texts

DBS N Goyal Garcia
J P Misra Raghu

ADBS J P Misra Connolly

Garcia

© Prof. Navneet Goyal, BITS, Pilani

4 NF
 1 NF Version

CTX COURSE TEACHER TEXTS

DBS N GOYAL GARCIA

DBS N GOYAL RAGHU R
DBS J P MISRA GARCIA
DBS J P MISRA RAGHU R
ADBS J P MISRA GARCIA
ADBS J P MISRA CONNOLLY

NO FDs in this relation

© Prof. Navneet Goyal, BITS, Pilani

4 NF
Highest Normal Form?

BCNF?
CTX COURSE TEACHER TEXTS

DBS N GOYAL GARCIA

DBS N GOYAL RAGHU R
DBS J P MISRA GARCIA
DBS J P MISRA RAGHU R
ADBS J P MISRA GARCIA
ADBS J P MISRA CONNOLLY

© Prof. Navneet Goyal, BITS, Pilani

4 NF
Anomalies?

MANY!!
CTX COURSE TEACHER TEXTS

DBS N GOYAL GARCIA

DBS N GOYAL RAGHU R
DBS J P MISRA GARCIA
DBS J P MISRA RAGHU R
ADBS J P MISRA GARCIA
ADBS J P MISRA CONNOLLY

© Prof. Navneet Goyal, BITS, Pilani

4 NF
Anomalies
 New Teacher for DBS
 New Text for ADBS
 Teacher teaching DBS leaves

© Prof. Navneet Goyal, BITS, Pilani

4 NF
Points to note:
 If (c,t1,x1), (c,t2,x2) both appear, then
(c,t1,x2), (c,t2,x1) will also appear.
 Teachers and texts are completely independent of
one another
 CTX has no FDs at all
 CTX is in BCNF
 Any all key relation must necessarily be in BCNF!!
 But still there is a need to normalize CTX

© Prof. Navneet Goyal, BITS, Pilani

4 NF
 Decompose CTX into CT
& TX
CT TX
COURSE TEXT
COURSE TEACHER
DBS GARCIA
DBS N GOYAL
DBS RAGHU R
DBS J P MISRA
ADBS GARCIA
ADBS J P MISRA
ADBS CONNOLLY

© Prof. Navneet Goyal, BITS, Pilani

4 NF
 Decompose CTX into CT & TX is not done
on the basis of FDs (as there are no FDs)
 Decompose CTX into CT & TX is done on
the basis of MVDs
 MVDs
Represents a dependency between attributes of a
relation, such that for every value of A, there is a
set of values of B & a set of values of C, The set of
values for B & C are independent of each other
course  teacher (course multi-determines teacher)

course  text (text multi-dependent on course)

© Prof. Navneet Goyal, BITS, Pilani

 4 NF
 Interpretation of course  teacher
 Although a course does not have a single
corresponding teacher, i.e. the FD course
teacher does not hold
 Still each course must have a ‘well defined’ set
of teachers
 For a given course c and a given text x, the set
of teachers t matching the pair (c,x) depends
on value of c alone
 It makes no difference which particular value
of x we choose
Interpret course  text analogously
© Prof. Navneet Goyal, BITS, Pilani
 4 NF
Formal Definition
 Let R be a relation and A,B,C be subsets of attributes
of R, then we say that
A  B
iff, in every possible legal value of R, the set of B
values matching a given (A,C) pair depends only on
the value of A and is independent of the C value.
 It can be easily shown that for R(A,B,C), the MVD A
 B hold iff the MVD A  C also holds.
 MVDs always go together in pairs and we write them
as
A  B | C
course  teacher | text

© Prof. Navneet Goyal, BITS, Pilani

 4 NF
Fagin Theorem
 Let R(A,B,C) be a relation where A,B,C and
be subsets of attributes of R, then R is equal
to the join of its projections on {A,B} and
{A,C} iff R satisfies the MVD
A  B | C

© Prof. Navneet Goyal, BITS, Pilani

4 NF
 An MVDs A  B is trivial if
(a) B  A or
(b) A U B = R
 A relation that is in BCNF & contains no
non-trivial MVDs is said to be in 4NF
 CTX is not in 4NF because course  teacher
is a non trivial MVD

© Prof. Navneet Goyal, BITS, Pilani

Multi-Valued Dependencies
 Most common source of redundancy
in BCNF schemas is to put 2 or more
M:M relationships in a single relation

© Prof. Navneet Goyal, BITS, Pilani

Formal Definition of MVD
 The MVD
A1A2….An  B1B2…Bm
holds for a relation R if
for each pair of tuples t & u that
agree on As, we can find a tuple v
that agrees
1. With t & u on As
2. With t on Bs
3. With u on all attributes of R that are
not among As & Bs

© Prof. Navneet Goyal, BITS, Pilani

MVD
A  B
A’s B’s Others

© Prof. Navneet Goyal, BITS, Pilani

Problem Solving
 Consider a relation R (A,B,C,D,E,F)
with the following FDs:
F = {ABC, BCAD, DE, CFB}
(a) Find out whether AB is a key of R or not.
(b) Use the result of part (a) to find out
whether ABD is implied by F

AB+={ABCDE}

If D is in AB+, then ABD is implied by F

© Prof. Navneet Goyal, BITS, Pilani

Q&A
Thank You