Introduction to Information Technology - PowerPoint PPT Presentation

Loading...

PPT – Introduction to Information Technology PowerPoint presentation | free to download - id: 72f26b-NDBlZ



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

Introduction to Information Technology

Description:

Introduction to Information Technology Chapter 5: Managing Organizational Data and Information – PowerPoint PPT presentation

Number of Views:176
Avg rating:3.0/5.0
Slides: 98
Provided by: Rober1299
Learn more at: http://memberfiles.freewebs.com
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Introduction to Information Technology


1
Introduction to Information Technology
  • Chapter 5
  • Managing Organizational Data and Information

2
Chapter Preview
  • In this chapter, we will study
  • Basic data management terminology, updating,
    retrieval and access
  • Storing data in traditional files and problems
    with this approach/access
  • The data base approach to storing data
  • How data is organized/arranged to create a data
    base
  • Components and architecture of a DBMS
  • How companies utilize their stored data

3
Introduction
  • Back bone of IS is database and back bone of
    database is data management, efficient storage,
    easy access and all these are vital to business
    functions.
  • So in modern data management techniques H/W,S/W
    are integrated in an organization in order to
    survive in hypercompetitive global markets.

4
Basics of Data Arrangementand Access
  • The Data Hierarchy
  • Recall8 bits gt 1 byte gt 1 character
  • Word-A logical grouping of characters forms a
    word
  • Field - a logical grouping of words, a small
    group of words, or a complete number
  • Record - a logical grouping of related fields
  • File - a logical grouping of related records
  • Database - a logical grouping of related files
  • Storage Ways to store
  • Arrangement How they can be accessed either
    sequential or direct access. How the
    relationships between data entities are defined.
  • Sequential media (tape) stores records
    sequentially based on key values
  • Direct (or random) media (disks) use other
    techniques

5
Storing and Accessing Records
  • Indexed sequential access method (ISAM)-It uses
    an index of key to locate individual records ,and
    lists the key field of each record and where the
    record is physically located e.g. track and
    cylinder index.
  • Direct file access method-Uses the key field to
    locate the physical address of a record. It
    employs a mathematical formula called transform
    algo that performs mathematical calculations on
    record key and the result of this algo is the
    required particular records address

6
Traditional File Environment
  • The organization has multiple applications with
    related data files

Each application has a specific data file related
to it, containing all the data records needed by
the application
Each application comes with an associated
application-specific data file
7
Traditional File Environment (continued)
8
Disadvantages of file system
  • Slow information retrieval
  • Scattered information
  • Loose backup restore controls
  • Data integrity (Data values must be accurate and
    fit for their intended use)
  • Lack of security because of involvement of
    multiple applications and many people and
    scattered data.
  • Application and data dependence( Data files are
    dependent on their respective application in file
    environment)

9
Problems with traditional file system
  • Embedding same function for each application e.g.
    input, retrieval, querying, report generation etc
    ,As corporate applications typically share common
    core functions ,So great expenses for each
    application
  • Needs trained users
  • Data redundancy (Data duplication at several
    places)
  • Data inconsistency (Due to duplication there is
    no longer agreement of several copies) in-correct
    results
  • Data isolation (transparency) As data are
    accessed and stored in different formats in
    different applications

10
Definition
  • Meta Data
  • Data about data
  • Description of Files - types, storage, formats,
    locations, etc.
  • Description of Graphic Data - types, formats,
    layers, source, accuracy, dates,
  • Description of Attribute Data - fields, formats,
    types, widths, creators, dates, accuracy
  • Development Procedures - methods, outputs

11
  • Types of meta data
  • 1- Technical meta data
  • Includes where the data come from
  • How the data were changed, stored and organized
  • Who can access the data
  • Last date of data update
  • Who owns the data

12
  • 2- Business meta data
  • Includes info about what data are available
  • Where the data are
  • What the data mean (description/contents/units)
  • How to access/retrieve the data
  • What are characteristics of quality data and GIGO

13
Data Redundancy
  • A set of relations is said to be redundant, if
    it contains data in a relation that is derivable
    from other relations in any way i.e. by repeating
    whole row, by repeating some of the attributes or
    by combination of attributes from other
    relations. The redundant data may waste storage
    space and may effect access time.

14
Data Consistency
  • A database is said to be consistent, if it has
    no contradiction. If a database is fully
    normalized then in-consistency will not arise,
    because each fact is stated only once and no
    derivable facts are stored.

