Statistical Database System - National Open University of Nigeria

Statistical Database System - National Open University of Nigeria
NATIONAL OPEN UNIVERSITY OF NIGERIA
SCHOOL OF SCIENCE AND TECHNOLOGY
COURSE CODE: DAM 461
COURSE TITLE: STATISTICAL DATABASE SYSTEM.
COURSE CODE: DAM 461
COURSE TITLE: STATISTICAL DATABASE SYSTEM.
COURSE
GUIDE
DAM 461
.
Statistical Database System
Course Developer/Writer
Dr. ADEKUNLE, Yinka A.
Computer Science & Mathematics
Babcock University, Ilishan-Remo,
Ogun State, Nigeria.
Programme Leader
Course Coordinator
ONWODI, Greg. O
The study units in this course are as follow:
Module 1 Fundamentals of Database Systems
Unit 1
Databases and Database Users
Unit 2
Database System Concepts and Architecture
Unit 3
Data Modelling Using the Entity-Relationship Model
Module 2 The Statistical database system
Unit 1
Statistical Database Concepts
Unit 2
Statistical Data Analysis, Mining and Decision Tree
Unit 3
Computer Security and Statistical Databases
Module 3 Application of Statistical Database System
Unit 1
SPEA SMART Airport Statistical Data Management
System (SMART STAT)
Module 1 Fundamentals of Database Systems
Unit 1 Databases and Database Users
1.0
Introduction
2.0
Objective
3.0
Definitions and functions
3.1
Characteristics of the Database Approach
3.2
Actors on the Scene
3.3
Advantages of Using a Database Management System
3.4
Implications of the Database Approach
3.5
When Not to Use a Database Management System
4.0
Conclusion
5.0
Summary
6.0
Tutor Marked Assignment
7.0
Further Reading and Other Resources
1.0
Introduction
In our life we have to remember so much of data. And it is easier for us to
remember all information for a few individuals but it is too difficult for us
to memorise all these information for a large number of individuals. There
is simply too much data to be managed in the minds and in order to store all
the new information, humanity invented the technology of writing.
Then, in a Competitive and global economy - data resources are essential
for survival, Information required for competitive initiatives. Accuracy &
timeliness very important, Managers must understand the competitive
advantages available through innovative use data. To this end, Databases
should be understood within the larger context of Information Resources
Management. The concept that information is a major corporate resource
and must be managed using the same basic principles used to manage other
assets (e.g. employees, materials, equipment, financial resources).
2.0
Objectives
At the end of this unit, you should be able to:
explain the concept of database
appreciate the features of database management system (DBMS)
State the major advantages of database approach
Draw the major components of database environment
3.0
Definitions and Functions
Data are known facts that can be recorded, and have an implicit meaning.
(e.g., names, telephone numbers, addresses, etc). Data are raw facts
concerning things such as people, objects, or events.
Information on the other hand is data that have been processed and
presented in a form suitable for human interpretation, often with the
purpose of revealing trends or patterns.
What is a Database?
A database is a collection of related information (data) in a structured way.
It is simply a bunch of information (data) stored on a computer. This could
be a list of all your clients, a list of the products you sell, the results of a
chess tournament or everyone in your family tree. The most common type
of database is a relational database. Relational databases consist of tables of
data with clearly defined columns. Database could also be a piece of
software used to store data. This is because the word “database” can refer to
both the software and the actual data.
What is a DBMS and what are its functions?
A DBMS (Database Management System) is best described as a collection
of programs that manage the database structure and that control shared
access to the data in the database. It is a software package to facilitate the
creation and maintenance of a computerized database.Current DBMSs also
store the relationships between the database components; they also take care
of defining the required access paths to those components.
The functions of a current generation DBMS may be summarized as follow:
1. Stores the definitions of data and their relationships (metadata) in a data
dictionary; any changes made are automatically recorded in the data
dictionary.
2. Creates the complex structures required for data storage.
3. Transforms entered data to conform to the data structures in item 2.
4. Creates a security system and enforces security within that system.
5. Creates complex structures that allow multiple user access to the data.
6. Performs backup and data recovery procedures to ensure data safety.
7. Promotes and enforces integrity rules to eliminate data integrity
problems.
8. Provides access to the data via utility programs and from programming
languages interfaces.
9. Provides end-user access to data within a computer network environment.
3.1
Characteristics of the Database Approach
There are two approaches to database system, which include:
The Traditional file Processing (File based) approach: Under this
approach, each user defines and implements the files needed for a
specific application. The approach is characterized by redundancy
(duplication) in defining and storing data, inconsistencies or various
versions of data, and wastage of storage space.
The database approach: This approach emphasizes the integration and
sharing of data across the organization. It is characterized by a single
repository of data (i.e. data located at a single site).
Characteristics of database approach:
Self-Describing nature of a database system: complete definition or
description of the database structure (structure of each file, type &
storage format of each data item), and constraints on data are stored in
the system catalogue and the information stored in the catalogue is
called meta-data.
Insulation between Programs and Data (structure): This is called
program-data independence or program-operation independence; the
characteristics that allow changing data storage structures without
having to change the DBMS access programs.
Data Abstraction: A situation in which data model is used to hide
storage details and present the users with a conceptual view of the
database.
Multiple views of the data: Each user may see a different view of the
database, which describes only the data of interest to that user.
Sharing of data and multiuser transaction processing: concurrency
control (on-line transaction processing (OLTP)) application.
3.2
Actors on the Scene
These include Persons whose job involves daily use of a large database,
such persons as:
Database Administrator (DBA): This includes person’s administering database, DBMS and related software. That is, persons responsible
for managing the database system, authorizing access, coordinating
and monitoring uses, acquiring resources (database and DBMs).
Database designers: These are persons responsible for designing the
database, identifying the data to be stored, choosing the appropriate
structures to represent and store this data, and interacting with users.
End Users: The persons that use the database for querying, updating,
generating reports, etc.
Casual end users: Occasional users.(middle- or high-level
managers)
Parametric (or naive) end users: They use pre-programmed
canned transactions to interact continuously with the database. For
example, bank tellers or reservation clerk.
Sophisticated end users: Use full DBMS capabilities for
implementing complex applications.
Stand-alone users (personal databases).
System Analysts and Application Programmers (Software
engineers): These are people that determine requirements of end users
(specification), implement, test, debug and document.
Workers behind the scene
DBMS system designers and implementers: These are people who
design and implement the DBMS modules and interfaces as a software
package.
Tool developers: Persons who develops software packages and tools
that facilitate database system design and use, and help improve
performance. Tools include design tools, performance tools, special
interfaces, etc.
Operators and maintenance personnel: Work on running and
maintaining the hardware and software environment for the database
system.
3.3
Advantages of Using a DBMS
Controlling redundancy in data storage and in development and
maintenance efforts, thereby eliminating duplication of efforts, space
wastage and inconsistence.
Restricting unauthorized access (security and authorization) .
Providing persistent storage for program objects and data
structures: Object-oriented database. OODB are compatible with
C++, Java.
Permitting inference and actions using rules.
Providing multiple user interfaces, backup and recovery.
Representing complex relationships among data.
Enforcing integrity constraints.
Saving time and aiding communication: The database is a more
efficient solution than paper files held in a file folder. Then, larger
companies can benefit from databases when information must be
spread to various users.
Databases Are Inexpensive Managers: Smaller businesses are always
looking for ways to cut costs without cutting quality. A database can
be a hefty investment initially, but, over the long term, it will save
money by improving the efficiency of all employees, impressing
customers who will not need to repeat their information and saving on
paper costs.
3.4
Implications of the Database Approach
The followings are the implications of using database approach in an
organisation:
Potential for Enforcing Standards.
Reduced Application Development Time.
Flexibility.
Availability of Up-to-date Information.
Economies of Scale.
3.5
When Not to Use DBMS Main Costs of
Using a DBMS:
When there is high initial investment in hardware, software, training
and possible need for additional hardware.
When there is high overhead for providing generality, security,
recovery, integrity, and concurrency control.
Generality that a DBMS provides for defining and processing data.
When a DBMS may be unnecessary:
If the database and applications are simple, well defined, and not
expected to change. If there are stringent real-time requirements that
may not be met because of DBMS overhead.
If access to data by multiple users is not required.
When no DBMS may suffice:
If the database system is not able to handle the complexity of data
because of modeling limitations.
If the database users need special operations not supported by the
DBMS.
4.0
Conclusion
A database is a collection of related information (data) in a structured way.
While database management is a collection of programs that manage the
database structure and that control shared access to the data in the database,
with different characteristics and users, as well as implications.
5.0
Summary
In this unit we have learnt that:
A database is a collection of related information (data) in a structured
way.
Database management is a collection of programs that manage the
database structure and that control shared access to the data in the
database.
There are two approaches to data management with different
characteristics and users.
The advantages include controlling redundancy, restricting
unauthorized access, saving time etc, with flexibility, economies of
scale, and potential for enforcing standards, are some of the
implications of database approach.
6.0
1.
Tutor Marked Assignment
(a) Differentiate between the following:
(i) Data and information (ii) Database and database management.
(b) What are the functions of database management?
2 . (a) What are the characteristics of database approach?
(b) Mention the advantages of using database management.
7.0
Further Reading and Other Resources
Relational Database Data www.OLDI.com Automatic RDB Data
Logging Enterprise Transaction Modules
ACL Data AnalysisData analysis made easier with ACL's world leading
Auto Database RecoveryReal-time database recovery enables 24-7
availability to essential data.
Databases in the cloud: a work in progress; October 2009; ISBN 978-160558-765-3.
Advantages of Relational Databases
Database Administrator Job Description
Introduction To Relational Data Model
Database Design As little or as many leads as you need. Check this
out! ChainStoreGuide.com/Data
Database Management WinSQL - A Homogeneous Solution for
Heterogeneous Environment. www.synametrics.com
The Advantages of Using a Database by: Jacqueline Thomas
Data Masking Softwarewww.orpheus-it.com Anonymize your sensitive
data Protect personal and company data
Database/SQL Toolwww.dbvis.com For DB2, SQL Server, Derby, Mimer
Informix, Oraclé and more
Sample Database?www.ioglobal.net Hosted essay data management: Safe,
Efficient QA, Global 24/7 access.
Database Mgmt & Design Hansen & Hansen
MS-Access, MS FoxPro etc. manuals, books
MIS Demo Databases (Access, Lotus Notes, Paradox)
A Database Wizard for accessing ODBC-compliant databases from Web
applications (database-enabled Web applications).
Module 1 Fundamentals of Database Systems
Unit 2 Database System Concepts and Architecture
1.0
Introduction
2.0
Objective
3.0
Data Models and Their Categories
3.1
History of Data Models
3.2
Schemas, Instances, and States
3.3
Data and the Three-Schema Architecture
3.4
Data Independence
3.5
DBMS Languages and Interfaces
3.6
The Database System Environment
3.7
Database System Utilities and Tools
3.8
Centralized and Client-Server Architectures
3.9
Classification of DBMSs
4.0
Conclusion
5.0
Summary
6.0
Tutor Marked Assignment
7.0
Further Reading and Other Resources
1.0
Introduction
The purpose of this Unit is to introduce students to the database approach
to information systems development, and to the important concepts and
principles of this approach. This is highly imperative because, it should
convey a sense of the central importance of databases in today’s
information systems environment. The idea of an organizational database
is intuitively appealing to most students.
However, many students have little or no background or experience of
databases. Others have had some experiences with a PC database (such as
Microsoft Access), and consequently have a limited perspective concerning
an organizational approach to databases. Therefore in this Unit, the basic
concepts and definitions of databases are introduced.
2.0
Objective
At the end of this unit, you should be able to:
Explain what data and database model are all about
Mention different types of data model
Vividly explain data and schema architecture
Describe the data independence
Explain database environment and its components etc.
3.0
Data Models and Their Categories
Data Model
This is a set of concepts used to describe the structures of a database, and
the operations for manipulating these structures, as well as certain
constraints that the database should obey. It is simply a diagram that
describes the most important “things” in business environment from a datacentric point of view. For example, an Entity Relationship Diagram
(ERD) below describes the relationship between the data stored about
products, and the data stored about the organizations that supply the
products.
Product ID
Unit Price
Qty on hand
Product
Entity
Cardinal
Attributes
Supplied
by
Name
Relationship
Supplier
Address
FIGURE 1: An ERD showing a relationship between products and suppliers.
Data Model Structure and Constraints:
Constructs are used to define the database structure. Constructs typically
include elements (and their data types) as well as groups of elements (e.g.
entity, record, table), and relationships among such groups. Constraints
specify some restrictions on valid data; these constraints must be enforced
at all times.
