Transcript 沒有投影片標題 - National University of Kaohsiung

Fundamentals of
Database Systems
Chapter 4
Enhanced Entity-Relationship
and Object Modeling
IM ISU
Database
1
Chapter Outline
 EER stands for Enhanced ER or Extended ER
 EER Model Concepts
 Includes all modeling concepts of basic ER
 Additional concepts:
»
»
»
»
subclasses/superclasses
specialization/generalization
categories (UNION types)
attribute and relationship inheritance
 These are fundamental to conceptual modeling
The additional EER concepts are used to model
applications more completely and more accurately
 EER includes some object-oriented concepts, such as
inheritance
IM ISU
Database
2
EER Model Concepts
 Extended Entity-Relationship
(EER)Model
 An extension of ER model includes the
concepts of
» subclass and superclass
» specialization and generalization
» category
» attribute and relationship inheritance
IM ISU
Database
3
EER Model Concepts (cont.)
 Subclass and Superclass
 In many cases an entity type has
numerous subgroupings of its entities
» e.g., EMPLOYEE may be grouped into
SECRETARY, ENGINEER, MANAGER,
TECHNICIAN
 Each subgroup form a subclass of the
entity type (superclass) SECRETARY
» e.g., EMPLOYEE
superclass
IM ISU
Database
ENGINEER
MANAGER
TECHNICIAN
subclass
4
EER Model Concepts (cont.)
 Class/subclass (IS-A/AN) relationship
 The relationship between a superclass
and any one of its subclasses
» e.g., EMPLOYEE/MANAGER
 An entity being a member of a subclass
must also be a member of the superclass
» e.g., a SECRETARY entity ‘Joan Logano’ is
also the EMPLOYEE ‘Joan Logano’
 A subclass entity is the same as that in the
superclass, but in a distinct specific role
IM ISU
Database
5
IM ISU
Database
6
EER Model Concepts (cont.)
 Type inheritance
 A subclass entity inherits
» all the attributes of the entity as a member of
the superclass
» all the relationships in which the superclass
participates
 A subclass should have specific attributes
and can participate in specific relationship
types
IM ISU
» e.g., SECRETARY
has attribute TypingSpeed7
Database
EER Model Concepts (cont.)
 Specialization
 The process of defining a set of
subclasses of an entity type (superclass)
based on some distinguishing
characteristic, e.g.,
» {SECRETARY, ENGINEER,
TECHNICIAN} based on job type
» {SALARIED_EMPLOYEE,
HOURLY_EMPLOYEE} based on method
of pay
IM ISU
Database
8
EER Model Concepts (cont.)
IM ISU
Database
9
EER Model Concepts (cont.)
 Reasons for using (specialization)
subclass
 Certain attributes may apply to some but
not all entities of the superclass
 Some relationship types may be
participated in only by entities that are
members of the subclass
IM ISU
Database
10
EER Model Concepts (cont.)
 Generalization
 The process of identifying the common
features among several entity types, and
generalizing them into a single superclass
» e.g., CAR and TRUCK can be generalized
into VEHICLE
 Generalization can be viewed as the
inverse of specialization
IM ISU
Database
11
ER Model Concepts (cont.)
IM ISU
Database
12
Constraints on Specialization/
Generalization
 Predicate-defined (or conditiondefined) subclasses
 Placing a condition on the value the
attribute that identify the subclasses
 e.g., SECRETARY subclass by the
defining predicate JobType = ‘Secretary’
 A constraint specifying that members of
the subclass must satisfy the predicate
IM ISU
Database
13
Constraints on Specialization/
Generalization (cont.)
IM ISU
Database
14
Constraints on Specialization/
Generalization (cont.)
Attribute-defined specialization
 Specify that all subclasses having the
membership condition on the same
attribute of the superclass
 The attribute is called the defining
attribute of the specialization, e.g., Job
type
IM ISU
Database
15
Constraints on Specialization/
Generalization (cont.)
 Disjointness constraint
 Specify that the subclasses of the
specialization must be disjoint, i.e., an
entity belongs to at most one subclass
» e.g., specialization {HOURLY_EMPLOYEE,
SALARIED_EMPLOYEE}
» denoted as
symbol
d
» If the subclasses
are not disjoint, their sets of
entities may overlap, denoted as
symbol
o
IM ISU
Database
16
Constraints on Specialization/
Generalization (cont.)