15
Anomalies
  • An anomaly is raised by inconsistent or
    contradictory state of the database.
  • If anomalies are present, we will be unable to
    represent some information, we might loose
    information when certain updates are performed
    and we will run the risk of having data become
    inconsistent over the time.

16
  • Suppose we have a record of the format.
  • 01 Order_record.
  • 02Ord_no
  • 02Ord_date
  • 02 Prod_record occurs 20 times.
  • 03prodid
  • 03prod_descr
  • 03 prod_price
  • 03prod_qty

17
  • It will be impossible to insert Prod_No.
    Prod_Descr and Prod_Price for a new product if no
    customer had yet purchased the product. It is
    known as insertion anomaly. However, if fulfilled
    order were deleted. Deletion of the last order
    containing information about a particular product
    would remove the information of that product form
    the database. this is called as deletion anomaly.

18
Cont
  • Suppose the price of a product is to be changed.
    Then all the occurrences of the product are to be
    updated. If any record is missed then data will
    become inconsistent it is called as update
    anomaly.

19
Database Management System (DBMS)
Database The Modern Approach
The database management system provides access to
the data
20
Data base
  • An organized collection of related information,
    Which is used by the (single) application system
    of an enterprise. It can also be considered as an
    electronic filing cabinet

21
Components of database environment
22
Types of database
  • Desktop databases
  • Multiuser databases
  • Categories of multiuser databases
  • Centralized Databases
  • Distributed Databases

23
Locating Data in Databases
  • Centralized and Distributed
  • Choices will affect user accessibility, query
    response time, data entry, security, and cost
  • Option 1 Centralized database
  • All the related files are in one physical
    location Provides database administrators with
    the ability to work on a database as a whole at
    one location
  • In a centralized database, All data is stored at
    a central location. Remote users access the
    database using communication facilities.
    Centralized databases provide better control over
    data

24
Cont
  • Advantages
  • Saves expenses
  • Centralized administrations
  • Consistency of files
  • Data security
  • Good recovery from disaster
  • Good control on data transactions
  • Disadvantages
  • When centralized sys fails all users suffer
  • Delayed/slow data access

25
Locating Data in Databases (continued)
26
Distributed Databases
  • Complete copies of a database, or portions of a
    database, are
  • in more than one location, close to the user
  • Distributed database approach is used for the
    organizations who are spread over different
    locations (these locations may also be different
    countries). In such cases centralized database
    approach may not be feasible.
  • A distributed database is single logical
    database spread physically across the computers
    situated at different locations.
  • Distributed Databases may be Homogeneous or
    Heterogeneous
  • Databases (Databases under one platform) or
    Heterogeneous databases (Databases under
    different platforms)
  • A collection of multiple ,logically interrelated
    databases distributed over a computer network

27
Types of distributed database
  • Replicated database
  • Has complete copy of entire database in many
    locations
  • Copies of database in many location
  • Reduced single-point-of-failure problems
  • Increased user access responsiveness
  • Portioned database
  • Has subdivided portions of data in many locations
  • Provides good access speed, but causes
    consistency , security and single point- of
    -failure problem
  • Each location responsible for its own data

28
Database Development
  • To develop a data base a designer must develop a
    conceptual (an abstract model of the data base
    from user or business perspective, How data
    elements are grouped/related to be easily
    accessed) and physical (How the data base is
    actually arranged on storage device) design of
    data base

29
Database Development
  • First, develop a Conceptual design - an abstract
    model of the database from the user or business
    perspective
  • Second, organize with Entity-Relationship (ER)
    modeling, A process of planning the database
    design
  • Third, define Entity classes ? Instance ?
    Identifiers ? Relationships

30
Database Development
  • Fourth, physically implement the data structure
    in the database management system software
  • Create tables
  • Define fields properties
  • Establish primary keys
  • Define table relationships
  • Add actual data (records) to tables

31
Database design methods
  • Commonly two methods are employed to produce
    optimal database designs
  • 1- Entity relationship modeling
  • 2- Normalization

32
Architecture Of Database Management System
  • A database can be viewed at three different
    levels of abstraction. The combination of these
    three levels is called as three level
    architecture of databases and these levels are
    experssed as models, views or schemas of
    database.
  • The purpose of three level architecture is to
    separate the way database is physically
    represented from the way users think about it.
    Most of databases follow three level architecture
    but some small database management systems
    deviate from this architecture.
  • Transparency (Network, replication,
    fragmentation)

33
Database Architecture
User A
User B
User D
User C
Conceptual View
Conceptual View
Internal View
Stored Database
34
Key terms
  • Database model
  • It refers to the storage mechanism of the data
    at the logical level.
  • Database schema
  • Each levels refer to the permanent structure of
    the database, This permanent structure is
    referred as database schema.
  • Database instance
  • Where as database instance is the information
    stored at any moment in the database.