Data Model Operations:
These operations are used for specifying database retrievals and updates by
referring to the constructs of the data model. Operations on the data model
may include basic model operations (e.g. generic insert, delete, and update)
and user-defined operations (e.g. compute_student_gpa, update_ inventory).
Database Model
This is a theory or specification describing how a database is structured and
used. A database model is a collection of logical constructs used to
represent the database's data structure as well as the data relationship(s)
found within that structure. Several such models have been suggested which
include:
•
Flat model: This may not strictly qualify as a data model. The flat (or
table) model consists of a single, two-dimensional array of data
elements, where all members of a given column are assumed to be
similar values, and all members of a row are assumed to be related to
one another.
Figure 2
•
Flat model
Hierarchical model: In this model data is organized into a tree-like
structure, implying a single upward link in each record to describe the
nesting, and a sort field to keep the records in a particular order in
each same-level list.
Figure 3
•
Hierarchical model
Network model: This model organizes data using two fundamental
constructs, called records and sets. Records contain fields, and sets
define one-to-many relationships between records: one owner, many
members.
Figure 4
•
Network model
Relational model: This is a database model based on first-order
predicate logic. Its core idea is to describe a database as a collection of
predicates over a finite set of predicate variables, describing
constraints on the possible values and combinations of values.
Figure 5
•
Relational model
Object-relational model: This is similar to a relational database
model, but objects, classes and inheritance are directly supported in
database schemas and in the query language.
Figure 6
•
Concept-oriented model
Star schema: This is the simplest style of data warehouse schema.
The star schema consists of a few "fact tables" (possibly only one,
justifying the name) referencing any number of "dimension tables".
The star schema is considered an important special case of the
snowflake schema.
Figure 7
Star schema
Categories of Data Models:
Conceptual (high-level, semantic) data models: Provide concepts that are
close to the way many users perceive data (also called entity-based or
object-based data models).
Physical (low-level, internal) data models: Provide concepts that describe
details of how data is stored in the computer. These are usually specified in
an ad-hoc manner through DBMS design and administration manuals.
Implementation (representational) data models: Provide concepts that
fall between the conceptual and physical models, used by many commercial
DBMS implementations (e.g. relational data models used in many
commercial systems).
3.1
History of Data Models
One of the earliest pioneering works in modelling information systems was
done by Young and Kent (1958), who argued for "a precise and abstract
way of specifying the informational and time characteristics of a data
processing problem". They wanted to create "a notation that should enable
the analyst to organize the problem around any piece of hardware". Their
work was a first effort to create an abstract specification and invariant basis
for designing different alternative implementations using different hardware
components.
The next step in IS modelling was taken by CODASYL, an IT industry
consortium formed in 1959, who essentially aimed at the same thing as
Young and Kent: the development of "a proper structure for machine
independent problem definition language, at the system level of data
processing". This led to the development of a specific IS information
algebra.
In the 1960s data modelling gained more significance with the initiation of
the management information system (MIS) concept. According to Leondes
(2002), "during that time, the information system provided the data and
information for management purposes. The first generation database
system, called Integrated Data Store (IDS), was designed by Charles
Bachman at General Electric. Two famous database models, the network
data model and the hierarchical data model, were proposed during this
period of time". Towards the end of the 1960s Edgar F. Codd worked out
his theories of data arrangement, and proposed the relational model for
database management based on first-order predicate logic.
In the 1970s entity relationship modelling emerged as a new type of
conceptual data modelling, originally proposed in 1976 by Peter Chen.
Entity relationship models were being used in the first stage of information
system design during the requirements analysis to describe information
needs or the type of information that is to be stored in a database. This
technique can describe any ontology, i.e., an overview and classification of
concepts and their relationships, for a certain area of interest.
Also in the 1970s G.M. Nijssen developed "Natural Language Information
Analysis Method" (NIAM) method, and developed this in the 1980s in
cooperation with Terry Halpin into Object-Role Modelling (ORM). Further
in the 1980s according to Jan L. Harrington (2000) "the development of the
object-oriented paradigm brought about a fundamental change in the way
we look at data and the procedures that operate on data. Traditionally, data
and procedures have been stored separately: the data and their relationship
in a database, the procedures in an application program. Object orientation,
however, combined an entity's procedure with its data’’.
3.2
Schemas, Instances, and States
Schemas versus Instances:
Database Schema represents the description of a database, which includes
descriptions of the database structure, data types, and the constraints on the
database. Schema Diagram on the other hand, is an illustrative display of
(most aspects of) a database schema, while Schema Construct is a
component of the schema or an object within the schema, e.g., STUDENT,
COURSE.
Database State is the actual data stored in a database at a particular
moment in time. This includes the collection of all the data in the database.
It is also called database instance (or occurrence or snapshot). The term
instance is also applied to individual database components, e.g. record
instance, table instance, entity instance.
Database Schema versus Database State
Database State refers to the content of a database at a moment in time,
which includes:
Initial Database State: This refers to the database state when it is
initially loaded into the system.
Valid State: A state that satisfies the structure and constraints of the
database.
The distinction to be made here is that, database schema changes very
infrequently, whereas database state changes every time the database is
updated. Then the schema is also called intension, as state is called
extension.
Example of a Database Schema:
Figure 8
Database Schema
Example of a database state:
Table 1:
3.3
Database State
Data and The Three-Schema Architecture
Data architecture is the design of data for use in defining the target state
and the subsequent planning needed to hit the target state. It is usually one
of several architecture domains that form the pillars of an enterprise
architecture. It describes the data structures used by a business and its
applications. These are descriptions of data in storage and data in motion;
descriptions of data stores, data groups and data items; and mappings of
those data artefacts to data qualities, applications, locations etc.
And essential to realizing the target state, Data architecture describes how
data is processed, stored, and utilized in a given system. It provides criteria
for data processing operations that make it possible to design data flows and
also control the flow of data in the system.
Three-Schema Architecture defines DBMS schemas at three levels
which include:
Internal schema: This describes the physical storage structures and
access paths (e.g indexes) of database, at the internal level of DBMS
design. And it typically uses a physical data model.
Conceptual schema: This describes the structure and constraints for
the whole database, for a community of users, at the conceptual level.
This uses a Conceptual or an Implementation data model.
External schemas: This at the external level describes the various
user views. And it usually uses the same data model as the conceptual
schema.
Three-Schema Architecture is proposed to support DBMS characteristics
of Program-data independence, and multiple views of the data. Although it
not explicitly used in commercial DBMS products, it has been useful in
explaining database system organization.
Figure 9:
The Three-schema Architecture
Mappings among schema levels are also needed to transform requests and
data. Programs refer to an external schema, and are mapped by the DBMS
to the internal schema for execution. And data extracted from the internal
DBMS level is reformatted to match the user’s external view (e.g. formating the results of an SQL query for display in a Web page).
3.4
Data Independence
Data independence is the type of data transparency that matters for a
centralized DBMS. It refers to the immunity of user applications to make
changes in the definition and organization of data.
The physical structure of the data is referred to as "physical data description". Physical data independence deals with hiding the details of the storage
structure from user applications. The application should not be involved
with these issues since, conceptually; there is no difference in the operations
carried out against the data. And that, the data independence and operation
independence together gives the features of data abstraction.
There are two levels of data independence:
Logical Data Independence: This is the capacity to change the conceptual
schema (logical) without having to change the external schemas (user
views) and their associated application programs. For example, the addition
or removal of new entities, attributes, or relationships to the conceptual
schema should be possible without having to change existing external
schemas or having to rewrite existing application programs.
Physical Data Independence: This is the capacity to change the internal
schema (physical) without having to change the conceptual schema
(logical) For example, the internal schema may be changed when certain
file structures are reorganized or new indexes are created to improve
database performance, without having to change the conceptual or external
schemas.
And with knowledge about the three-scheme architecture, the term data
independence can be explained as follows: Each higher level of the data
architecture is immune to changes of the next lower level of the architecture. When a schema at a lower level is changed, only the mappings between
this schema and higher level schemas need to be changed in a DBMS that
fully supports data independence. The higher-level schemas themselves are
unchanged. Hence, the application programs need not be changed since
they refer to the external schemas.
3.5
DBMS Languages and Interfaces
DBMS Languages are of two categories which include:
Data Definition Language (DDL)
Data Manipulation Language (DML): This can either be
High-Level or Non-procedural Languages: These include the relational
language SQL, which may be used in a standalone way or may be
embedded in a programming language.
Low Level or Procedural Languages: These must be embedded in a
programming language.
Data Definition Language (DDL): This is the language used by the
database and database designers to specify the conceptual schema of a
database. In many DBMSs, the DDL is also used to define internal and
external schemas (views). In some DBMSs, separate storage definition
language (SDL) and view definition language (VDL) are used to define
internal and external schemas. SDL is typically realized via DBMS
commands provided to the DBA and database designers.
Data Manipulation Language (DML): This is the language used
to specify database retrievals and updates. DML commands (data
sublanguage) can be embedded in a general-purpose programming
language (host language), such as COBOL, C, C++, or Java. A library of
functions can also be provided to access the DBMS from a programming
language. Alternatively, stand-alone DML commands can be applied
directly (called a query language).
Types of DML
For example, the SQL relational languages are “set”-oriented and specify
what data to retrieve rather than how to retrieve it. It is also called
declarative languages.
Low Level or Procedural Language: This is used to retrieve data, one
record-at-a-time; constructs such as looping are needed to retrieve
multiple records, along with positioning pointers.
DBMS Interfaces
Stand-alone query language interfaces
Example: Entering SQL queries at the DBMS
Interactive SQL interface (e.g. SQL*Plus in ORACLE).Programmer
interfaces for embedding DML in programming languages
User-friendly interfaces
Menu-based, forms-based, graphics-based, etc.
DBMS Programming Language Interfaces
Programmer interfaces for embedding DML in a programming languages:
Embedded Approach e.g. embedded SQL (for C, C++, etc.), SQLJ(for Java)
Procedure Call Approach e.g. JDBC for Java,
ODBC for other programming languages
Database Programming Language Approach:
e.g. ORACLE has PL/SQL, a programming language based on SQL;
language incorporates SQL and its data types as integral components.
User-Friendly DBMS Interfaces
Menu-based, popular for browsing on the web.
Forms-based, designed for naïve users.
Graphics-based:(Point and Click, Drag and Drop, etc.).
Natural language: requests in written English.
Combinations of the above: For example, both menus and forms used
extensively in Web database interfaces.
Other DBMS Interfaces
Speech as Input and Output
Web Browser as an interface
Parametric interfaces, e.g., bank tellers using function keys.
Interfaces for the DBA: Creating user accounts, granting authorizations
Setting system parameters Changing schemas or access paths.
3.6
The Database System Environment
The DBMS is an important component of the database environment. This
environment also includes different types of hardware and software, people
who perform different functions within the environment, procedures
designed to accomplish desired activities, and data. The data constitute the
database's central component through which information is generated. So,
the main purpose of the database environment is to help an organization to
perform its mission and to achieve its goals.
It is worthy of note that, the database administrator writes and enforces the
procedures and standards that are then used by designers, analysts,
programmers, and end users. The end users use the application programs
created by analysts and programmers. In turn, the application programs
make use of the DBMS, which manages the data. Note also that the
database designer and the database administrator perform their jobs through
the interface – that would be shown on the computer. The interface is the
gateway to the database, which resides within the hardware. Finally, the
System administrator manages the entire system.
Components of DB environment:
CASE tools : Automated tools to design DBs & apps. programs.
Repository : Centralised knowledge base containing all data
definitions, screen & report formats & definitions.
DBMS : Commercial S/W used to create, maintain & provide
controlled access to the DB & the repository (definitions of data).
DB : A shared collection of data, designed to meet the information
needs of users in an organisation (occurrences of data).
Application programs : Computer programs that are used to create &
maintain DB & provide information to users.
User Interface: Languages, menus, and other facilities by which
users interact with various system components, such as CASE tools,
application programs, the DBMS and the repository.
Data administrators: Responsible for overall information resources
of an organization .
System developers: System analysts, programmers who design new
application programs.
End users: add, delete & modify data in DBs & receive information
from it throughout the organization .
Figure 10: The Database Environment Components
3.7
Database System Utilities and Tools
Database System Utilities: These are used to perform certain functions
such as:
• Loading data stored in files into a database. Includes data conversion
tools.
• Backing up the database periodically on tape.
• Reorganizing database file structures.
• Report generation utilities.
• Performance monitoring utilities.
• Other functions, such as sorting, user monitoring, data compression,
etc.
Other Tools: These include data dictionary / repository, used to store
schema descriptions and other information such as design decisions,
application program descriptions, user information, usage standards, etc.
