Scribd
Upload
17 views

CH 14 - Basics of Functional Dependencies and Normalization For Relational Databases

The document discusses the basics of functional dependencies and database normalization. It covers topics like lossless joins, functional dependencies, different normal forms including 1NF, 2NF, 3NF and BCNF, denormalization, candidate keys, prime and non-prime attributes, and decomposition.

Uploaded by

joni
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
17 views

CH 14 - Basics of Functional Dependencies and Normalization For Relational Databases

The document discusses the basics of functional dependencies and database normalization. It covers topics like lossless joins, functional dependencies, different normal forms including 1NF, 2NF, 3NF and BCNF, denormalization, candidate keys, prime and non-prime attributes, and decomposition.

Uploaded by

joni
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

Ch 14 – Basics of Functional Dependencies and Normalization for Relational Databases

1. Frequent NULL attributes could place tuple in other relations (with primary key)
2. Reasons for NULL – Attribute not applicable, unknown, or unavailable
3. “Lossless join” guarantees meaningful results for join
4. Decomposition –Not additive or losslessness of join, preserves functional dependencies
5. Functional Dependencies – measure how good relational design is
a. Used to define normal forms for relations
b. Constraints come from meaning and interrelationship of data attributes
c. Basically, an attribute should yield the result that it intends, and breaks up one table into multiple
6. Normalization – Breaking up bad relations’ attributes into smaller relations
a. Done so designs are high quality
b. 2NF, 3NF, BCNF, 4NF, 5NF (usually 3NF and BCNF)
7. Denormalization – Storing join of higher normal form relations as base
8. Normal Form – Using keys and FDs to see if a relation schema is in a kind of normal form
9. Superkey – Set of attributes in a subset where no two tuples are in relation to superset
10. Candidate Key – Relation schema has more than one key
a. One is primary key, others are secondary keys
11. Prime Attribute – member of some candidate key
12. Nonprime attribute – nonmember of candidate key
13. Full Functional Dependency – removing an attribute of one FD makes it not a FD anymore
14. Normal Form – primary key only, multiple candidate keys, prime attribute, nonprime attributes
a. BCNF relation is in 3NF, 3NF relation is in 2NF, 2NF relation is in 1NF
b. 1NF – NO composite attributes, multivalued attributes, or nested relations
i. Attributes depend on KEY
c. 2NF – FDs, primary key
i. Attributes depend on WHOLE KEY
d. 3NF – Transitive FD - derive FD from first two FDs, only occurs when schema is 2NF and no non-prime
attribute
i. Attributes depend on NOTHING BUT THE KEY
e. BCNF (Boyce-Codd NF) – Stronger than above NFs
15. Binary Decomposition – decomposing R into two relations (needs to make sense course -> teacher, student)
16. Multivalued Dependency – R is in 4NF with dependencies F (FD & multi dependencies) F+ = closure of F

You might also like