35
Why Three Level Architecture?
  • Different users need different views of the same
    data.
  • View of a user may be changed over the time.
  • Users are not required to deal with the complex
    database storage/physical structures.
  • The DBA should be able to change logical or file
    structure of the database without effecting the
    users or physical structure.
  • Data structures should be unaffected, if changes
    are made to the physical aspects of storage such
    as change in storage device.
  • It should be independent of software.

36
External View
  • It is the individual users view of the database
    that means different users may have different
    views of the same data. The external view (user
    interface developed by the DBA) creates the
    working environment of the user, accepting and
    displaying the information in the format, that
    user expects. It also acts as a boundary below
    which user is not permitted to see. It hides
    hardware details from users. External view may
    consist of many occurrence of the same data.

37
Logical (Conceptual) View
  • This community view of data includes the
    description of all data available in the
    database. All the entities, their attributes and
    their relationships are represented in this
    level.
  • Conceptual view is representation of data in a
    form that is abstract in comparison with the way
    in which data is physically stored but in the
    user terms.

38
Internal View
  • It is the low level representation of entire
    database. the internal view is not basically the
    physical view but it is closest to the physical
    view because it does not deal in terms of
    physical records, blocks, pages, cylinders or
    tracks. The internal view uses addresses and
    these addresses are mapped to the physical
    storage.

39
Logical versus Physical View
  • Logical view - represents data in a format that
    is meaningful to a user (e.g., tables with fields
    and records)
  • Physical view - deals with the actual, physical
    arrangement and location of data in the direct
    access storage devices (DASD)
  • DBMS shields the user from having to know about
    the physical location of the data user only has
    to know the logical way its organized

40
Entity-Relationship Model
  • Used for conceptual data modeling
  • Conceptual data modeling is a representation of
    the structure of a database that is independent
    of the software that will be used to implement
    the database
  • Made up of entities, relationships, and
    attributes
  • Can be used to model any kinds of entities and
    relationships

41
E-R Model
  • Is a detailed, logical representation of the data
    of an organization (entities), relationships (or
    associations) among entities and their
    attributes.
  • Allows us to identify relationships between
    entities - at best it provides a broad overview
    of relationships
  • An entity relationship diagram is a graphical
    representation of an E-R model
  • Components of an E-R model
  • Entities
  • Attributes (and identifiers)
  • Relationships

42
Entity-relationship model
  • Types of relationships
  • One-to-oneA student has one ID an ID belongs to
    one student
  • One-to-many A course has one professor a
    professor has one or more courses
  • Many-to-many A student have one or more courses
    a course has one or more students

43
ENTITIES
  • Entity is any distinguishable object, person,
    place, Concept or event about which we collect
    data and store it in the database or in other
    words entity is any object that exists and is
    distinguishable from other objects.
  • Entities share common properties but also have
    one or more distinct properties

44
Cont..
  • Person Employee, Student, Patient
  • Place Region, Country, Province
  • Object Machine, Building, Automobile
  • Event Sale, Registration
  • Concept account, course

45
ENTITY TYPE
  • A collection of entities that share common
    properties or characteristics is referred as
    ENTITY TYPE.

Employee
Account
Course
46
ENTITY INSTANCE
A single occurrence of an entity type is Entity
Instance
ATRIBUTES
A property or characteristic of an entity that
is of interest to the organization
Student No.
Name
Address
Phone No
Student
47
Multivalued attributes
  • An attribute which can have more than one value
    for each entity instance (or repeating groups) is
    referred as Multivalued attribute. i.e.
  • EMPLOYEE (EmpNum, EmpName, EmpAddr, EmpSkill)
    ,EmpSkill is a multivalued attribute.
  • These are removed during normalization process
    (occurs during logical design)

48
RELATIONSHIPS

Relationships is an association between the
instance of one or more entity types.
complete
Student
Course
49
Relationships
  • Are an association between instances of one or
    more entity types of interest to the organization
  • The degree of the relationship is determined by
    the number of entity types that participate in a
    relationship
  • Relationships can be unary, binary, and ternary
  • Relationships can also be HAS-A or IS-A
    relationships

