Semantic Data Modeling

Transcription

1 Semantic Data Modeling owhat is Conceptual Data Modeling oentity-relationship (E-R ) Modeling olimitations of E-R Modeling oobject-oriented Modeling: Another semantic model (discussed later in this course) Semantic Modeling 1

2 What Is Conceptual Data Modeling? A process that represents the entities, relationships, and activities of an enterprise in terms of a set of abstract concepts of a chosen data model for specific purposes. Enterprise Modeling, Business Modeling Conceptual Perception of of an an Enterprise Semantic Modeling 2

3 Bridge the Gap STUDENT( STUDENT( ID, ID, Name, Name, Age, Age, Address, Address, GPA GPA ) ) INSTRUCTOR INSTRUCTOR ( ( Emp#, Emp#, Name, Name, Rank, Rank, Dept Dept ) ) COURSE COURSE ( ( Course#, Course#, Credits, Credits, Title Title ) ) CLASS CLASS ( ( Emp#, Emp#, ID, ID, Course#, Course#, Time, Time, Room Room ) ) Semantic Modeling 3

4 Data Semantics ostatic Information u Data -- Entities u Associations -- Relationships among entities odynamic Information u Activities -- Operations/transactions u Integrity constraints -- Business rules/regulations and data meanings Semantic Modeling 4

5 Conceptual Data Model Revisited A conceptual data model consists of: o A collection of formal concepts o A set of usage rules Different model has different modeling capability Object-Oriented data data modeling Semantic data data modeling -- E-R -- EER -- etc. Conventional (Logical data data modeling) Semantic Modeling 5 -- Hierarchical -- Network -- Relational

6 E-R Modeling ointroduced by Peter Chen in 1976 obasic modeling concepts: u Entities, entity types, and attributes u Relationships Date Works_for 1 Department Language Emp# N Instructor Date 1 1 Assigned Office FName MInit Name LName N Teaches Time Location M Course Semantic Modeling 6

7 E-R Notation Entity Relationship Attribute Primary Key Semantic Modeling 7

8 Entities oan entity is a conceptual object ophysically exists 4Usually a noun in requirement specification Jose Alice Steve Class Student Acct Semantic Modeling 8 Department

9 Entity Types oa collection of similar entities oan abstraction of "physical" entities 4A noun in requirement specifications 4Having "independent" meaning Department Student Student Jose Steve Acct Alice Course Course Semantic Modeling 9

10 Weak Entity Types ocan't exist in DB independently omust be identified by its owner u Owner entity type u Identify relationship u Partial key u Total participation Parking Lot# Date Color PNum N 1 Student Registers Car Semantic Modeling 10

11 Attributes Properties or characteristics of entities and entity types u Attribute values -- Properties of entities u Value set - All acceptable attribute values u Attributes (definitions) -- Properties of entity types 4A noun or an adjective in requirement specifications 4No "independent" meaning Student Student ID Age Jose " " 25 Semantic Modeling 11

12 Key Attributes One or a group of attributes that can uniquely identify individual entities of an entity type u A key refers to one or a group of attributes as a whole u A key attribute is a component attribute of a key u Key changes with data semantics An entity type may have several qualified keys u Primary key -- One of the candidate keys u Alternate key - Candidate keys not used as the primary key u Secondary key -- An identifier of records with similar properties of interest 4The primary key attribute(s) is(are) underlined Semantic Modeling 12

13 More Attributes osimple attribute Contains atomic values only ocomposite attribute Student Student degrees id age Has component attributes name osingle-valued attribute Has exactly one value per entity Fname Mname Lname omulti-valued attribute Contains repeating values per entity oderived attribute CommRate CommEarned Salesrep EmpNo Attribute values computed by means of other attributes Semantic Modeling 13 Fname Mname Lname

14 Relationships Associations among entities u Relationships -- Associations among entities 4Usually a verb in requirement specification Occurrence Diagram or Semantic Net Course Student Joseph Alice Sue Tom Peter... Takes Student Takes Course Semantic Modeling 14

15 Relationship Degrees The number of entity types associated with that relationship. Each entity in a relationship is known as a participant. Unary Binary Employee Supervises Employee Work Department Ternary n-ary Project Project Supply Part Part Supplier Semantic Modeling 15

16 N-ary Relationships oa link must associate with all participants ocardinality is with respect to individual relationships 4A N-ary relationship is not equivalent to N binary relationships Customer N Salesperson M 1 ships Order Semantic Modeling 16

17 Relationship Attribute odescribes the association 4A adverb or noun in requirement specification sect# time Instructor teaches Course room# date Semantic Modeling 17

