Normalization & BCNF – 5 Marks Notes
1. “Boyce–Codd Normal Form is found to be stricter than Third Normal Form.” Justify.
BCNF is stricter than 3NF because every functional dependency X → Y in BCNF requires X to be a
super key, while 3NF allows non-superkey dependencies if Y is a prime attribute.
Example: R(StudentID, Course, Instructor)
FDs: (StudentID, Course) → Instructor and Instructor → Course
→ R is in 3NF but not BCNF since Instructor is not a key.
2. Prepare a Database to Illustrate BCNF
Relation: R(StudentID, Course, Instructor)
FDs: (StudentID, Course) → Instructor, Instructor → Course
Decompose:
R1(Instructor, Course)
R2(StudentID, Instructor)
Both are in BCNF since each determinant is a key.
3. Briefly Discuss about the Functional Dependency Concepts
A functional dependency (FD) is a relationship where one attribute uniquely determines another.
Example: RollNo → Name
Types: Trivial (Y ⊆ X), Non-trivial (Y ⊄ X), Transitive (A → B, B → C ⇒ A → C).
4. Develop a Database to Illustrate 3NF
Student(RollNo, Name, DeptNo, DeptName, HOD)
FDs: RollNo → Name, DeptNo; DeptNo → DeptName, HOD
Transitive dependency exists → Decompose:
R1(RollNo, Name, DeptNo) and R2(DeptNo, DeptName, HOD)
Now both are in 3NF.
5. Find Candidate Keys for R(A,B,C,D,E), F = {AC→E, B→D, E→A}
ACB■ = {A,B,C,D,E} ⇒ Candidate Key = {A,B,C}.
6. Given R(A,B,C,D,E), FDs: {CE→D, D→B, C→A}
(a) CE■ = {A,B,C,D,E} ⇒ Candidate Key = CE
(b) Highest Normal Form = 3NF (violates BCNF due to D→B)
(c) Decompose to BCNF:
R1(D,B) and R2(A,C,D,E).
8. What is Normalization? Explain the need for normalization.
Normalization is organizing data to reduce redundancy and improve integrity.
Need: Avoid anomalies, ensure consistency, reduce duplication.
Example: Student(StuID, Name, DeptID) and Department(DeptID, DeptName, HOD).
9. Compare and Contrast BCNF with 3NF
3NF: Allows dependencies if right side is prime.
BCNF: For every FD X→Y, X must be a super key.
BCNF is stricter and removes more redundancy.
10. Illustrate Multivalued Dependencies and Fourth Normal Form
A■B means A multi-determines B.
Example: Student ■ Hobby, Student ■ Skill
Decompose: R1(Student, Hobby), R2(Student, Skill)
Now both in 4NF.
11. Lossless Decomposition
R(A,B,C,D,E,F), FDs: A→BC, F→A, C→A, D→E, E→D
Decompose: R1(A,C,D), R2(B,C,D), R3(E,F,D)
Condition: Common attributes must form key in one relation.
Since D↔E, decomposition is lossless.