IM ISU
Database
17
Constraints on Specialization/
Generalization (cont.)
 Completeness constraint
 Total specialization constraint
» Specify that every entity in the superclass
must be a member of some subclass in the
specialization
 Partial specialization
» Allow an entity not to belong to any of the
subclasses
IM ISU
Database
18
Characteristics on Specialization/
Generalization
 Specialization Hierarchies and Lattices
 A subclass itself may have further subclasses
specified on it, forming a hierarchy or a lattice
of specializations
 For a specialization hierarchy every subclass
participates in only one class/subclass
 For a specialization lattice a subclass can be a
subclass in more than one class/subclass
IM ISU
Database
19
Characteristics on Specialization/
Generalization (cont.)
IM ISU
Database
20
Characteristics on Specialization/
Generalization (cont.)
 A subclass with more than one superclass
is called a shared subclass, leading to
multiple inheritance
» e.g., ENGINEERING_MANAGER
 Any inherited attribute should be
included only once in a shared subclass
IM ISU
Database
21
IM ISU
Database
22
Utilizing Specialization/Generalization in
Conceptual Modeling
 Top-down conceptual refinement
 Start with an entity type and then define
subclasses by successive specialization
 e.g., A university database schema
PERSON => {EMPLOYEE, ALUMNUS, STUDENT}
{STAFF, FACULTY, STUDENT_ASSISTANT}
{GRADUATE_STUDENT,UNDERGRADUATE_STUDENT}
IM ISU
Database
23
Utilizing Specialization/Generalization
in Conceptual Modeling (cont.)
 Bottom-up conceptual synthesis
 Start with entity types and then define
superclasses by successive generalization
 e.g., {STAFF, FACULTY, ALUMNUS}
EMPLOYEE
 Hybrid conceptual process
 A combination of the two processes
IM ISU
Database
24
UNION Types Using Categories
 Union type (Category)
 Modeling a single superclass/subclass
relationship with more than one superclass
 Superclasses can represent different entity types
 e.g., In a vehicle registration database, the
vehicle owner can be a person, a bank, or a
company
=> OWNER: a subclass of the UNION of COMPANY,
BANK, and PERSON
 The union type (category) is specified via 
symbol, called set union operation
IM ISU
Database
25
IM ISU
Database
26
IM ISU
Database
27
UNION Types Using Categories
(cont.)
 A category can be total or partial
» e.g., Fig 4.9, where c1 and c2 are predicate
conditions that specify which COMPANY
and PERSON entities are members of
ACCOUNT_HOLDER
 The superclasses of a category may
» have different key attributes
– e.g., OWNER category of Fig 4.8
» have the same key attribute
– e.g., REGISTERED_VEHICLE category
IM ISU
Database
28
UNION Types Using Categories
(cont.)
IM ISU
Database
29
UNION Types Using Categories
(cont.)
 Category vs shared subclass
 A category is a subset of the union of its
superclasses
 An entity that is a member of a category
must exist in only one of the superclasses
» e.g., an OWNER may be
– a COMPANY
– a BANK, or
– a PERSON
IM ISU
Database
30
UNION Types Using
Categories(cont.)
 A shared subclass is a subset of the
intersection of the superclasses
 An entity that is a member of the shared
subclass must exist in all superclasses
» e.g., An engineering manager must be
– an ENGINEER,
– a MANAGER, and
– a SALARIED_EMPLOYEE
IM ISU
Database
31
UNION Types Using
Categories(cont.)
 Category vs generalized superclass
 A total category can be represented as a
specialization (or a generalization)
» e.g., Fig 4.9(b)
 A partial category cannot be represented
as a partial specialization (or a
generalization)
IM ISU
» e.g., category REGISTERED_VEHICLE
(Fig 4.8) and generalized superclass
VEHICLE (Fig
4.3(b))
Database
32
UNION Types Using Categories
(cont.)
IM ISU
Database
33
An Example EER Schema
 A UNIVERSITY Database Example
 Requirements: The database keeps track
» Students and their majors, transcripts, and
registration
» The university’s course offerings
» The sponsored research projects of faculty
and graduate students
» The advisor and thesis committee of each
graduate student
» Colleges and their related departments
IM ISU
Database
34
An Example EER Schema (cont.)
 Initial design
 Entity types
»STUDENT
»COURSE
»FACULTY
»GRADUATE_STUDENT
»GRANT (project)
IM ISU
Database
35
An Example EER Schema (cont.)
