1/44
CS 4210 Chapter 5
Name | Mastery | Learn | Test | Matching | Spaced |
---|
No study sessions yet.
data model
plan or blueprint for database design
three design stages, and the one we focus on
conceptual design, logical design, and physical design. we focus on conceptual design
the er model
set of concepts and symbols to create conceptual schemas
what er model do we use now
extended er model
entity
something that can be identified and kept track of
entity class
a collection of entities of a given type
entity instance
occurence of a particular entity
attributes
an entity’s characteristics
identifiers
attributes that name/identify entities
identifier vs key
identifiers in data models become keys in database designs
composite identifiers
consists of two or more attributes
relationships
entities associated with one another
relationship classes
associations among entity classes; can involve two or more entity classes
relationship instances
associations among entity instances
degree of the relationship
number of entity classes in the relationship, binary, ternary
entities vs tables
you can express a relationship between entities without using a foreign key
maximum cardinality
the maximum # of relationship in which an entity can participate in
minimum cardinality
minimum # of relationship instances an entity must participate in
types of maximum cardinality
1:1, 1:N, N:M
who is parent and child is 1:N relationship
entity on the one side of the relationship is the parent. the one on the many side is the child
HAS-A relationship
each entity instance “has a” relationship with another entity instance
minimum cardinality types
zero (optional) , one (mandatory)
strong entity
entity that can exist on its own (basically a normal entity)
weak entity
entity whose existence depends on another entity
ID-dependent entity
an entity whos identifier includes the identifier of another entity. logical extension of the parent. minimum cardinality is always one
identifying relationship
solid line connecting an id dependent entity and its parent
nonidentifying relationship
dashed line connecting strong entities or non-id dependent weak entity to its parent
exclusive subtypes
one supertype relates to at most one subtype (student can only be grad or undergrad)
inclusive subtypes
one supertype can relate to one or more subtypes (a student can be in multiple clubs)
total subtypes
if a supertype has to be in one of the subtypes
partial subtypes
if a supertype doesnt have to be in any of the subtypes
IS-A relationships
relationship connecting supertypes and subtypes
the identifier of the supertypes and all the subtypes must be
identical
subtypes are used to avoid
value-inappropriate nulls
ID-Dependent relationship patterns
multivalued attribute pattern, archetype/instance pattern, association pattern
association pattern
includes a third entity to hold additional attributes. look at slides for examples
multivalued attribute pattern
idk like someone can have multiple phone numbers
archetype/instance pattern
when the id dependent child is the physical manifestation of an abstract parent (ie painting→print or house model→house)
patterns for mixing identifying and nonidentifying patterns
line-item pattern and others ig
look for a mixed pattern where
a strong entity has a multivalued composite attribute, and one of the elements of the composite group is an identifier of another strong entity
recursive relationship
when an entity has a relationship with itself, like a train a boxcar has a recursive relationship with the next boxcar
IDEF1X
Integrated Defined 1, Extended (national standard since 1993)
UML
Unified Modeling Language
IDEF1X relationship terms
nonidentifying connection relationships (same), identifying connection relationships (same), nonspecific relationships (N:M relationship), categorization relationships
default cardinality for identifying and non-identifying relationships
1:N, M:O