Active data dictionary is accessed by DBMS software and users/DBA.
Passive data dictionary is accessed by users/DBA only.
Application Development Environments and CASE (computer-aided
software engineering) tools:
Examples: PowerBuilder (Sybase) , JBuilder (Borland), JDeveloper
10G (Oracle), Databrid, DTM ODBC Manager, mysql data syncronization,
myDBR, AeroSQL, Data-tier Applications etc.
Figure 11: Typical DBMS Component Module
3.8
Centralized and Client-Server Architectures
Centralized DBMS: Combines everything into single system includingDBMS software, hardware, application programs, and user interface
processing software. User can still connect through a remote terminal –
however, all processing is done at centralized site.
Figure 12:
A Physical Centralized Architecture
Client-server: This is a software architecture model consisting of two parts,
client systems and server systems, both communicate over a computer
network or on the same computer. A client-server application is a
distributed system consisting of both client and server software. The client
process always initiates a connection to the server, while the server process
always waits for requests from any client.
When both the client process and server process are running on the same
computer, this is called a single seat setup but in peer-to-peer, each host or
application instance can simultaneously act as both a client and a server
(unlike centralized servers of the client-server model) and because each has
equivalent responsibilities and status. Peer-to-peer architectures are often
abbreviated using the acronym P2P.
A one server based network.
A peer-to-peer based network.
Figure 13: Client Server Networks
Basic 2-tier Client-Server Architectures
Specialized Servers with Specialized functions
Print server
File server
DBMS server
Web server
Email server
Clients can access the specialized servers as needed
Figure 14 : Logical two-tier client server architecture
Two Tier Client-Server Architecture:
This is a client program that may connect to several DBMSs, sometimes
called the data sources. In general, data sources can be files or other nonDBMS software that manages data. Other variations of clients are possible:
e.g., in some object DBMSs, more functionality is transferred to clients
including data dictionary functions, optimization and recovery across
multiple servers, etc.
Three Tier Client-Server Architecture:
This is very common for Web applications, Intermediate Layer called
Application Server or Web Server. Stores the web connectivity software
and the business logic part of the application used to access the
corresponding data from the database server. Acts like a conduit for sending
partially processed data between the database server and the client.
Three-tier Architecture Can Enhance Security:
Database server only accessible via middle tier
Clients cannot directly access database server
Figure 15: Three-tier client-server architecture
Clients: The client–server model of computing is a distributed application
structure that partitions tasks or workloads between the providers of a
resource or service, called servers, and service requesters, called clients.
It provides appropriate interfaces through a client software module to access
and utilize the various server resources. Clients may be diskless machines
or PCs or Workstations with disks with only the client software installed. It
is connected to the servers via some form of a network (LAN: local area
network, wireless network, etc.)
Clients characteristics
•
•
•
•
•
Always initiates requests to servers.
Waits for replies.
Receives replies.
Usually connects to a small number of servers at one time.
Usually interacts directly with end-users using any user interface such
as graphical user interface.
DBMS Server provides database query and transaction services to the
clients. Relational DBMS servers are often called SQL servers, query
servers, or transaction servers. Applications running on clients utilize an
Application Program Interface (API) to access server databases via
standard interface such as:
ODBC: Open Database Connectivity standard
JDBC: for Java programming access
Client and server must install appropriate client module and server
module software for ODBC or JDB
Server characteristics
•
Always wait for a request from one of the clients.
•
Serve clients requests then replies with requested data to the clients.
•
A server may communicate with other servers in order to serve a
client request.
3.9
Classification of DBMSs
Based on the data model used:
Traditional: Relational, Network, Hierarchical.
Emerging: Object-oriented, Object-relational.
Other classifications
Single-user (typically used with personal computers) vs. multi-user
(most DBMSs).
Centralized (uses a single computer with one database) vs. distributed
(uses multiple computers, multiple databases)
4.0
Conclusion
With the introduction to concepts of database system, students must have
acquired the basic sense of the central importance of databases in today’s
information systems environment. And with this, they should be able to
function very well in such environment.
5.0
Summary
In this unit we have learnt that:
Data model is a set of concepts used to describe the structures of a
database, and the operations for manipulating these structures, as well
as certain constraints that the database should obey.
Database models include flat, hierarchical network, relational,
concept-oriented and star-schema.
Data architecture is the design of data for use in defining the target
state and the subsequent planning needed to hit the target state.
Database system environment includes different types of hardware
and software, people who perform different functions within the
environment, procedures designed to accomplish desired activities,
and data.
6.0
Tutor Marked Assignment
1. (a) Define data and database model
(b) What do you understand by data independence and its levels
2. (a) Explain database environment and its components
7.0
Further Reading and Other Resources
Paul R. Smith & Richard Sarfaty (1993). Creating a strategic plan for
configuration management using Computer Aided Software Engineering
(CASE) tools. Paper For 1993 National DOE/Contractors and Facilities
CAD/CAE User's Group.
Data Modeling Made Simple 2nd Edition", Steve Hoberman, Technics
Publications, LLC 2009
Michael R. McCaleb (1999). "A Conceptual Data Model of Datum
Systems". National Institute of Standards and Technology. August 1999.
Matthew West and Julian Fowler (1999). Developing High Quality Data
Models. The European Process Industries STEP Technical Liaison
Executive (EPISTLE).
American National Standards Institute. 1975. ANSI/X3/SPARC Study Group
on Data Base Management Systems; Interim Report. FDT (Bulletin of
ACM SIGMOD) 7:2.
Data Warehouse tool DWE - The most easy to use Data Warehouse tool.
BI.dwexplorer.com/datawarehousetool
Riversand MDM Multi-entity Master Data Mgmt. Global, enterprise-class
solutions www.riversand.com
"Distributed Application Architecture". Sun Microsystem.
http://java.sun.com/developer/Books/jdbc/ch07. p
Module 1 Fundamentals of Database Systems
Unit 3 Data Modelling Using the Entity-Relationship Model
1.0
Introduction
2.0
Objective
3.0
Using High-Level Conceptual Data Models for Database Design
3.1
A Sample Database Application
3.2
Entity Types, Entity Sets, Attributes, and keys
3.3
Relationships, Relationship Types, Roles, and Structural
Constraints
3.4
Weak Entity Types
3.5
Refining the ER Design for the Company Database
3.6
ER Diagrams, Naming Conventions, and Design Issues
4.0
Conclusion
5.0
Summary
6.0
Tutor Marked Assignment
7.0
Further Reading and Other Resources
1.0
Introduction
Traditionally, the design and testing of application programs have been
considered to be more in the realm of the software engineering domain than
in the database domain. As database design methodologies include more of
the concepts for specifying operations on database objects, and as software
engineering methodologies specify in more detail the structure of the
databases that software programs will use and access, it is clear that these
activities are strongly related. Conceptual modelling is a very important
phase in designing a successful database application.
Generally, the term database application refers to a particular database and
the associated programs that implement the database queries and updates.
In this unit, the traditional approach of concentrating on the database
structures and constraints during database design is strictly followed. The
modelling concepts of the Entity-Relationship(ER) model is herewith
presented, which is a popular high-level conceptual data model.
This model and its variations are frequently used for the conceptual design
of database applications, and many database design tools employ its
concepts. The basic data-structuring concepts and constraints of the ER
model as well as their use in the design of conceptual schemas for database
applications are described in this unit. Diagrammatic notation associated
with the ER model, known as ER diagrams is also presented.
2.0
Objective
At the end of this unit, you should be able to:
appreciate the features of Entity Types, Entity Sets, Attributes, and
keys of database management system (DBMS)
Use High-Level Conceptual Data Models for Database Design
State the major concepts Relationships, Relationship Types, Roles,
and Structural Constraints
Draw the major components of ER Diagrams, Naming Conventions,
and Design Issues in data management system.
3.0
Using High-Level Conceptual Data Models
for Database Design
Figure below shows a simplified description of the database design
process. The first step shown is requirements collection and analysis.
During this step, the database designers interview prospective database
users to understand and document their data requirements. The result of
this step is a concisely written set of users’ requirements. These requirements should be specified in as detailed and complete a form as possible.
In parallel with specifying the data requirements, it is useful to specify the
known functional requirements of the application. These consist of the
user-defined operations (or transactions) that will be applied to the
database, including both retrievals and updates. In software design, it is
common to use data flow diagrams, sequence diagrams, scenarios, and
other techniques for specifying functional requirements. We will not discuss
any of these techniques here because they are usually described in detail in
software engineering texts.
Once all the requirements have been collected and analysed, the next step is
to create a conceptual schema for the database, using a high-level
conceptual data model. This step is called conceptual design. The concepttual schema is a concise description of the data requirements of the users
and includes detailed descriptions of the entity types, relationships, and
constraints; these are expressed using the concepts provided by the highlevel data model. Because these concepts do not include implementation
details, they are usually easier to understand and can be used to
communicate with non-technical users.
The high-level conceptual schema can also be used as a reference to ensure
that all users’ data requirements are met and that the requirements do not
conflict. This approach enables the database designers to concentrate on
specifying the properties of the data, without being concerned with storage
details. Consequently, it is easier for them to come up with a good
conceptual database design.
During or after the conceptual schema design, the basic data model
operations can be used to specify the high-level user operations identified
during functional analysis. This also serves to confirm that the conceptual
schema meets all the identified functional requirements. Modifications to
the conceptual schema can be introduced if some functional requirements
cannot be specified using the initial schema.
The next step in database design is the actual implementation of the
database, using a commercial DBMS. Most current commercial DBMSs use
an implementation data model, such as the relational or the object-relational
database model—so the conceptual schema is transformed from the highlevel data model into the implementation data model. This step is called
logical design or data model mapping, and its result is a database schema
in the implementation data model of the DBMS.
The last step is the physical design phase, during which the internal storage
structures, indexes, access paths, and file organizations for the database
files are specified. In parallel with these activities, application programs are
designed and implemented as database transactions corresponding to the
high-level transaction specifications.
Figure 16: A Simplified Diagram of Phases of database Design
3.1
A Sample Database Application
In this section we describe an example database application, called
COMPANY that serves to illustrate the basic ER model concepts and their
use in schema design. We list the data requirements for the database here,
and then create its conceptual schema step by step as the modelling
concepts of the ER model are introduced. The COMPANY database keeps
track of a company’s employees, departments, and projects. Suppose that
after the requirements collection and analysis phase, the database designers
provided the following description of the “miniworld”—the part of the
company to be represented in the database:
1. The company is organized into departments. Each department has a
unique name, a unique number, and a particular employee who manages the
department. We keep track of the start date when that employee began
managing the department. A department may have several locations.
2. A department controls a number of projects, each of which has a unique
name, a unique number, and a single location.
3. We store each employee’s name, social security number, address, salary,
sex, and birth date. An employee is assigned to one department but may
work on several projects, which are not necessarily controlled by the same
department. We keep track of the number of hours per week that an
employee works on each project. We also keep track of the direct
supervisor of each employee.
4. We want to keep track of the dependents of each employee for insurance
purposes. We keep each dependent’s first name, sex, birth date, and
relationship to the employee.
Figure 3.2 shows how the schema for this database application can be
displayed by means of the graphical notation known as ER diagrams. We
describe the step-by-step process of deriving this schema from the stated
requirements—and explain the ER diagrammatic notation—as we introduce
the ER model concepts in the following section.
3.2
Entity Types, Entity Sets, Attributes, and keys
The ER model describes data as entities, relationships, and attributes:
Entities and Their Attributes: The basic object that the ER model
represents is an entity, which is a “thing” in the real world with an
independent existence. An entity may be an object with a physical existence
(for example, a particular person, car, house, or employee) or it may be an
object with a conceptual existence (for example, a company, a job, or a
university course). Each entity has attributes—the particular properties that
describe it. For example, an employee entity may be described by the
employee’s name, age, address, salary, and job. A particular entity will have
a value for each of its attributes. The attribute values that describe each
entity become a major part of the data stored in the database.
Figure 17:
Alternative ER Notation
Figure 3.3 shows two entities and the values of their attributes. The
employee entity e1 has four attributes: Name, Address, Age, and
HomePhone; their values are “John Smith,” “2311 Kirby, Houston, Texas
77001,” “55,” and “713-749-2630,” respectively. The company entity c1
has three attributes: Name, Headquarters, and President; their values are
“Sunco Oil,” “Houston,” and “John Smith,” respectively.
Figure 18:
1
Several types of attributes occur in the ER model:
In ER model we have the following attributes: simple versus composite,
single-valued versus multi-valued, and stored versus derived.