»DEPARTMENT (student’s majors)
»SECTION (course offerings, student’s
registration)
»COLLEGE
 Superclass/subclass
»STUDENT/GRAD_STUDENT
»SECTION/CURRENT_SECTION (for
quarter system universities)
IM ISU
Database
36
An Example EER Schema (cont.)
 Generalization
»PERSON: {FACULTY, STUDENT}
 Category (Union type)
»INSTRUCTOR_RESEARCHER:
{FACULTY, GRAD_STUDENT}
IM ISU
Database
37
An Example EER Schema (cont.)
 Iterative refinement
 Relationship types
» STUDENT <REGISTERED>
CURRENT_SECTION
» STUDENT <MAJOR> DEPARTMENT
» STUDENT <TRANSCRIPT> SECTION
» FACULTY <BELONGS> DEPARTMENT
» FACULTY <ADVISOR> GRAD_STUDENT
» FACULTY <COMMITTEE>
GRAD_STUDENT
» FACULTY <PI> GRANT
IM ISU
Database
38
An Example EER Schema (cont.)
» INSTRUCTOR_RESEARCHER <TEACH>
SECTION
» GRANT <SUPPORT>
INSTRUCTOR_RESEARCHER
» COURSE <CS> SECTION
» DEPARTMENT <DC> COURSE
» COLLEGE <CD> DEPARTMENT
 Structure constraints
IM ISU
Database
39
ER Conceptual Design (cont.)
IM ISU
Database
40
Alternative Notation
IM ISU
Database
41
Conceptual Object Modeling
 UML (Universal Modeling Language)
 Defines several diagrams that can be
used in the process of object oriented
software design
 Among the UML diagrams, class
diagrams are similar to EER diagrams
in many ways
IM ISU
Database
42
Conceptual Object Modeling
(cont.)
 UML class diagram vs EER
diagram
 Entity vs Object.
 Entity type vs Class
»Each class comprises three parts
– class name
– attributes
– operations
IM ISU
»A composite attribute is modeled as a
structured domain
»A multivalued attribute is modeled as a
Database
separate class
43
Conceptual Object Modeling
IM ISU
Database
44
Conceptual Object Modeling
(cont.)
IM ISU
Database
45
Conceptual Object Modeling (cont.)
 Relationship type vs Association
» UML has two types of relationships: association
and aggregation
» Aggregation is meant to represent a relationship
between a whole object and its component parts
» Relationship constraints, called multiplicities in
UML, are specified in the form min..max,
corresponding to (min, max) notation in ERD
» The placement of multiplicity is opposite when
compared to ERD
– e.g., cp. WORKS_FOR in Figs 3.15 and 4.11
IM ISU
Database
46
Conceptual Object Modeling (cont.)
»An association/aggregation may or may
not have a name
»It is not clear that when to choose
aggregation or association
»Weak entities can be modeled as a
qualified association (or qualified
aggregation)
– The partial key is placed in a box attach to
the owner class, e.g., Dependent Name
 Relationship vs link
IM ISU
»In UML, a relationship attribute is called a
link attribute Database
47
Conceptual Object Modeling
(cont.)
 UML notation for
generalization/specialization
» A blank triangle indicates a disjoint
specialization/generalization
» A filled triangle indicates overlapping
» cp., Figs 4.7 and 4.12
IM ISU
Database
48
Conceptual Object Modeling
(cont.)
IM ISU
Database
49
Conceptual Object Modeling
(cont.)
IM ISU
Database
50
Chapter Summary
 Introduced the EER model concepts
 Class/subclass relationships
 Specialization and generalization
 Inheritance
 These augment the basic ER model concepts
introduced in Chapter 3
 EER diagrams and alternative notations
were presented
IM ISU
Database
51
In-Class Exercise
Consider the BANK ER schema of Figure 3.21, and
suppose that is is necessary to keep track of
different types of ACCOUNTS
(SAVINGS_ACCTS, CHECKING_ACCTS,...) and
LOANS (CAR_LOANS, HOME_LOANS, ...).
Suppose that is is also desirable to keep track of each
account's TRANSACTIONs (deposits, withdrawals,
checks, ...) and each loan's PAYMENTs; both of
these include the amount, date, time, ... .
Modify the BANK schema, using ER and EER
concepts of specialization and generalization.
IM ISU
Database
52