-
BIS3635 - Database Systems
School of Management, Business Information Systems,Assumption
University
A.Thanop SomprasongChapter # 2Data Models
-
To learn about data modeling and why datamodels are importantTo
learn about the basic data-modeling building blocksTo learn what
business rules are and how they influence database designTo learn
how the major data models evolvedTo learn how data models can be
classified by level of abstractionObjectives
-
Designers, programmers, and end users see data in different
waysDifferent views of same data lead to designs that do not
reflect organizations operationData modeling reduces complexities
of database designVarious degrees of dataabstraction help reconcile
varying views of same dataIntroduction
-
Data models Relatively simple representations of complex
real-world data structuresOften graphicalModel: an abstraction of a
real-world object or eventUseful in understanding complexities of
the real-world environmentData modeling is iterative and
progressiveData Modeling and Data Models
-
Facilitate interaction among the designer, the applications
programmer, and the end userEnd users have different views and
needs for dataData model organizes data for various usersData model
is an abstractionCannot draw required data out of the data modelThe
Importance of Data Models
-
Entity: anything about which data are to becollected and
storedAttribute: a characteristic of an entityRelationship:
describes an association among entitiesOne-to-many (1:M)
relationship Many-to-many (M:N or M:M) relationshipOne-to-one (1:1)
relationshipConstraint: a restriction placed on the dataData Model
Basic Building Blocks
-
Entity can be classified into 3 major parts :GeneralsPersonals:
student, employee, instructor, doctor etc.Places: restaurant,
company, hospital, zoo, classroomObjects: machine, car,
bookConceptssubject, faculty, departmentEventsregistration,
enrolment, borrowing, returningEntity Types (+)
-
Descriptions of policies, procedures, orprinciples within a
specific organizationApply to any organization that stores and uses
data to generate informationDescription of operations to
create/enforce actions within an organizations environmentMust be
in writing and kept up to dateMust be easy to understand and widely
disseminatedDescribe characteristics of data as viewed by the
companyBusiness Rules
-
Sources of business rules:Company managersPolicy
makersDepartment managersWritten
documentationProceduresStandardsOperations manualsDirect interviews
with end usersDiscovering Business Rules
-
Standardize companys view of dataCommunications tool between
users and designersAllow designer to understand the nature, role,
andscope of dataAllow designer to understand business
processesAllow designer to develop appropriate relationship
participation rules and constraintsDiscovering Business Rules
(2)
-
Generally, nouns translate into entitiesVerbs translate into
relationships among entitiesRelationships are bidirectionalTwo
questions to identify the relationship type:How many instances of B
are related to oneinstance of A ?How many instances of A are
related to oneinstance of B ?Translating Business Rules intoData
Model Components
-
The Evolution of Data Models
-
Sometimes called top-down or parent-child structure
modelDeveloped in the 1960s to manage largeamounts of data for
manufacturing projectsBasic logical structure is represented by an
upside-down tree (Tree-like structure)Hierarchical structure
contains levels or segmentsSegment analogous to a record typeSet of
one-to-many (1:M) relationships between each particular segmentThe
Hierarchical Model
-
The Hierarchical Model (2)
-
Searching methodology will be initialized from top-down and
left-right format respectivelyFoundation for current data
modelsEasy to understand for this modelDisadvantages of the
hierarchical model:Complex to implementDifficult to manageLacks
structural independenceRelationships do not conform to M:N formNo
standards for how to implementThe Hierarchical Model (3)
-
Created to represent complex data relationshipsmore effectively
Improves database performanceImposes a database standardConference
on Data Systems Languages (CODASYL) created the DBTGDatabase Task
Group (DBTG): defined environment to facilitate database
creationThe Network Model
-
SchemaConceptual organization of entire databaseas viewed by the
database administratorSubschemaDatabase portion seen by the
application programsData management language (DML) Defines the
environment in which data can be managedThe Network Model (2)
-
Resembles hierarchical modelRecord may have more than one
parentCollection of records in 1:M relationshipsSet composed of two
record typesOwnerEquivalent to the hierarchical models
parentMemberEquivalent to the hierarchical models childThe Network
Model (3)
-
The Network Model (4)
-
Disadvantages of the network modelCumbersome (Too difficult)Lack
of ad hoc query capability placed burden on programmers to generate
code for reportsStructural change in the database could
producehavoc in all application programsThe Network Model (5)
-
Developed by E. F. Codd (IBM) in 1970Table (relations) Matrix
consisting of row/column intersectionsEach row in a relation called
a tupleRelational models considered impractical in 1970Model
conceptually simple at expense of computer overheadThe Relational
Model
-
Relational data management system (RDBMS)Performs same functions
provided by hierarchical modelHides complexity from the
userRelational diagramRepresentation of entities, attributes, and
relationshipsRelational table stores collection of related
entitiesThe Relational Model (2)
-
The Relational Model (3)
-
The Relational Model (4)
-
SQL-based relational database application involves three
parts:User interfaceAllows end user to interact with the dataSet of
tables stored in the databaseEach table is independent from
anotherRows in different tables related based on common values in
common attributesSQL engineExecutes all queriesThe Relational Model
(5)
-
Widely accepted standard for data modeling Introduced by Chen in
1976Graphical representation of entities and their relationships in
a database structureEntity relationship diagram (ERD)Uses graphic
representations to model database componentsEntity is mapped to a
relational tableThe Entity Relationship Model
-
Entity instance (or occurrence) is row in table Entity set is
collection of like entitiesConnectivity labels types of
relationshipsRelationships expressed using Chen
notationRelationships represented by a diamond Relationship name
written inside the diamondCrows Foot notation used as design
standard in this book and courseThe Entity Relationship Model
(2)
-
The Entity Relationship Model (3)
-
Data and relationships contained in singlestructure known as an
objectOODM (object-oriented data model) is the basis for
OODBMSSemantic data modelObjects contain operations Object is
self-contained: a basic building-block for autonomous
structuresObject is an abstraction of a real-world entityThe
Object-Oriented (OO) Model
-
Attributes describe the properties of an objectObjects that
share similar characteristics aregrouped in classesClasses are
organized in a class hierarchyInheritance: object inherits methods
and attributes of parent classUML based on OO concepts that
describe diagrams and symbolsUsed to graphically model a systemThe
Object-Oriented (OO) Model (2)
-
The Object-Oriented (OO) Model (3)
-
Extended relational data model (ERDM)Semantic data model
developed in response to increasing complexity of
applicationsIncludes many of OO models best featuresOften described
as an object/relational database management system
(O/RDBMS)Primarily geared to business applicationsThe Convergence
of Data Models
-
The Development of Data Models
-
Internet drastically changed role and scope of database
marketFocus on Internet makes underlying data model less
importantDominance of Web has resulted in growing need to manage
unstructured informationCurrent databases support XMLXML: the
standard protocol for data exchange among systems and Internet
servicesDatabase Models and the Internet
-
Common characteristics: Conceptual simplicity with semantic
completenessRepresent the real world as closely as
possibleReal-world transformations must comply with consistency and
integrity characteristicsEach new data model capitalized on the
shortcomings of previous modelsSome models better suited for some
tasksData Models: A Summary
-
Data Models: A Summary (2)
-
THE END