Composite versus Simple (Atomic) Attributes: Composite attributes
can be divided into smaller subparts, which represent more basic attributes
with independent meanings. For example, the Address attribute of the
employee entity shown in Figure 3.3 can be subdivided into StreetAddress,
City, State, and Zip,3 with the values “2311 Kirby,”“Houston,” “Texas,”
and “77001.” Attributes that are not divisible are called simple or atomic
attributes. Composite attributes can form a hierarchy; for example,
StreetAddress can be further subdivided into three simple attributes:
Number, Street, and ApartmentNumber, as shown in Figure 3.4. The value
of a composite attribute is the concatenation of the values of its constituent
simple attributes.
Figure 19:19
Single-Valued versus Multivalued Attributes: Most attributes have a
single value for a particular entity; such attributes are called single-valued.
For example, Age is a single-valued attribute of a person. In some cases an
attribute can have a set of values for the same entity—for example, a Colors
attribute for a car, or a CollegeDegrees attribute for a person. Cars with one
color have a single value, whereas two-tone cars have two values for
Colors. Similarly, one person may not have a college degree, another person
may have one, and a third person may have two or more degrees; therefore,
different persons can have different numbers of values for the
CollegeDegrees attribute. Such attributes are called multivalued.
Stored versus Derived Attributes. In some cases, two (or more) attribute
values are related—for example, the Age and BirthDate attributes of a
person. For a particular person entity, the value of Age can be determined
from the current (today’s) date and the value of that person’s BirthDate.
The Age attribute is hence called a derived attribute and is said to be
derivable from the BirthDate attribute, which is called a stored attribute.
In general, composite and multi-valued attributes may be nested arbitrarily to any
number of levels although this is rare. For example, PreviousDegrees of a STUDENT
is a composite multi-valued attribute denoted by {PreviousDegrees (College, Year,
Degree, Field)}.
Entity Types, Entity Sets, and Keys
Entity Types and Entity Sets: A database usually contains groups of
entities that are similar. For example, a company employing hundreds of
employees may want to store similar information concerning each of the
employees. These employee entities share the same attributes, but each
entity has its own value(s) for each attribute. An entity type defines a
collection (or set) of entities that have the same attributes. Each entity type
in the database is described by its name and attributes. Figure shows two
entity types, named EMPLOYEE and COMPANY, and a list of attributes
for each.
A few individual entities of each type are also illustrated, along with the
values of their attributes. The collection of all entities of a particular entity
type in the database at any point in time is called an entity set; the entity set
is usually referred to using the same name as the entity type. For example,
EMPLOYEE refers to both a type of entity as well as the current set of all
employee entities in the database.
An entity type describes the schema or intension for a set of entities that
share the same structure. The collection of entities of a particular entity type
is grouped into an entity set, which is also called the extension of the entity
type.
Key Attributes of an Entity Type: An important constraint on the entities
of an entity type is the key or uniqueness constraint on attributes. An
entity type usually has an attribute whose values are distinct for each
individual entity in the entity set. Such an attribute is called a key attribute,
and its values can be used to identify each entity.
Figure 20:
Figure 21:
Figure 22:
.
3.3
Relationships, Relationship Types, Roles,
and Structural Constraints
In Figure 17 above, there are several implicit relationships among the
various entity types. In fact, whenever an attribute of one entity type refers
to another entity type, some relationship exists. For example, the attribute
Manager of DEPARTMENT refers to an employee who manages the
department; the attribute Controlling Department of PROJECT refers to the
department that controls the project; the attribute Supervisor of
EMPLOYEE refers to another employee (the one who supervises this
employee); the attribute Department of EMPLOYEE refers to the
department for which the employee works; and so on. In the ER model,
these references should not be represented as attributes but as
relationships, which are discussed in this section.
A relationship relates two or more distinct entities with a specific
meaning.
For example: EMPLOYEE John Smith works on the ProductX PROJECT or
EMPLOYEE Franklin Wong manages the Research DEPARTMENT.
A relationship type R among n entity types E1, E2, . . . , En defines a set
of associations or a relationship set—among entities from these entity
types. Relationships of the same type are grouped or typed into a
relationship type.
For example, the WORKS_ON relationship type in which
EMPLOYEEs and PROJECTs participate, or the MANAGES
relationship type in which EMPLOYEEs and DEPARTMENTs
participate. The degree of a relationship type is the number of
participating entity types. Both MANAGES and WORKS_ON are
binary relationships.
Figure 23:
Role Names and Recursive Relationships: Each entity type that
participates in a relationship type plays a particular role in the relationship.
The role name signifies the role that a participating entity from the entity
type plays in each relationship instance, and helps to explain what the
relationship means. For example, in the WORKS_FOR relationship type,
EMPLOYEE plays the role of employee or worker and DEPARTMENT
plays the role of department or employer. Role names are not technically
necessary in relationship types where all the participating entity types are
distinct, since each participating entity type name can be used as the role
name.
However, in some cases the same entity type participates more than once in
a relationship type in different roles. In such cases the role name becomes
essential for distinguishing the meaning of each participation. Such
relationship types are called recursive relationships.
CONSTRAINTS ON RELATIONSHIP TYPES
Relationship Types (also known as ratio constraints) usually have certain
constraints that limit the possible combinations of entities that may
participate in the corresponding relationship set. These constraints
are determined from the miniworld situation that the relationships represent.
For example, in Figure 22, if the company has a rule that each employee
must work for exactly one department, then we would like to describe this
constraint in the schema. We can distinguish two main types of relationship
constraints: cardinality ratio and participation:
Cardinality Ratios for Binary Relationships. The cardinality ratio for a
binary relationship specifies the maximum number of relationship instances
that an entity can participate in. For example, in the WORKS_FOR binary
relationship type, DEPARTMENT:EMPLOYEE is of cardinality ratio 1:N,
meaning that each department can be related to (that is, employs) any
number of employees, but an employee can be related to (work for) only
one department. The possible cardinality ratios for binary relationship types
are 1:1, 1:N, N:1, and M:N. Cardinality ratios for binary relationships are
represented on ER diagrams by displaying 1, M, and N on the diamonds as
shown in Figure 17 above.
Maximum Cardinality can be in the form of:
One-to-one (1:1)
One-to-many (1:N) or Many-to-one (N:1)
Many-to-many
The participation constraint specifies whether the existence of an entity
depends on its being related to another entity via the relationship type. This
constraint specifies the minimum number of relationship instances that each
entity can participate in, and is sometimes called the minimum cardinality
constraint.
Minimum Cardinality (also called participation constraint or existence
dependency constraints) can also be in the form of:
3.4
zero (optional participation, not existence-dependent)
one or more (mandatory, existence-dependent)
Weak Entity Types
Entity types that do not have key attributes of their own are called weak
entity types. In contrast, regular entity types that do have a key attribute
are called strong entity types. Entities belonging to a weak entity type are
identified by being related to specific entities from another entity type in
combination with one of their attribute values. We call this other entity type
the identifying or owner entity type, and we call the relationship type that
relates a weak entity type to its owner the identifying relationship of the
weak entity type. A weak entity type always has a total participation
constraint (existence dependency) with respect to its identifying
relationship, because a weak entity cannot be identified without an owner
entity. However, not every existence dependency results in a weak entity
type. For example, a DRIVER_LICENSE entity cannot exist unless it is
related to a PERSON entity, even though it has its own key
(LicenseNumber) and hence is not a weak entity.
In ER diagrams, both a weak entity type and its identifying relationship are
distinguished by surrounding their boxes and diamonds with double lines
(see Figure 17). The partial key attribute is underlined with a dashed or
dotted line.
Weak entity types can sometimes be represented as complex (composite,
multi-valued) attributes.
3.5
Refining the ER Design for the Company
Database
We can now refine the database design of Figure 22, by changing the
attributes that represent relationships into relationship types. The cardinality
ratio and participation constraint of each relationship type are determined
from the requirements listed earlier. If some cardinality ratio or dependency
cannot be determined from the requirements, the users must be questioned
further to determine these structural constraints. In our example, we specify
the following relationship types:
1. MANAGES, a 1:1 relationship type between EMPLOYEE and
DEPARTMENT. EMPLOYEE participation is partial. DEPARTMENT
participation is not clear from the requirements. We question the users, who
say that a department must have a manager at all times, which implies total
participation.14 The attribute StartDate is assigned to this relationship type.
2. WORKS_FOR, a 1:N relationship type between DEPARTMENT and
EMPLOYEE. Both participations are total.
3. CONTROLS, a 1:N relationship type between DEPARTMENT and
PROJECT. The participation of PROJECT is total, whereas that of
DEPARTMENT is determined to be partial, after consultation with the
users indicates that some departments may control no projects.
4. SUPERVISION, a 1:N relationship type between EMPLOYEE (in the
supervisor role) and EMPLOYEE (in the supervisee role). Both participations are determined to be partial, after the users indicate that not every
employee is a supervisor and not every employee has a supervisor.
5. WORKS_ON, determined to be an M:N relationship type with attribute
Hours, after the users indicate that a project can have several employees
working on it. Both participations are determined to be total.
3.6
ER Diagram and Naming Conventions
Figure 3.2 above displays the COMPANY ER database schema as an ER
diagram. We now review the full ER diagram notation.
General Database Naming Conventions
The following conventions apply to all of the elements of a database:
•
All names used throughout the database should be lowercase only.
This will eliminate errors related to case-sensitivity.
•
Separate name parts by underlines, never by spaces. This way, you
improve the readability of each name (e.g. product_name instead of
productname). You will not have to use parentheses or quotes to
enclose names using spaces as well. The use of spaces in a database
name is allowed only on some systems, while the underline is an
alphanumeric character, allowed on any platform. Thus, your database
will become platform independent.
•
Do not use numbers in the names (e.g. product_attribute1). This is
proof of poor design, indicating a badly divided table structure. If you
need a many-to-many relation, the best way to achieve it is by using a
separate linking table. See how here. Moreover, using numbers to
differentiate between two columns that store similar information
might be an indication that you need an extra table, storing that
information. For instance, having a location1 column in a
manufacturers table and a location2 column in a distributors table
could be solved by creating a separate table that stores all locations,
and that is referenced by both the manufacturers and distributors
tables via foreign keys.
•
Do not use the dot (.) as a separator in names. This way you will avoid
problems when trying to perform queries, as the dot is used to identify
a field in a specific column. In SQL language,
manufacturer_man.address_man means the address column from the
table that stores information about manufacturers.
•
Do not use any of the reserved words as names of database elements.
Each database language uses some words as names for internal
functions, or as part of the SQL syntax. For instance, using order as
the name of a table that stores product orders from an online shop is
bad practice, because order is also used in SQL language to sort
records (ascending or descending).
For a complete list of the reserved words that you should not use when
naming the database elements, consult your specific database software
manual. See the list of reserved words for MySQL here.
•
When naming the elements, do not use long or awkward names. Keep
them as simple as you can, while maintaining a clear meaning. It's
also a good idea to use names which are close to the natural language:
description_prd is certainly a better name for a column that stores
product descriptions than dscr_pr or some generic name as field.
Database names
Each database must have a name of its own, which should also follow some
conventions:
•
•
Use the project name as the name of the database.
Prefix the database name with the owner name, separated by an
underscore. The owner might be a person (e.g. the project manager) or
the application for which the database is created. For example, acme
catalog can be a good name for the database storing
the product catalog of the ACME company.
Table names
Tables are some of the most common elements used in an application, as
they store
the columns, and as such are mentioned in each query. Therefore, the
following conventions should apply to tables:
•
Table names contain the name of the entity that is being defined,
followed by a three letter acronym of that name (e.g. category_ctg).
Optionally, you can use the same prefix for all tables in the same
database. For example, acme_product_prd, acme_manufacturer_prd,
acme_category_prd can be tables from the acme_catalog database.
•
Prefix tables that define the same larger entity with a 2 or 3-letter
acronym that identifies it. For example, e.g. hr_applicant_app,
hr_job_job and hr_resume_rsm are
all tables that belong to the Human Resources Department of a large
corporation database.
•
Do not use generic prefixes, such as tbl_, or db_, as they are
redundant and useless.
•
Use short, unambiguous names for each table, restricted to one word,
if possible. This way tables can be distinguished easily.
•
Use singular for table names. This way, you avoid errors due to the
pluralization of English nouns in the process of database development.
For instance, activity becomes activities, box becomes boxes, person
becomes people or persons, while data remains data.
•
Use clear names. Do not overdo it using abbreviations and acronyms.
While using a shorter name might help the developers, it makes the
meaning less clear to other members of the team. Using clear names
makes the design self-explanatory.
•
Prefix lookup tables with the name of the table they relate to. This
helps group related tables together (e.g. product_type, product_status),
and also helps prevent name conflicts between generic lookup tables
for different entities. You can have more than one generic lookup
table for an existing master table, but which address different
properties of the elements in the table.