50
Unary Relationship
If the relationship is between entities in a
single entity type then it is called as unary
relationship for example Manager and Employee are
of the same entity type that is Employee,
therefore the relationship between manager and
employee is of the form Unary.
Manages
Employee
51
BINARY RELATIONSHIP If the relationship is
between entities in one entity type and entities
in an other entity type then it is called as
binary relationship. For example  
Works on
Employee
Project
Registered
Student
Course
52
TERNARY RELATIONSHIP A ternary relationship
involved more than two entity types i.e. a Vender
ships certain Parts from a Warehouse
Part
Ships
Vendor
Warehouse
53
Cardinalities
  • Cardinality of a relationship is the number of
    instances of entity B that can be (or must) be
    associated with each instance of entity A

54
Cardinalities continued
  • Cardinality can be maximum or minimum
  • Minimum cardinality - minimum number of instances
    of entity B that may be associated with instance
    of entity A
  • Maximum cardinality - maximum number of instances
    (can also be a fixed number) Minimum cardinality
    0 means optional participation
  • Minimum cardinality 1 means mandatory
    participation

55
Existence Dependency
If an instance of one entity can not exist
without the existence of an instance of some
other entity then we say that these entities have
Existence Dependency
Weak Entity A weak entity depends for its
existence on another entity A weak entity cannot
be created without its proper parent When the
parent entity is deleted, the weak entity should
also be deleted
56
HAS-A Relationships
  • One-to-one (11)
  • One-to-many (1M)
  • Many-to-many (MN)

57
Relationship
  • Described as 11
  • a single-entity instance of one entity type is
    related to a single entity instance of another
    entity type
  • Described as 1M
  • a single-entity instance of one entity type
    related to many instances of another entity type
  • parent-child relationship
  • parent is the one side (primary table)
  • child is the many side (related table)

58
One-to-Many relationship
  • Described as MN
  • many entity-instances of one entity type relate
    to many instances of another entity type

59
IS-A Relationships
  • IS- A relationship occurs among entities that are
    subtypes of a common logical type (or supertype)
  • example Employee can be hourly or salaried
  • Relationship between supertype and subtype is
    called IS-A
  • cardinality from subtype to supertype is 1 (i.e.
    mandatory)
  • cardinality from supertype to subtype is 0 or 1
  • Supertype - A generic entity type subdivided into
    subtypes
  • Subtype - a subset of a supertype that shares
    common attributes or relationships distinct from
    other subset

60
Supertype and Subtype
  • EMPLOYEE (EmpNum, EmpName, EmpAddr, DateHired)
  • HOURLY (EmpNum, HourlyRate)
  • SALARIED (EmpNum, AnnualSal)
  • Supertype Employee
  • Subtype Hourly, Salaried

61
Supertype and Subtype
  • Exclusive relationship - subtypes are mutually
    exclusive (such as a person is either hourly or
    salaried but not both)
  • Inheritance - property by which all attributes of
    a supertype become attributes of its subtypes

62
Normalization
  • It is a process of grouping and regrouping the
    attributes in such a way that duplication of data
    be avoided and data may be stored more
    efficiently. It is more detailed and non
    diagrammatical process, which database designer
    carries out while designing the database.
  • Helps achieve
  • minimum redundancy
  • Free from anomalies
  • maximum data integrity
  • best processing performance

63
Types of Relations
  • a) Base Relation

A table that exists in the system at logical
level is known as base table.
b) Derived Relation
A table that is derived from other table(s) by
means of some relational operations.
c) View and snap shots
These are named derived tables.
64
Database Management Systems
  • A set of software programs that provide access to
    a database
  • Data is stored in one location, from which it can
    be updated and retrieved
  • Application programs are given access to the
    stored data by various mechanisms
  • Maintaining the integrity of stored information
  • Managing security and user access
  • Recovering information when the system fails
  • Accessing various database functions from within
    an application

65
Advantages of DBMS
  • Sharing of data/improved access
  • Control of redundancy
  • Data consistency
  • Batter data security
  • Improved data integrity
  • Development of new applications
  • Economic
  • Batter backup recovery procedures
  • Providing transaction support
  • Reduced development, maintenance and
    costs/complexity
  • Application/data independence

