Week 3

Question 1

What does a weak entity set reduce to in the relational schema?

Answer 1

A schema that incldues the foreign key column for the primary key of the identifying entity set

Question 2

How is a many to many relationship set represented in a schema?

Answer 2

With a schema with attributes for the primary keys of the two participating entity sets

Question 3

How are many-to-one and one-to-many relationship sets represented in the schema?

Answer 3

By adding a foreign key attribute to the side whose entities can only participate once in the relationship set

Question 4

What kind of relationship set doesn’t require a schema?

Answer 4

An identifying relationship set

Question 5

How are multivalued attributes represented in a schema?

Answer 5

By creating a separate relation with each value of the multivalued attribute mapping to a separate tuple in this new schema

Question 6

What are the two approaches for how to deal with generalisation in relational schema?

Answer 6

<ul>
<li>Form a schema for the superclass with a foreign key to the superclass in each of the subclass schemas</li>
<li>Form a schema for each lower level set with all inherited attributes</li>
</ul>

Question 7

What are the drawbacks of forming a schema for the superclass and having all the subclasses have a foreign key to that?

Answer 7

Getting information is harder, requires accessing information about two relations

Question 8

What are the drawbacks of having the information stored at each level of the inheritance in the super and sub classes?

Answer 8

Data can be stored redundantly if inheritance is not total

Question 9

What are some problems that can occur if data redundancy is present in the database?

Answer 9

<ul>
<li>Insert anamolies</li>
<li>Update anamolies</li>
<li>Delete anamolies</li>
</ul>

Question 10

Describe what it means for B to be functionally dependent on A (A->B)

Answer 10

If each possible value of A is associated with exactly one value of B

Question 11

If A->B then what are A and B?

Answer 11

A - determinant

B - dependent set

Question 12

What are functional dependencies the property of instead of instances?

Answer 12

Schemas

Question 13

Define a trivial functional dependency

Answer 13

If the dependent set is a subset of the determinant

Question 14

Define a partial functional dependency

Answer 14

If the dependent set is functionally dependent on a strict subset of the determinant

(you can remove an attribute from the determinant and the dependency still exists)

Question 15

Define a full functional dependency

Answer 15

If the dependent set is functionally dependent on the determiner but not on any strict subset of determiner

(you can’t remove an attribute from the determinant and the depedenceny will still exist)

Question 16

What do we need to say that C is transivitely dependent on A via B?

Answer 16

A->B and B->C (because we have A->C)

Question 17

Define a transitive functional dependency

Answer 17

If there is a set of attributes B that is the dependent set in one functional dependency and the determinant in another

Question 18

What is the main purpose of identifying a set of functional dependencies?

Answer 18

To specify the set of constraints that hold on a relation, so we can identify candidate keys

Question 19

Define a superkey in terms of functional dependencies

Answer 19

K is a superkey for a relational schema R if and only K->R

Question 20

Define a candidate key in terms of functional dependencies

Answer 20

K is a candidate key for R if and only if K -> R is a FULL functional dependency (no subset)

Question 21

What is the closure of F?

Answer 21

The set of all functional dependecies logically implied by F

Question 22

What is the closure of F denoted by?

Answer 22

F+

Question 23

What are the three of Armstrong’s axioms?

Answer 23

Reflexivity
Augmentation
Transitivity

Question 24

What is the reflexivity Armstrong axiom?

Answer 24

If B is a subset of A, then A -> B

Question 25

What is the augmentation Armstrong axiom?

Answer 25

If A->B then cA->cB where cA is any additional set of attributes

Question 26

What is the transitivity Armstrong axiom?

Answer 26

If A->B and B->C then A->C

Question 27

What does the Armstrong axioms being sound mean?

Answer 27

That they generate only functional dependencies that actually hold

Question 28

What does the Armstrong axioms being complete mean?

Answer 28

That they generate all functional dependencies that hold

Question 29

Describe the union closure rule

Answer 29

If A->B and A->C then A->BC holds

Question 30

Describe the decomposition closure rule

Answer 30

If A->BC holds then A->B and A->C holds

Question 31

Describe the pseudotransitivity rule

Answer 31

If A->B holds and cB->d holds then cA->d holds