Column names
Columns are attributes of an entity, describing its properties. Therefore, the
name they carry should be natural, and as meaningful as possible. The
following conventions are recommended:
•
All keys are used for indexing and identifying records. Therefore, it's
a good practice to put the id particle in their name. This way, you'll
know the field is used as a key.
•
The primary key is used to uniquely identify each record. That is why
its name should be made up of the id particle, followed by the table
name acronym. For instance, for the table product_prd, the primary
key is id_prd.
•
The foreign key name should be composed by the id particle, followed
by the acronym of the referred table, and then by the acronym of the
table it belongs to. For example, the idctg_prd foreign key belongs to
the products table (product_prd), but it refers to the categories table
(category_ctg). This way, the table being referenced is obvious from
the key name.
•
Each column name should be followed by the 3-letter table acronym.
This way, each column has a unique name across the database.
Without the table acronym, you would end up having two columns
called "name", one storing the product name and the other the
manufacturer name. Instead, name_prd and name_man can easily be
distinguished.
•
Date columns should use the “date_” prefix, and the boolean type
columns should use the “is_” prefix. For instance, date birth stores the
birth date of a person, while is confirmed could indicate the order
status for a product in a shop, using true/false values (or 0/1).
4.0
Conclusion With the introduction of High-Level
Conceptual Data Models for Database Design. We have explained all
the features of Entity Types, Entity Sets, Attributes, and keys of
database management system (DBMS) in relationships to the major
concepts Relationships and Structural Constraints. Students must have
acquired the basic concepts of the conceptual importance of data
models in today’s designing systems. And with this, they should be
able to function very well in such environment.
5.0
Summary
In this unit we have learnt that:
The features of Entity Types, Entity Sets, Attributes, and keys of
database management system (DBMS) as a set of concepts used to
describe the structures of a database management and the operations
for manipulating these structures, as well as certain constraints that the
database should obey.
The use of High-Level Conceptual Data Models for Database Design
The major concepts of
Relationships, Relationship Types, Roles, and
Structural Constraints in database management systems (DBMS)
How to draw the major components of ER Diagrams, Naming
Conventions and Design Issues in data management system
6.0
Tutor Marked Assignment
Explain the use of High-Level Conceptual Data Models for Database Design is
all about.
Differentiate between the major concepts of Relationship Types and
Structural Constraints in database management systems (DBMS)
Draw the major components of ER Diagrams, Naming
Conventions and Design Issues in database management system
7.0
Further Reading and Other Resources
Stephen M. Richard (1999). Geologic Concept Modelling. U.S. Geological
Survey Open-File Report 99-386.
Joachim Rossberg and Rickard Redler (2005). Pro Scalable .NET 2.0
Application Designs.. Page 27
Terry Halpin (2001). "Object-Role Modelling: an overview"
The ORM Foundation home page
Terry Halpin (2001), Object-Role Modelling: an overview
Terry Halpin (2005), ORM2 On the Move to Meaningful Internet Systems
2005: OTM 2005 Workshops, eds R. Meersman, Z. Tari, P. Herrero et al.,
Cyprus. Springer LNCS 3762, pp 676–87.
Halpin, Terry (1989), Conceptual Schema and Relational Database Design,
Sydney: Prentice Hall, ISBN 978-0131672635
Rossi, Matti; Siau, Keng (April 2001), Information Modelling in the New
Millennium, IGI Global, ISBN 978-1878289773
The Entity Relationship Model: Toward a Unified View of Data" for entityrelationship modelling.
A.P.G. Brown, "Modelling a Real-World System and Designing a Schema
to Represent It", in
Douque and Nijssen (eds.), Data Base Description, North-Holland, 1975,
ISBN 0-7204-2833-5.
Paul Beynon-Davies (2004). Database Systems. Houndmills, Basingstoke,
UK: Palgrave
"Gliffy March 2007 NewsLetter", March 1, 2007, accessed January 13,
2011
Richard Barker (1990). CASE Method: Tasks and Deliverables.
Wokingham, England: Addison-Wesley.
Paul Beynon-Davies (2004). Database Systems. Houndmills, Basingstoke,
UK: Palgrave
Module 2
The Statistical database system
Unit 1 Statistical Database Concepts
1.0
Introduction
2.0
Objective
3.0
Basic Definition
3.1
Concept of Public Policies on Statistical Data
3.2
The Design of a Statistical Database
3.3
Features and Requirements of a Statistical Database
3.4
Statistical Database Modelling
3.5
Some Statistical Data Models
4.0
Conclusion
5.0
Summary
6.0
Tutor Marked Assignment
7.0
Further Reading and Other Resources
1.0
Introduction
A statistical database contains confidential information about individuals or
events. These databases are mainly used to generate statistical information
about the stored information. But Statistical databases have some nonstandard characteristics which cannot be well supported by commercially
available database management systems. This concerns the underlying data
structures, the various abstraction levels of the data, and the type of
operations on the data and the kind of processing requirements. The current
research is aiming at an appropriate conceptual modelling, an efficient
representation of the data and a powerful and user-friendly query processing
on each data level.
2.0
Objectives
At the end of this unit, you should be able to:
Appreciate the Features of Statistical database system (DBMS)
Define Statistical database system.
Use High-Level Concept of Statistics data in policies formulation.
State some major concepts of Statistical database Models.
Design the major components of Statistical database and Modelling.
3.0
Basic Definition
Databases that are mainly used for statistical analysis are called statistical
databases (SDB). And a statistical database management system
(SDBMS) may be defined as a database management system that provides
capabilities to model, store, and manipulate data in a manner suitable for the
needs of SDB users. It is also a database management system that provides
capabilities to apply statistical data analysis techniques that range from
simple summary statistics to advanced procedures.
In addition, a statistical database (SDB) is one that provides data of a
statistical nature, such as counts and averages. The term statistical database
is used in two contexts:
•
Pure statistical database: This type of database only stores statistical
data. An example is a census database. Typically, access control for a
pure SDB is straightforward: Certain users are authorized to access the
entire database.
•
Ordinary database with statistical access: This type of database
contains individual entries; this is the type of database discussed so far
in this chapter. The database supports a population of non-statistical
users who are allowed access to selected portions of the database
using DAC, RBAC, or MAC. In addition, the database supports a set
of statistical users who are only permitted statistical queries. For these
latter users, aggregate statistics based on the underlying raw data are
generated in response to a user query, or may be pre-calculated and
stored as part of the database.
3.1
Concept of Public Policies on Statistical Data
A statistical database contains confidential information about individuals
or events. These databases are mainly used to generate statistical
information about the stored information. Such databases accept only
statistical queries , which involve statistical functions such as SUM, AVG,
COUNT, MIN, MAX , and so on. However, users are not allowed to
retrieve information about a particular individual.
Consider a relation BankEmp with the attributes ECode, EName, Sex,
State, Salary, Branch , and Designation . Using statistical queries, one can
retrieve the number of clerks, maximum salary, average salary of clerks,
and so on. However, users are not allowed to retrieve the salary of a
particular employee.
Data Confidentiality and Disclosure
Confidentiality should mean that, the dissemination of data in a manner
that would allow public identification of the respondent or would in any
way be harmful to him is prohibited and that the data are immune from
legal process. Confidentiality differs from privacy because it applies to
business as well as individuals. Privacy is an individual right whereas
confidentiality often applies to data on organizations and firms.
Statistical disclosure occurs when released statistical data (either tabular or
individual records) reveal confidential information about an individual
respondent.
Disclosure relates to inappropriate attribution of information to a data
subject, whether an individual or an organization. Disclosure occurs when a
data subject is identified from a released file (identity disclosure), sensitive
information about a data subject is revealed through the released file
(attribute disclosure), or the released data make it possible to determine
the value of some characteristic of an individual more accurately than
otherwise would have been possible (inferential disclosure).
Note that each type of disclosure can occur in connection with the release of
either tables or microdata. The definitions and implications of these three
kinds of disclosure are discussed below: Identity disclosure occurs if a
third party can identify a subject or respondent from the released data.
Revealing that an individual is a respondent or subject of a data collection
may or may not violate confidentiality requirements.
For tabulations, revealing identity is generally not disclosure, unless the
identification leads to divulging confidential information (attribute
disclosure) about those who are identified.
For microdata, identification is generally regarded as disclosure, because
microdata records are usually so detailed that identification will
automatically reveal additional attribute information that was not used in
identifying the record. Hence disclosure limitation methods applied to
microdata files limit or modify information that might be used to identify
specific respondents or data subjects.
Attribute disclosure occurs when confidential information about a data
subject is revealed and can be attributed to the subject. Attribute disclosure
occurs when confidential information about a person or firm’s business
operations is revealed or may be closely estimated. Thus, attribute
disclosure comprises identification of the subject and divulging confidential
information pertaining to the subject.
Attribute disclosure is the primary concern of most statistical agencies in
deciding whether to release tabular data. Disclosure limitation methods
applied to tables assure that respondent data are published only as part of an
aggregate with a sufficient number of other respondents to disguise the
attributes of a single respondent.
Inferential disclosure, occurs when individual information can be
inferred with high confidence from statistical properties of the released
data. For example, the data may show a high correlation between income
and purchase price of home. As purchase price of home is typically public
information, a third party might use this information to infer the income of a
data subject. There are two main reasons that some statistical agencies are
not concerned with inferential disclosure in tabular or micro data. First a
major purpose of statistical data is to enable users to infer and understand
relationships between variables.
Tables, Microdata, and On-Line Query Systems
The choice of statistical disclosure limitation methods depends on the
nature of the data products whose confidentiality must be protected. Most
statistical data are released in the form of tables, microdata files, or through
on-line query systems. Tables can be further divided into two categories:
tables of frequency (count) data and tables of magnitude data. For either
category, data can be presented in the form of numbers, proportions or
percentages.
A microdata file consists of individual records, each containing values of
variables for a single person, business establishment or other unit. Some
microdata files include direct identifiers, such as name, address or Social
Security number. Removing any of these identifiers is an obvious first step
in preparing for the release of a file for which the confidentiality of
individual information must be protected.
Historically, disclosure limitation methods for tables were applied directly
to the tables. Methods include redesign of tables, suppression, controlled
and random rounding. More recent methods have focused on protecting the
microdata underlying the tables using some of the microdata protection
techniques. In this way all tables produced from the protected microdata
are also protected. This may be done whether there is an intention to release
the microdata or not. It is a particularly useful way to protect tables
produced from on-line query systems.
Restricted Data and Restricted Access
The confidentiality of individual information can be protected by restricting
the amount of information provided or by adjusting the data in released
tables and microdata files (restricted data) or by imposing conditions on
access to the data products (restricted access), or by some combination of
these.
Tables of Magnitude Data Versus Tables of Frequency Data
The selection of a statistical disclosure limitation technique for data
presented in tables (tabular data) depends on whether the data represent
frequencies or magnitudes. Tables of frequency count data present the
number of units of analysis in a cell. Equivalently the data may be
presented as a percent by dividing the count by the total number presented
in the table (or the total in a row or column) and multiplying by 100. Tables
of magnitude data present the aggregate of a "quantity of interest" that
applies to units of analysis in the cell. Equivalently the data may be
presented as an average by dividing the aggregate by the number of units in
the cell. To distinguish formally between frequency count data and
magnitude data, the "quantity of interest" must measure something other
than membership in the cell. Thus, tables of the number of establishments
within the manufacturing sector by SIC group and by county-within-state
are frequency count tables, whereas tables presenting total value of
shipments for the same cells are tables of magnitude data.
3. 2
The Design of a Statistical Database (Micro-,
Macro and Metadata Modelling)
A database is called a statistical database (SDB) if it contains data of three
kinds:
Microdata as primary or basis data on individuals, objects or events
representing sampled, census or collected data.
Macrodata as grouped or aggregated data (summarized data) which are
cross-classified by a set of categorical attributes(variables). The summary
attribute represents counts(frequencies), means, indices or other statistics
characterizing a set(population) of individuals, objects or events.
Metadata describing the micro- and macrodata on the semantic, structural,
statistical and physical level in such a way that they can be stored, transformed retrieved and transmitted in a reasonable way. It covers the whole data
life cycle, i.e. the data collecting from the data source, the data storing, the
data processing and retrieval, and the data disseminating within the
electronic data interchange (EDI). As a matter of fact, the metadata itself
must be easily accessible similiar to the micro- and macrodata. We close the
introduction with some examples of typical retrieval operations (queries) on
the data of the above kind. We make use of a pseudo-code notation.In some
cases the system response (i^) is given.