66
DBMS Components
  • Data Model
  • Defines the way data are conceptually structured
  • Data Definition Language (DDL)
  • Used to define the content and structure of the
    data base
  • Users define their logical view (schema) of the
    database using the DDL
  • Physical characteristics of records and fields
    are defined
  • Relationships, primary keys, and security can be
    established

67
More DBMS Components
  • Data Manipulation Language (DML)
  • Used to query the contents of the database, store
    or update information in the database, and
    develop database applications
  • Structured query language (SQL) - most popular
    relational database language
  • Data Dictionary
  • Stores definitions of data elements and data
    characteristics

68
Logical Data Models
  • Database model (Conceptual model or enterprise
    model) is the main consideration in database
    design. It is the tool to represent the
    conceptual organization of the data. It supports
    all the external views supported by the internal
    view. Where as internal model is only the
    physical implementation of the conceptual model.
    Some of the popular database models are listed
    below
  • In a logically structuring database, consider the
    characteristics of the data and how the data will
    be accessed.
  • Using these models, database designer can build
    logical or conceptual view of data that can then
    be physically implemented.

69
Database (Logical) Models
  1. Entity-Relationship Model
  2. Hierarchical Model
  3. Network Model
  4. Relational Model
  5. Object Relational Model
  6. Object Oriented Model
  7. Flat file model

70
Emerging and Specialized Data Models
  • Multidimensional
  • Hypermedia
  • Geographic information system database
  • Knowledge database
  • Multimedia database
  • Small-footprint database

71
HIERARCHICAL DATA MODEL
  • Hierarchical model is the oldest model. The
    oldest hierarchical database management system
    was developed to organize information needed by
    the Apollo moon-landing project. First version of
    IMS (information management system) was developed
    by a joint venture of IBM and north American
    aviation in 1968, which used hierarchical data
    model. The latest version of IMS is IMS/VS, which
    runs under MVS operating system. Another
    important hierarchical database management system
    is system-2000 by Intel corporation.

72
HIERARCHICAL DATA MODEL
  • The hierarchical model uses entity tree as its
    basic structure, which is a data structure that
    consists of hierarchy of nodes, with a single
    node called the root is at the highest level. A
    root node may have any number of children but
    each child node will have only one parent node on
    which it is dependent, which means the parent to
    child relationship is one-to-many but child to
    parent is one-to-one. Drawing a line between
    parent and child node shows parent child
    relationships. A node that has no children is
    called as leaf. Leafs can occur at different
    levels within a tree. Nodes that are children of
    the same parents are called siblings.
    Hierarchical path of a node is single path from
    root node to that particular node.

73
HIERARCHICAL DATA MODEL
  • The nodes along hierarchical path are called
    that nodes ancestors. Any node along a path from
    some given node to a leaf is called its
    descendant.. The height of a tree is the number
    of nodes on longest hierarchical path from root
    to leaf. A tree is said to be balanced if every
    path from the root node to a leaf has same
    length. A binary tree is one, in which each node
    has no more than two children.
  • Twins are the multiple occurrence of same
    child node type within the same parent segment
    occurrence. In hierarchical model, each segment
    type is assigned a unique type code according to
    its position in the tree structure diagram i.e.
    According to the preorder traversal with type
    code 1 for the root node.
  • Hierarchical data model gives best processing
    speeds, but poor query flexibility.

74
Relational Model
  • The relational model is based on the Dr. E.F
    Codd theory, Which was presented in 1970 and
    consists of 12 different rules. Dr. Codd claimed
    that any Database management system, to be
    relational, should comply with all the 12 rules
    of his theory.
  • In the relational model, Entity types and
    relationships are represented as relations
    (tables), attributes of these entity types and
    relationships are represented as Columns of the
    relations. Each row of the entity type (Relation)
    is known as tuple

75
Cont..
  • Relational model is common in PC environment
    because it is simple to understand.
  • Relational model provides high flexibility and
    ease of use.
  • Relational model provides slower search and
    access times a problem in high-volume business
    settings.

76
RELATIONAL MODEL
  • 1- Consists of several tables or Entities
  • 2- Database defined both in terms of Entities
    and Relationships between Entities
  • 3- Entities and Relationships require unique
    identifiers, termed as primary keys
  • 4- Constraints specified on relational database
    can include Domain Constraints, Key Constraints
    and Integrity Constraints
  • 5- Easier to control Redundancy, increase
    Database Consistency

77
Basic operations on relational database
  • 1) Select operation Creates a subset consisting
    of all records in the file that meet the certain
    criteria
  • 2) Join operation Combines more then one
    relational tables for comfort.
  • 3) Project operation Create subset consisting of
    columns creating new table consisting only
    required information's.

