Chapter 5 Flashcards
(24 cards)
What is the Extended Entity Relationship Model (EERM)?
An extension of the ER model that adds semantic constructs to capture more complex data structures and application requirements.
What is an EERD?
An Extended Entity Relationship Diagram
What is a supertype in EERM?
A generic entity type that has common attributes shared by one or more subtypes.
What is a subtype in EERM?
A specialized entity type that inherits attributes from its supertype and has additional unique attributes.
What is a specialization hierarchy?
A hierarchy organizing supertypes and subtypes
What is inheritance in EERM?
The ability of subtypes to inherit attributes and relationships from their supertype.
What is a subtype discriminator?
An attribute in the supertype that determines the appropriate subtype for each instance.
What does a disjoint constraint mean?
An entity instance can only be a member of one subtype. Represented by ‘d’ in a circle in EERDs.
What does an overlapping constraint mean?
An entity instance can belong to multiple subtypes. Represented by ‘o’ in a circle in EERDs.
What are completeness constraints?
They indicate whether every instance of a supertype must be a member of a subtype (total) or not (partial).
What is an entity cluster in an ERD?
A virtual entity used to group multiple related entities and relationships to simplify complex ERDs.
Are entity clusters implemented in the final database design?
No
What should be avoided when using entity clusters?
Displaying attributes
What is a primary key?
An attribute (or combination) that uniquely identifies each entity instance.
What are the characteristics of a good primary key?
Unique
What is a natural key?
A real-world identifier used as a primary key
When are composite primary keys useful?
For composite entities (bridge tables) or weak entities in identifying relationships.
What is a surrogate primary key?
A system-generated numeric key used when no suitable natural key exists.
How should multivalued attributes be handled in a data model?
By creating a new related entity to avoid structural problems and nulls.
How is a 1:1 recursive relationship handled?
By placing the foreign key in the entity where it causes the fewest nulls or represents the optional side.
How is a 1:M recursive relationship handled?
By placing the PK of the “one” side as a FK in the “many” side (same entity).
How is a M:N recursive relationship handled?
With a bridge table containing two foreign keys referencing the same entity.
What is time-variant data
and how is it handled?
Why might denormalization be used in database design?
To improve performance at the cost of increased redundancy and risk of anomalies.