Microdata
list name, age, sex
from labourcensusemployees
where industry = 'whole industry' and year = 1980
Macrodata
list number (employees), average (employees.income)
from labourcensus
where industry - 'whole industry' and year - 1980
cross-classified by age^group and sex
Metadata
household All the people belong to a household who live there together and have a joint budget
Each person who has an own
budget forms her own household,
summary-attribute (employees)
income categoryattribute (employees)
domain (industry)
3.3 Features and Requirements of a Statistical Database
The data structures offered by conventional database management systems
are rather inconvenient for storing and retrieving statistical data. Moreover,
the functionality of the systems is inappropriate and inadequate for
interpreting, validating and analyzing such kind of data. The main reasons
are the followings:
• Set- (tupel- or row-)oriented as well as ordered structures must
berepresented.
• Vectors, matrices, nested arrays besides non-fixed formatted data
types like documents, images, and plots must be stored.
• Almost static (historical) micro- and macrodata with very slow
update rates exist.
• GBytes of data are to be stored some of which are sparse.
• Very complex semantic integrity constraints are defined on the microand macrodata, espescially in order to map their links.
• Micro- and macrodata without the corresponding metadata are useless.
• The locality (^clusterings) of data according to attributes (columns) or
records (rows) is context-dependent.
• The querying and the processing of statistical data can cause »long
transactions«.
• The creation of macrodata by grouping, aggregation etc. implies the
bookkeeping of metadata.
• The frontier (interface) between retrieval and statistical analysis is not
crisp.
• The retrieval of even »anonymous« data must care about privacy
rules.
The above special features of a statistical database (STDB) give raise to the
following requirements for the design of a statistical database:
• The conceptual data model must include the micro-, macro- and
metadata level.
• The data-structures for micro-, macro- and metadata must be efficient.
• The set of operators should be complete which is defined on the
micro-, macro- and meta data level.
• The user-interfaces for the different user-groups must be user-friendly.
• Privacy handling mechanism must be incorporated.
3.4
Statistical Data Modelling
In this section, we propose to use the general location model to model
database and use model learnt to generate synthetic database. We will
examine in detail how to extract statistics and rules to estimate parameters
of the general location model and how to resolve the potential disclosure of
confidential information in data generation using model learnt.
The General Location Model
Let A1;A2; _ _ _ ;Ap denote a set of categorical attributes and Z1;Z2; _ _ _
;Zq a set of numerical ones in a table with n entries. Suppose Aj takes
possible domain values 1, 2, _ _ _ ; dj , the categorical data W can be
summarized by a contingency table with total number of cells equal to
D = πpj =1 dj . Let x = xd : d = 1; 2; _ _ _ ;D denote the number of entries
in each cell. Clearly ∑Dd =1 xd = n.
Database Modelling through the General Location Model
Our approach is to derive an approximate statistical model from the
characteristics (e.g., constraints, statistics, rules, and data summary) of the
real databases and generate a synthetic data set using model learned. A
major advantage of our system over [1] is that, in addition to constraints, we
extract more complex characteristics (e.g., statistics and rules) from data
catalogue and data and use them to build statistical model. As we discussed
in introduction, even if one synthetic database satisfies all constraints, it
does not mean it can fulfil users’ testing requirement as it may have
different data distribution than production database.
Our intuition is that, for database applications, if two databases are
approximately the same from a statistical viewpoint, then the performance
of the application on the two databases should also be approximately the
same 1. Furthermore, our system includes one disclosure analysis
component which helps users remove those characteristics which may be
used by attackers to derive confidential information in production database.
As all information used to generate synthetic data in our system are
contained in characteristics extracted, our disclosure analysis component
can analyze and preclude the potential disclosure of confidential
information at the characteristics (i.e., statistics and rules) level instead of
data level.
Model Learning
The characteristics of production databases can be extracted from three
parts: DDL, Data Dictionary, and Data. In order to ensure that the data is
close looking or statistically similar to real data, or at least from the point of
view of application testing, we need to have the statistical descriptions, S,
and non-deterministic rules, NR, of real data in production databases. These
two sets describe the statistical distributions or patterns of underlying data
and may affect the size of relations derived as a result of the evaluations of
queries the application will need to execute. Hence, they are imperative
for the statistical nature of the data that determines the query performance
of database application.
Extracting characteristics from data dictionary Data dictionary consists
of read only base tables that store information about the database. When
users execute an SQL command (e.g., CREATE TABLE, CREATE
INDEX, or CREATE SEQUENCE) to create an object, all of the
information about column names, column size, default values, constraints,
index names, sequence starting values, and other information are stored in
the form of metadata to the data dictionary. Most commercial DBMSs also
collect statistical information regarding the distribution of values in a
column to be created. This statistical information can be used by the query
processor to determine the optimal strategy for evaluating a query. As the
data in a column changes, index and column statistics can become out-ofdated and cause the query optimizer to make less-than-optimal decisions on
how to process a query. SQL server automatically updates this statistical
information periodically as the data in the tables changes. In our system, we
have one component which simply accesses tables in data dictionary and
fetch characteristics related.
Extracting characteristics from data The statistics information about
columns extracted directly from data dictionary is usually with high
granularity which may be insufficient to derive relatively accurate model. In
practice it is usually true that userscan collect more statistics at low
granularity from original data or a sample of real data themselves.
SELECT
Zip, Race, Age, Gender, COUNT(*),
AVG(Balance), AVG(Income), AVG(InterestPaid),
VAR POP(Balance), VAR POP(Income), VAR POP(InterestPaid),
COVAR POP(Balance,Income), COVAR POP(Balance,InterestPaid),
COVAR POP(Income,InterestPaid)
FROM
Mortgage
GROUP BY Zip, Race, Age, Gender
HAVING COUNT(*) > 5
Example 1: extracting statistics using SQL
Example 1above presents one SQL command to extract statistics at the
finest granularity level. It returns all information needed to derive
parameters of general location model. For example, the value from
aggregate function COUNT(*) is the estimate, xd, of the number of tuples in
each cell while the values from aggregate functions (i.e., AVG, VAR POP,
COVAR POP) are the estimates of mean vectors and covariance matrices
respectively of the multi-variate normal distribution of each cell. It is worth
pointing out all aggregate functions can be used with GROUP BY clause
with CUBE or ROLLUP option if we want to extract statistics at all
possible granularities. In practice, it may be infeasible to extract statistics at
the finest level because statistics at the finest level may contain too much
information and may be exploited by attackers to derive some confidential
information about production databases.
As our goal is to generate synthetic data for database application testing, we
may extract statistics which are only related to queries in workload of
database software. For the workload which contains two queries shown in
Figure 2, it is clear that the distribution of underlying data at some high
level (instead of at the finest level) is sufficient to capture the relation with
the execution time of queries in workload. For example, the approximate
distribution on Zip;Race would satisfy the performance requirements of Q1
in workload. In this case, we may only extract statistics necessary for query
performance of queries. Example 3 presents two SQL commands to extract
statistics for two queries shown in Example 2.
Q1: SELECT AVG(Income), AVG(InterestPaid) FROM Mortgage WHERE Zip = z AND Race = r
Q2: SELECT AVG(Income) FROM Mortgage WHERE Age = a AND Zip = z
Example 2. Workload example of Mortgage database
Statistics 1: SELECT
Zip, Race, COUNT(*), AVG(Balance), AVG(InterestPaid)
FROM
Mortgage
GROUP BY Zip, Race
HAVING COUNT(*) > 5
Statistics 2: SELECT
Zip, Age, COUNT(*), AVG(Income)
FROM
Mortgage
GROUP BY Zip, Age
HAVING
COUNT(*) > 5
Example 3. SQLs of extracting statistics for workload
Extracting characteristics from rule sets To derive the deterministic rule
set R, we take advantage of the database schema, which describes the
domains, the relations, and the constraints the database designer has
explicitly specified. Some information (function dependencies, correlations,
hierarchies etc.) can be derived from database integrity
constraints such as foreign keys, check conditions, assertions, and triggers.
Furthermore, users may apply some data mining tools to extract nondeterministic rules NR from production database. The non-deterministic
rule set NR helps describe the statistical distributions or patterns of
underlying data and may affect the size of relations derived as a result of the
evaluations of queries the application will need to execute. Formally, each
rule in R and NR can be represented as a declarative rule and is generally of
the form:
IF <premise> THEN <conclusion> [with support s and confidence c]
The rules may include exact, strong, and probabilistic rules based on the
support and confidence. We note here that complex predicates and external
function references may be contained in both the condition and action parts
of the rule. Anyone with subject matter expertise will be able to understand
the business logic of the data and can develop the appropriate conditions
and actions, which will then form the rule set.
Rule 1: IF Zip = 28223, Race = Asian, and Age in (25, 40) THEN Balance is in
(20k,30k) with support s = 900 and confidence c= 90 %.
Rule 2: IF Zip = 28262 THEN Race = White with support s = 5000 and
confidence c = 80 %
Example 4: The non-deterministic rules for Mortgage dataset
The above example 4 shows two non-deterministic rules for Mortgage
database. We can interpret Rule 1 as there are 1000 customers with Zip =
28223, Race = Asian, and Age in (25, 40) and 90 % of them with Balance
in the range of (20k, 30k). It is straightforward to see these rules can be
mapped to statistics of general location model at some granularity. For
example, the number of data entries in cell (28223, Asian, 25- 40, All) is
1000 and we can derive average balance of data entries in this cell from the
clause Balance in (20k,30k) with confidence c= 90 %.
Balance >1000 All All All
All
>1000
Statistics 1 28223 Asian All All 2800 23000
75000
28223 Black All All 3100 35000
89000
.
.
. .
.
.
.
.
28262 White All All 2500 123000
112000
Statistics 2 28223 All
20 All 300
56000
28223 All
21 All 570
38000
.
.
.
.
.
.
.
.
.
28262 All
40 All 210
73000
Rule 1
28223 Asian 25-40 All 900 (20k,30k)
Rule 2
28262 All All All 5000
28262 White All All 4000
Table 2:
The table of rule sets
Fitting model using characteristics It is easy to see all characteristics (i.e.,
S, R, NR) extracted from production database can be mapped to constraints
of parameters of general location model at the finest level. Table 2 shows
parameter constraints derived from examples (e.g., constraint Balance >
1000, two statistics from Example 3, and two rules from Example 4.) we
discussed previously.
Given those constraints, we can apply linear programming techniques
to derive parameters of general location model at the finest level. However,
it is infeasible to apply linear programming techniques directly in practice
due to high complexity (the number of variables is linear of the number of
cells D while the number of constraints is large). As we know, the problem
of estimating the cell entries at the finest granularity subject to some linear
constraints is known to be NP-hard. In our system, we combine some
heuristics to derive parameters from high level constraints. For example,
from Rule 1, we can initialize the number of tuples in those 30 cells (28223,
Asian, Age, Gender) where Age is in (25,40) and Gender in (Male, Female)
as 30 ( 900/15x2 ) when we assume the tuples are uniformly distributed
among Age and Gender in cell (28223, Asian, Age, Gender).
3.5
Some Statistical Data Models
According to Ullman (1988) a data model is a notation for describing data
and a set of operations used to manipulate the data. Many proposals for
such models have been published where most of them use a graph-theoretic
approach. The most prominent models are presented in the next subchapters. A common feature of these models is the fact that most of them
deal only with macro- and metadata. The macrodata are linked to a specific
topic (context) or a field of interest Several statistical populations are part of
such a field of interest. A specific population is considered as a set or class
of units (instances) each of which are characterized by a set of properties
(attributes) which are related to each other. The attributes can be classified
(Shoshani, 1982) according to their role. Attributes used to categorize or to
index data are called category attributes and the attributes expressing
summary properties are called summary attributes. The value of a category
attribute is sometimes called »category« instead of »category value«.
3.5.1
SUBJECT
Chan, Shoshani (1981) consider a single multi-dimensional table which is
modelled as a root tree (V, E) where the set V of nodes includes C-nodes
(cluster-nodes): a cluster is a set of subordinate categories, e.g. the node
»Year« has subordinates 1980, 1985, 1990. X-nodes (cross-product nodes):
a cross product is the Cartesian product of sets of categories, e.g. domain
(Sex) x domain (Year); note that the root node is of this type and refers to a
complete table.
Figure 24: A Subject Tree
The left subtree represents the stub of a statistical table and is modelled by
Cand X-nodes.The right subtree models the heading of the table and
consists of C- and X-nodes. The different summary attributes are clustered
in a C-node, indicating that the subordinate nodes are pointing to data
columns in the multidimensional table. Besides of being marked with either
a »C« or a »X« , the nodes are labeled according to the role they play in a
context, i.e. the name of the table, the names of category attributes, the
corresponding subordinate categories and the names of the summary
variables are given.