78
Relation
  • A relation is a named two dimensional table of
    data. Each relation consists of a set of named
    columns and a number of unnamed rows. Each
    column in a relation corresponds to an attribute
    of that relation whereas each row corresponds to
    a record that contains data values for an entity.
  • The rules of Dr Codd are discussed in the next
    slides

79
Zeroth Rule
  • Any RDBMS that is claimed to be relational should
    be able to manage databases entirely through its
    relational capabilities.
  • That means it should not be the case that some
    DBMS functions can be performed using relational
    approach but some other must be performed by
    other means.
  • RULE-1 Information Rule
  •   All information in an RDBMS is represented at
    logical level in exactly one way i.e. by values
    in table, how so ever the information is stored
    at physical level.

80
  • RULE-2 Guaranteed Access
  • Rule
  • Every data item in an RDBMS is guaranteed to be
    accessable by a combination of table name,
    primary key value and the column name.
  • RULE-3 SYSTEMATIC TREATMENT OF NULL VALUES
  •   In a fully relational database management
    system, Representation of missing or inapplicable
    information should be systematic, independent of
    data types. 

81
  •  RULE4 DYNAMIC ON LINE CATALOG.
  • The database description is represented at the
    logical level in the same way as ordinary data so
    that authorized users can apply same relational
    language for its retrieval as they apply to the
    regular data.
  • RULE-5 DATA SUB LANGUAGE-
  • A relational Database management system may
    support several languages, However there must be
    at least one language whose statements are
    expressible in some well defined syntax and that
    is comprehensive in supporting following items
  • Data Definition
  • View Definition
  • Data manipulation
  • Integrity constraints
  • Authorization (user authorization)
  • Transaction boundaries

82
  • Rule-6 View Updation
  • All views that are theoretically Updateable are
    also updateable by the system.
  • Rule-7 HIGH LEVEL INSERT
  • UPDATE AND DELETE
  • The capacity of handling base relation or a
    derived relation as a single operand applies not
    only to the retrieval of data but also to the
    insertion updation and deletion of data. For
    example consider following statement.
  • UPDATA PRODUCT
  • SET RICE PRICE 0.9
  • Where QIS 2 RE_ORD_L

83
  • Rule-8 PHYSICAL DATA INDEPENDENCE
  • Application programs and terminal activities
    remain logically unchanged, whenever any type of
    changes are made in either storage representation
    or access method.
  • RULE-9 LOGICAL DATA INDEPENDENCE
  • Application programs and terminal activities
    remain logically unchanged, when information
    preserving changes of any kind, that
    theoretically permits, are made to the base
    table. For example break one table to more tables
    on the basis of rows then create a view.

84
  • RULE-10 INTEGRITY INDEPENDENCE
  • Integrity constraints specific to a particular
    RDBMS must be definable in the relational
    database sub-language and storable in the catalog
    not in the application programs.
  • RULE-11 DISTRIBUTION INDEPENDCE
  •   An RDBMS may be distributed at different
    places.
  • RULE-12 NON SUB LANGUAGE VERSION
  •   If an RDBMS has a low level (single record
    processing at a time) language. Then that low
    level language can not be used to bypass the
    integrity constraints expressed in the high level
    relational language.

85
RELATIONAL MODEL-ADVANTAGES
  • Sharing of Data
  • Multi-user Transaction Processing
  • Multiple Views of Database, distinct GUIs
  • Flexibility, Scale Economies
  • Authorized Access Control
  • Ease in Constraint Enforcement
  • Redundancy Control
  • Avoidance of Cascading Errors
  • Fewer Data Entry Errors, higher Data Consistency
  • Ease in Backup and Recovery

86
NETWORK DATA MODEL
  • The network data model was first time presented
    in a report published in 1971 and COBOL language
    was used for its implementation. Subsequently
    this report was revised in 1978 and 1981 to
    include new concepts. There are two basic data
    structures in the network model namely records
    and sets.