18 Relationship Cardinality How entities are connected through a relationship u One-to-One -- An entity of E1 is connected to at most one entity of E2 and vice versa. u One-to-Many -- An entity of E1 may be connected to one or more entities of E2, but an entity of E2 can only be mapped to at most one entity of E1. u Many-to-Many -- An entity of E1 may be linked to one or more entities of E2, and vice versa. E1 a b c... a b c... a b c... R 1 [ ] [ ] [ ] M [ ] [ ] [ ]... M... N E2 x y z... x y z... x y z... Semantic Modeling 18

19 Relationship Types Recursive Relationship Relationship type where same entity type participates more than once in different roles. Relationships may be given role names to indicate purpose that each participating entity type plays in a relationship. Semantic Modeling 19

20 Recursive Relationship called Supervises with Role Names Semantic Modeling 20

21 Entities associated through two distinct Relationships with Role Names Semantic Modeling 21

22 Participation Constraints Let R be a relationship type involves entity type E as a participant. If every instance of E participates in at least one instance of R, then the participation is said to be total; other wise it is said to be partial [Date, 2004] opartial (optional) participation An entity of E does not have to be mapped to another entity through the relationship. Writer Writer 1 Publishes ototal (mandatory) participation Every entity of E must be connected through the relationship to other entity ( or entities ). M Book Book Semantic Modeling 22

23 Structural Constraints Main type of constraint on relationships is called multiplicity. Multiplicity - number (or range) of possible occurrences of an entity type that may relate to a single occurrence of an associated entity type through a particular relationship. Represents policies (called business rules) established by user or company. Semantic Modeling 23

24 Multiplicity Faculty Teaches Course Takes Student (0,4) (1,1) (1,45) (?,?) Semantic Modeling 24

25 Other E-R Models Relationship Entity 1 Entity 2 attr. 1 attr. 2 Entity 1 Relationship Entity 2 ( 1:m ) ( 0:5 ) Entity 1 Relationship Entity 2 Semantic Modeling 25

26 Advantages Pros and Cons of E-R Emp#, Name, Address Salary, Skill u Simple and easy to understand. u Very popular. u Semantic richer than classical data models. Disadvantages: u Not a formally defined data model. u Deals with some integrity constraints. u Difficult to distinguish entities from relationships. u Has redundant modeling information. Project Member m-has Dependent has Project Manager ID, Name Address Birth-date Works-on Project Manages Emp#, Name, Address Salary, Skill Semantic Modeling 26

27 Is-A Relationship ogeneralization and specialization hierarchy u Supertypes -- Hide the differences of subtypes u Subtypes -- Reveal specific properties Generalization Person Is-A Student Is-A Staff Is-A Is-A Is-A Is-A UnderGrad Grad Faculty Secretary Specialization Semantic Modeling 27

28 Inheritance oa supertype contains the common properties of all its subtype entities. osubtype inherits properties of its supertype and may have its own properties. Person Person ID, Name, Address Is-A Is-A Salary, Skill Employee Dependent Birth-date Is-A Is-A Project Project Project Project Manager Member Ratio-of-success,... Languages,... Semantic Modeling 28

29 ER Example In-class exercise A database is being constructed to keep track of the teams and games of a baseball league. A team has a number of players, not all of whom participate in each game. It is desired to keep track of the players participating in each game for each team, the positions they played in that game, and the result of the game. Design an ER schema diagram for this application Assumptions: Each game in the schedule is identified by a unique Game#, and a game is also identified uniquely by the combination of Date, starting Time, and Field where it is played. A performance attribute is used to store information on the individual batting performance of each player in a game. Semantic Modeling 29

30 EER Modeling Example American Airlines Company The American Airlines Company publishes a monthly flight log report that tracks which type of aircraft and the number of hours that were flown by an individual pilot. A separate report is prepared for each pilot and is used to monitor pilot flight proficiency for the two types of aircraft (fixed-wing and rotorcraft) which a pilot may be qualified to fly. The following business rules apply to this report. Pilots may be assigned to fly different aircraft each day by the flight scheduling manager. Each aircraft has a single crew chief permanently assigned to perform maintenance on the aircraft, although a crew chief may crew more than one aircraft. Each aircraft is identified by an aircraft number. There are several aircraft types in the fleet. An aircraft number is unique within an aircraft type. Identify the entities in this situation and draw an E-R diagram of the entities to include their relationships and any attributes identified in this example. Give examples of additional attributes that might be associated with each entity. Semantic Modeling 30