The set E of edges reflects the linking of nodes of type (X, X), (X, C), (C,
C), and (C,X). There are some special features of this data model:
- The tree is representing more the physical structure of a statistical table
(i.e. stab and heading of a table are modelled as subtrees).
- The tree is asymmetric in its subtrees. The right subtree models the
heading of the table and refers columnwise to the data.
The following set of SUBJECT operations are available:
- Browsing is achieved by traversing the graph
- Search is provided to locate the file nodes directly using specific
keywords
- Examine locating of nodes that contain specific keywords
- Include allowing for the specification of predicate conditions for queries
- Aggregation aggregating a selected set of terminal nodes
- Display displaying the result of a query in a table form
- Document displaying the text document associated with a node
3.5.2
NF2-table structures
Ozsoyoglu and Yuan (1987) proposed a non-first-normal form (NF2) of a
nested relationship type. This form is called a multi-dimensional or multiway table in statistics, cf Nelder (1974). It is a more natural representation
of a complex data-structure because it is not »flattened out« like a
normalized relationship. It is characterized by a crossing and nesting of the
categorical attributes which together make up the stub and heading of the
table. Each cell contains the corresponding value of the summary attribute
linked to the relationship type.
For example, consider the table »Professional Position in California«
published in Ferri, Pisano, Rafanelli (1992) and printed in a slightly
modified form below. It shows the absolute frequency distribution on the
category attributes sex, year, professional category and qualification. The
attributes sex and year are nested while sex and professional category are
crossed.
3.5.3 STORM (Statistical Object Representation Model)
Rafanelli, Shoshani(1990) proposed a graph-oriented data model which is
an enhanced version of the SUBJECT model.
Table 3
A complex data structure is used which describes a statistical object(StO). It is
defined by the quadruple
StO = (N, Ca, S, f)
where
N is the name of the statistical object
Ca is a (finite) set of category attributes
S is a (single) summary
summary attribute
f links Ca and S by a root tree.
A STORM model is a directed, acyclic graph(DAG) consisting of several
topic nodes (T-nodes). A T-node is either a root node, or a node with at
least one proceeding T-node and/or at least one succeeding S-node. A Snode is the root node of subtree corresponding to a single StO.
Allowing for:
T-nodes .... topic nodes
S-nodes summary attribute nodes
X-nodes cross product nodes
C-nodes cluster nodes
a StO is represented by a STORM tree and a STORM model by a DAG.
A specific feature of STORM is to model non-balanced (»non-symmetric«)
and heterogeneous statistical objects, cf. Rafanelli (1991). Non-symmetry
arises when one category attribute is classified in a classification hierarchy
(nomenclature), in which the number of levels is different depending on the
category attribute referred to. For example, state, county and city form a
hierarchy. Evidently, there exist states having cities but missing counties.
Non-mogeneity arises when the instances of a category attribute are, in turn,
classified with regard to different criteria. For example, the category
attribute'Professional Category ' has categories 'Engineer', 'Secretary' and
'Teacher'. While 'Teacher' may have subordinate categories like elementary,
grammar or high school teacher, 'Engineer' may be sub-classified according
to the diploma degree etc.
3.5.4
CSM (Conceptual Statistical Model)
The data models presented so far were concerned with macro- and
metadata. Di Battista and Batini (1988) introduced a model which covers all
kind of data, however, using different paradigms.
1. Microdata or elementary data are represented by an ER-model
2. Macrodata or summary data are represented by a graph-based model
similar to the SUBJECT data-model.
3. Metadata are embedded in the ER-model and in the nodes of model graph
representing the macrodata.
The conceptual schema of the macrodata is modelled by a labeled and
marked DAG. Its nodes have the following semantic (Tab. 4). The
corresponding proauction rules used for designing a feasible DAG are of
the type (if a node is oftype A then it has as parents nodes of type C and/or
A)
Table 4:
3.5.5
SDM4S (Statistical Data Model based on a 4
Schema Concept)
This data model is devoted to macro- and metadata. The conceptual
approach is strictly object-oriented. It has been developed and improved
over a couple of years. Similar ideas have been developed in order to
improve the semantics of data modelling. The concepts of an object graph
with formal definitions in an (infologica) language called INFOL, of a
metaobject graph with type, series and occurrence layers and of the socalled alfa-beta-gamma-tau-analysis have been introduced, which represent
metadata from a substantial, regional and temporal point of view. The main
features of this model are the following:
The model considers only macro- and metadata.
- The macrodata are stored physically in a relational database.
- The metadata are embedded in a statistical data dictionary (StDD). Its
logical structure is described by an object-oriented (frame-oriented) system
with 4 levels.
There are three classes of objects forming the data-dictionary frames.
- Statistical object
- Category value (domain)
- Summary value (domain).
The hierarchical structure is mapped by arcs of the type
- a_kind_of representing a superclass-subclass relationship,
- is_a representing a class-instance relationship,
- a_part_of representing a whole-part relationship.
Note, that the definition of the »is-a« relationship is rather unusual. The
main manipulating facilities are:
• An editor for insert, update and delete operations.
• A browser uses the links between frames to explore the StDD and to
list its items.
The statistical data-dictionary of the SDM4S is represented by a frame or
object-oriented system. It consists of 3 types of classes (frames) and has 4
levels of abstraction.
(1) The data model level (root level of the frames):
There exist 3 root levels :
- Statistical objects categorized by
• the categorical attributes with category values
• and the summary attributes with summary values
- Category Values (Domain)
- Summary Values (Domain)
(2) The Conceptual Level:
On this level the data is described which is conceptually obtainable in the
realm (the real object world) of a database regardless of its availability.
Ex.:
Persons are categorized by sex, age and summarized by the population size, i.e. by counting the
number of persons classified according to their sex and age group.
Persons, Employees represent statistical objects on the conceptual level Age and Sex Category,
Population Size represent conceptual domains
Example 5: The Conceptual Level
(3) DB Schema Level:
On this level one describes which part of the conceptual data is available as
actual and stored data in the SDB. There may be gaps due to sparse data.
The distinction between conceptually obtainable and actual available data
makes it possible to give second best answers to a query. For example, a
response could be »Census data on persons not available on a annual basis
but on a 10 years basis!«.
Ex.:
Persons for Census are categorized by
sex = male/female
age = 5 years age group and summarized with
population = number with unit = 1000.
Persons for Census represent an actual statistical object
5 years age group represent an actual domain
Example 6: DB schema Level
4) Instance Level
On this level the individual metadata and individual values are described which are
linked to the objects on the instance level.
Ex.:
The metadata corresponding to a cell of a multi-dimensional table are kept here
together with the name
of the statistical object and the values of the categorical and summary attributes.
Example 7: Instance Level
3.5.6
Modelling metadata using an eER-diagram
A comprehensive formal way to model metadata is to use an extended
entity relationship model (eERD). An eER diagram is a graph which
consists of the nodes of type (rectangular) representing entity types or
classes of meta-objects which have similiar characteristics.
The nodes of type (rhombs) representing types of sub- or supordinating as
well as a temporal preordering the arcs reflecting structural and semantic
relationships between the metadata. The arcs are marked with the maximum
complexity number.In order to simplify the notation the eER diagram is
drawn without any characteristics, i.e. attributes of the entity types of the
metadata. Of course, adetailed structure can be visualized by a stepwise
refinement of each subset of entity types or classes. This is demonstrated
below (Fig. 4).
The ER-diagram shown below (Fig. 5) is visualizing die attributes of the
(Meta) entity types (attribute), (statistical object) and (file). It should be
considered as an example for a further refinement of an eER-diagram. it
reflectsthe semantic, statistical and storage views on the metadata.
The operations needed to create, update and explore the metadata are:
- loading, editing a metadata graph
- scrolling
- browsing
- searching
- zooming
- listing, printing, storing.
Figure 25
Fig. 26
4.0
Conclusion
With the introduction to basic concepts Statistics database system, students
must have acquired the basic sense of the central importance of statistical
databases in today’s information systems environment. And with this, they
should be able to design and model various Statistical database for real life
situations.
5.0
Summary
In this unit we have learnt that:
Basic features and different definitions of Statistical database system
and the operations for manipulating these structures, as well as certain
constraints that the database should obey.
Various concepts of Statistical database policies and models were
extensively discussed.
We also design Statistics database models for use in defining the
target state and the subsequent planning needed to hit the target state..
6.0
Tutor Marked Assignment
1. (a) Define Statistical database system
(b) Design a model to solve one real life problem using a
Statistical database approach.
2. (a) Explain the basic concepts in Statistical database system.
7.0
Further Reading and Other Resources
Duncan et. al., 1993, Private Lives and Public Policies,. p. 157
“Restricted Access Procedures” by the Confidentiality and Data Access
Committee (April
2002) at http://www.fcsm.gov/committees/cdac/cdacra9.doc.
http://www.fcsm.gov/committees/cdac/resources.html
Di Battista, G. and Batini, C.(1988), Design of Statistical Databases: A
Methodology for the Conceptual Step, Inform. Systems, Vol.13, No. 4, pp.
407-422.
Van den Berg, G.M., de Feber, E., and de Greef, P. (1992), Analysing
StatisticalData Processing, in Eurostat, New Technologies and Techniques
for Statistics,Luxembourg, pp 102-111.
Chan, P., and Shoshani, A. (1981), SUBJECT: A Directory Driven System
for Large Statistical Databases, in Proc. of the LBL Workshop on Statistical
Database Management, Lawrence Berkeley Lab, Berkeley, CA.
Cubitt, R. et al.(eds.) (1986), Proceedings of the Third International
Workshop on Statistical and Scientific Database Management, Eurostat,
Luxembourg. Eurostat (1992), New Technologies and Techniques for
Statistics, Luxembourg.
Ferri, F., Pisano, M.T., and Rafanelli, M.(1992), A Object Oriented Visual
Definition Language for Statistical Data, in New Technologies and
Techniquesfor Statistics, Proc. of the conference, Bonn, pp.320-339.
Hinterberger, H., French, J.C.(eds.) (1992), Proceedings of the Sixth
InternationalWorkshop on Statistical and Scientific Database Management,
Departmentof Informatik, ETH Zürich, Zürich.
Lenz, H.-J.(1993), M3-Database Design, Micro-, Macro- and Metadata
Modelling, in F. Faulbaum (ed.), SoftStat '93 Advances in Statistical
Software
D. Chays, S. Dan, P. Frankl, F. Vokolos, and E. Weyuker. A framework for
testing database applications. In Proceedings of the ISSTA. Portland,
Oregon, 2000.
A. Dobra and S. E. Fienberg. Bounds for cell entries in contingency tables
induced by fixed marginal totals with applications to disclosure limitation.
Statistical Journal of the United Nations ECE, 18:363–371, 2001.
J.L. Schafer. Analysis of Incomplete Multivariate Data. Chapman Hall,
1997.
Module 2 The Statistical Database System
Unit 2 Statistical Data Analysis, Mining and Decision Tree
1.0
Introduction
2.0
Objective
3.0
Statistical analysis of Oracle performance data using R
3.1
Data Mining and the Semantic Conference Organizer
3.2
Data Mining and Decision Tree
4.0
Conclusion
5.0
Summary
6.0
Tutor Marked Assignment
7.0
Further Reading and Other Resources
1.0
Introduction
A Statistical data analysis of Database can be carried out using Basic RStatistical function with little laborious configuration. This analysis can be
made possible with the use of Statistical tables and graphs for statistic
inferences. In the same manner, the organization of a technical meeting,
workshop, or conference involving submitted abstracts or full-text
documents can be quite an onerous task. To gain a sense of what topic each
submission addresses may require more than just a quick glimpse at the title
or abstract. The use of automated indexing and text mining revolutionises
the manner and speed of information assessment and organization.
2.0
Objective
At the end of this unit, you should be able to:
Appreciate the features of Statistical analysis in Oracle performance
data using R database management system (RDBMS)
Use High-Level Conceptual Data mining and the Semantic conference
organizer Database system.
State the major concepts of Data mining and Decision Tree.
Distinguishes between the major components of Data mining and
Semantic conferencing.
3.0
Statistical analysis of Oracle performance
data using R
R is without doubt the Open Source tool of choice for statistical analysis, it
contains a huge variety of statistical analysis techniques – rivalled only by
hugely expensive commercial products such as SAS and SPSS.
Installing R:
R can be install in linux as a standard package: On Windows, one may
wish to use the Revolution R binaries. At times, there may trouble
installing the 32-bit binaries on the system as they conflicted with the 64-bit
JDBC. The easiest way to setup a connection to Oracle to install the RJDBC
package is hereby stated:
[[email protected] ~]$ R
R version 2.12.1 (2010-12-16)
Copyright (C) 2010 The R Foundation for Statistical Computing
ISBN 3-900051-07-0
Platform: x86_64-redhat-linux-gnu (64-bit)
<snip>
Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.