87
NETWORK DATA MODEL
  • A network data model creates relationships among
    data through a linkedlist structure in which
    subordinate records are called members ,not
    children.
  • Like hierarchical data model it uses links
    called pointers that have storage address of
    related record. linked with owner record (as root
    in hierarchal model) and member.
  • It is every complex and difficult to maintain
  • For every data element a pair of pointers is
    to be maintained

88
RECORDS
  • In the network model, data is stored in
    records. Whereas each record consists of a group
    of related data items. Records are classified
    into record types. Each record type describes the
    structure of a group of records that stores the
    same type of information. Each record type is
    given a name and a name along with data type is
    also given to each data item with in the record
    type.
  • Typical database application may have more than
    one record types. To represent relationships
    between the records, the network model provides
    the constructs called as set type.

89
SYSTEM OWNED SET OR SINGULAR SET
  • In the network model, a system-owned set is a
    set with no owner record. Actually database
    management system is the owner of such set.
    System owned set provides the entry point in the
    database. Processing can start by accessing the
    members of the system owned set. A system owned
    set is also called as singular set because there
    is only one set occurrence of it. There are also
    some other special set types, these

90
Cont
  • Each set type definition consists of
  • A name for the set type
  • An owner record type
  • A member record type
  • Recursive set type
  • A set type having the same record serving as an
    owner as well as member record is called as
    recursive set type. Recursive set types are
    dis-allowed.
  • Multimember set type
  • A set type, which contains multiple record types
    as members in the same set types, is called as
    multiple set types.

91
Storage representation
  • A set instance is commonly represented as a
    ring linking an owner record and all member
    records of the set. This is also called as
    circular chain. A special field called the type
    field is included in the storage to distinguish
    the owner and member record sets. The type field
    has a unique value for each occurrence of a
    record type. Along with the type field, DBMS
    also stores a next pointer with the set. Apart
    from ring representation, there are also some
    other representations such as
  • 1) Doubly linked list along with the next
    pointer in a member record type, a previous
    pointer points back to the previous record.
  • 2) Owner pointer representation in this
    representation, for each set type, an additional
    owner pointer is included in the member record
    type, which points directly to the owner record.

92
  • 3)Contiguous member records In this
    representation, member records are placed in
    contagious physical locations following the owner
    record.
  • 4) Pointer array In this representation, an array
    of pointers is stored with the owner record, nth
    element in the array points to the nth member
    record of the set instance.
  • 5) Indexed representation In this representation,
    a small index is maintained with the owner
    record. Index entry contains the value of the
    index field and a pointer to the actual member
    record that has its field values.
  • Network data model gives pretty good processing
    speeds and pretty good query flexibility, but is
    very complex.

93
Data warehouse
  • Data warehouse
  • To have easy approach to modify ,access and
    retrieve huge data in global market place when
    data is integrated with the internet, they can be
    accessed from any location at any time
  • Then to make use if these data for decision
    making and analysis.
  • A data warehouse is a relational and
    multidimensional database management system
    designed to support management decision making.
  • Data warehouses are oriented around the major
    business subjects of the enterprise such as
    customer, vendor, product or activity.
  • Emphasis is on organizing/grouping data in
    convenient, meaningful ways so that users can get
    their queries answered more usefully
  • Current and historical, detail and summarized
    data are included.
  • Metadata (data about data) is included to help
    keep track of the data warehouse content.

94
Cont.
  • Gigantic computer (storage) of large amounts of
    data.
  • Stored at a single place and are with agreed
    format even operational applications store the
    data differently.
  • Data mart Small scale, simpler data warehouse.
    Easier to implement. Targets smaller business
    segments. Where as large scale data warehouse in
    called enterprise data warehouse.

95
Data mining
  • A sophisticated analysis technique that
    automatically discovers previously undetected
    relationships and dynamics among the data
    elements.
  • E.g. how the sale of one product might drive the
    sale of another.
  • Identifies facts or suggest conclusions based on
    intelligence
  • Extracting new insights from data warehouse
  • Sophisticated tools employ algorithms to discover
    hidden patterns, correlations, statistical
    measures and relationships.
  • It is integration of AI and databases.

96
Functions of data mining
  • Classifying (Certain groups of data having
    certain characteristics)
  • Clustering (Makes clusters of data groups which
    share some common characteristics)
  • Associating (Identifies relationships among these
    groups)
  • Sequencing (Makes sequencing of groups having
    similar associations)
  • Forecasting (Estimates future data values based
    on detected relations)
  • What can we learn (examples)?
  • Market segments and customer characteristics
  • Customer buying patterns
  • Fraudulent behavior

97
Thank You
About PowerShow.com