> install.packages("RJDBC")
Getting data from Oracle into R
Once R is installed, it’s pretty simple to get data out of Oracle and into R.
Here’s a very short snippet that grabs data from the V$SQL table:
1: library(RJDBC)
2:
3: drv <- JDBC("oracle.jdbc.driver.OracleDriver",
4:
"/ora11/home/jdbc/lib/ojdbc6.jar")
5:
6: conn <- dbConnect(drv,"jdbc:oracle:thin:@hostname:1521: service","username","password")
7: sqldata<-dbGetQuery(conn, "SELECT cpu_time cpu,elapsed_time ela,disk_reads phys,
8:
buffer_gets bg,sorts sorts
9:
FROM V$SQL ")
10: summary(sqldata)
Let’s look at that line by line:
Line
Comments
1
The library command loads the RJDBC module, which will provide connectivity to
Oracle.
3
We create a driver object for the Oracle JDBC driver. The second argument is the
location of the Oracle JDBC jar file, almost always
$ORACLE_HOME/jdbc/lib/ojdbc6.jar.
6
Connect to the Oracle database using standard JDBC connections strings
7
Create an R dataset from the result set of a query. In this case, we are loading the
contents of the V$SQL table.
10
The R “summary” package provides simple descriptive statistics for each variable in the
provided dataset.
Basic R statistical functions
R has hundreds of statistical functions, in the above example we used
“summary”, which prints descriptive statistics. The output is shown below;
mean, medians, percentiles, etc:
Figure 27.
Correlation
Statistical correlation reveals the association between two numeric
variables. If two variables always increase or decrease together the
correlation is 1; if two variables are absolutely random with respect of each
other then the correlation tends towards 0.
correlation prints the correlation between every variable in the data set:
Figure 28.
Correlation test calculates the correlation coefficient and prints out the
statistical significance of the correlation, which allows you to determine if
there is a significant relationship between the two variables. So does the
number of sorts affect response time? Let’s find out:
Figure 29
The p-value is 0.19 which indicates no significant relationship – p values
of no more than 0.05 (one chance in 20) are usually requires before we
assume statistical significance.
On the other hand, there is a strong relationship between CPU time and Elapsed time:
Figure 30
Plotting
plot prints a scattergram chart. Here’s the output from plot(sqldata$ELA,sqldata$CPU):
Figure 31
Regression
Regression is used to draw “lines of best fit” between variables. In the
simplest case, we use the “lm” package to create a linear regression model
between two variables (which we call “regdata” in the example). The
summary function prints a summary of the analysis:
Figure 32
This might seem a little mysterious if your statistics is a bit rusty, but the
data above tells us that there is a significant relationship between elapsed
time (ELA) and physical reads (PHYS) and gives us the gradient and Y axis
intercept if we wanted to draw the relationship. We can get R to draw a
graph, and plot the original data by using the plot at abline functions:
Figure 33
Testing a hypothesis
One of the benefits of statistical analysis is you can test hypotheses about
your data. For instance, what about we test the until recently widely held
notion that the buffer cache hit rate is a good measure of performance. We
might suppose if that were true that SQL statements with high buffer cache
hit rates would show smaller elapsed times than those with low buffer cache
hit rates. To be sure, there are certain hidden assumptions underlying that
hypothesis, but for the sake of illustration let’s use R to see if our data
supports the hypothesis.
Simple correlation is a fair test for this; all we need to do is see if there is a
statistically significant correlation between hit rate and elapsed time.
Here’s the analysis:
Figure 34
The correlation is close to 0, and the statistical significance way higher
than the widely accepted .05 threshold for statistical significance.
Statements with high hit ratios do not show statistically significantly lower
elapsed times that SQLs with low hit ratios.
3.1
Statistical Data Mining and the Semantic
Conference Organizer
The organization of a technical meeting, workshop, or conference involving
submitted abstracts or full-text documents can be quite an onerous task. To
gain a sense of what topic each submission addresses may require more
than just a quick glimpse at the title or abstract. The use of automated
indexing and text mining can revolutionize the manner and speed of
information assessment and organization. In this section, the use of Latent
Semantic Indexing (LSI) for probing and labelling conference abstracts
using an intuitive Web interface and client-server internal software design
using grid-based middleware such as NetSolve, is demonstrated.
3.1.1
Background
Creating a conference manually can be a burdensome task. After all papers
have been submitted, the human organizer must then group the papers into
sessions. The session topics can be decided either before or after the
organizer has a feel for the material covered in the papers. And since the
average conference has around one hundred papers submitted to it, the
organizer must shuffle these papers between topics trying to find a
workable fit for the papers and the sessions to which they are assigned. Of
course, one person trying to fit fifty to one hundred papers into about
twenty sessions will lose context very quickly. Switching rapidly between
sessions will cause confusion, and renaming sessions or assigning different
topics may cause the entire conference to get reworked. Many times the
human organizer will only work with document surrogates such as an
abstract or simply the paper title, so often papers will be misclassified due
to summarization errors.
Also note that a significant amount of time must be spent reading and rereading abstracts to remember what each paper’s subject is. Manually
creating a conference takes anywhere from a day to a week or longer. With
such a combinatorial problem confronting the person who manually
organizes the conference, the need for some sort of automated assistance is
justified in hopes of reducing the hours spent in creating a conference.
3.1.2
Latent Semantic Indexing
In order for the Semantic Conference Organizer to be useful, it must replace
the most time-consuming of tasks undertaken when creating a conference—
reading. There are several techniques and algorithms used in the field of
information retrieval that enable relevant documents to be retrieved to meet
a specific need without requiring the user to read each document. The
model used by the Semantic Conference Organizer is latent semantic
indexing or LSI.
Once the document collection is received, it must be parsed into bare words
called tokens. All punctuation and capitalization is ignored. In addition,
articles and other common, non-distinguishing words are discarded. In
effect, each document is viewed as a bag of words upon which operations
can be performed. Once the bag of words has been formed, a term-bydocument matrix is created where the entries of the matrix are the weighted
frequencies associated with the corresponding term in the appropriate
document.
The weight of a term within a document is a nonnegative value used to
describe the correlation between that term and the corresponding document.
A weight of zero indicates no correlation. In general, each weight is the
product of a local and global component. A simplistic method of obtaining
weights is to assign the local component as the frequency of the word
within the document and the global component as the log of the proportion
of total documents to the number of documents in which the term appears.
Such a method is known as a tf-idf (term frequency, inverse-document
frequency) weighting scheme. The aim of any scheme is to measure
similarity within a document while at the same time measuring the
dissimilarity of a document from the other documents within the collection.
FIGURE 35. Sample layout of the Semantic Conference Organizer.
3.2
Data Mining and Decision Tree
since Data Mining is all about automating the process of searching for
patterns in the data. It is necessary to known which patterns are interesting,
which might be mere illusions and how can they be exploited? And the
answer will turn out to be the engine that drives decision tree learning.
3.2.1
Learning Decision Trees
A Decision Tree is a tree-structured plan of a set of attributes to test in order
to predict the output, and to decide which attribute should be tested first,
simply find the one with the highest information gain.
Table 5: Table of Dataset
Figure 36
Figure 37
Summary of Basic Decision Tree Building
BuildTree(DataSet,Output)
• If all output values are the same in DataSet, return a leaf node that says
“predict this unique output”
• If all input values are the same, return a leaf node that says “predict the
majority output”
• Else find attribute X with highest Info Gain
• Suppose X has nX distinct values (i.e. X has arity nX).
• Create and return a non-leaf node with nX children.
• The i’th child should be built by calling BuildTree(DSi,Output)
Where DSi built consists of all those records in DataSet for which X = ith
distinct value of X.
4.0
Conclusion
There’s tons of data in our Oracle databases that could benefit from
statistical analysis – not the least the performance data in the dynamic
performance views, ASH and AWR. We use statistical tests in Spotlight on
Oracle to extrapolate performance into the future and to set some of the
alarm thresholds. Using R, you have easy access to the most sophisticated
statistical analysis techniques and as I hope I’ve shown, you can easily
integrate R with Oracle data.
5.0
Summary
In this unit we have learnt that:
How to use Statistical analysis tools in database system and collection
of related information (data) in a structured way using ‘R’ method
Statistical Database analysis is a tool of program that manages the
database structure and that control shared access to the data in the
database.
The advantages include controlling redundancy, restricting
unauthorized access, saving time etc, with flexibility, economies of
scale, and potential for enforcing standards, are some of the
implications of Statistics database approach.
6.0
1.
Tutor Marked Assignment
(a) Differentiate between the following:
(i) Data mining and information (ii) Semantic conferencing and
database management.
(b) Demonstrate a real life situation where you can use statistical
analysis to solve human’s problem.
2. (a) What are the characteristics of Semantic conferencing.
(b) Mention the advantages of using Decision Tree management.
7.0
Further Reading and Other Resources 2010
Annual Subscription to STAN: OECD Structural Analysis Statistics Online
Database on the OECD iLibrary:
Wiebren de Jonge, Compromising statistical databases responding to
queries about means, ACM Transactions on Database Systems, Volume 8,
Issue 1 (March 1983), Pages: 60 – 80
S. Deerwester, S. Dumais, G. Furnas, T. Landauer, and R. Harshman.
Indexing by Latent Semantic Analysis. Journal of the American Society for
Information Science 41:391-407, 1990.
R. Baeza-Yates and B. Ribeiro-Neto. Modern Information Retrieval.
Addison-Wesley, Boston, MA, 1999.
M. Berry and M. Browne. Understanding Search Engines: Mathematical
Modeling and Text Retrieval. SIAM, Philadelphia, PA, 1999.
M. Berry, Z. Drmaˇc, and E. Jessup. Matrices, Vector Spaces, and
Information Retrieval. SIAM Review 41:335-362, 1999.
L. Breiman, J. H. Friedman, R. A. Olshen, and C. J. Stone. Classification
and Regression Trees. Wadsworth, Belmont, CA, 1984.
C4.5 : Programs for Machine Learning (Morgan Kaufmann Series in
Machine Learning) by J. Ross Quinlan
Learning Classification Trees, Wray Buntine, Statistics and Computation
(1992), Vol 2, pages 63-73
Kearns and Mansour, On the Boosting Ability of Top-Down Decision Tree
Learning Algorithms, STOC: ACM Symposium on Theory of Computing,
1996“
Module 2 The Statistical Database System
Unit 3 Computer Security and Statistical Databases
1.0
Introduction
2.0
Objective
3.0
Overview of Statistical Database
3.1
Inference from a Statistical Database
3.2
Query Restriction
3.3
Partitioning
3.4
Query Denial and Information Leakage
3.5
Perturbation
4.0
Conclusion
5.0
Summary
6.0
Tutor Marked Assignment
7.0
Further Reading and Other Resources
1.0
Introduction
In this unit, it is imperative for us to take a look at the unique security issues
that relate to statistical databases and know why the database administrator
must prevent, or at least detect, the statistical user who attempts to gain
individual information through one or a series of statistical queries.
The security problem of a statistical database is to limit the use of the
database so that no sequence of statistical queries is sufficient to deduce
confidential or private information.
Statistical databases often incorporate support for advanced statistical
analysis techniques, such as correlations, which go beyond SQL. They also
pose unique security concerns, which were the focus of much research,
particularly in the late 1970s and early to mid 1980s. In a statistical
database, it is often desired to allow query access only to aggregate data,
not individual records.
And securing such a database is a difficult problem, since intelligent users
can use a combination of aggregate queries to derive information about a
single individual. Some common approaches are:
•
only allowing aggregate queries (SUM, COUNT, AVG, STDEV, etc.)
•
rather than returning exact values for sensitive data like income, only
return which partition it belongs to (e.g. 35k-40k)
•
return imprecise counts (e.g. rather than 141 records met query, only
indicate 130-150 records met it.)
•
don't allow overly selective WHERE clauses
•
audit all users queries, so users using system incorrectly can be
investigated
•
use intelligent agents to detect automatically inappropriate system use
2.0
Objective
At the end of this unit, you should be able to:
Students should be able to use the features of Query Restriction,
Partitioning, Perturbation and information leakage of database
management system (DBMS)
Use High-Level Conceptual Query Denial and Information leakage.
State the major concepts Perturbation and Query restriction.
Draw the major components of Perturbation, Naming
Conventions and Design Issues in data management system
3.0
Overview of Statistical Database
A statistical database (SDB) is one that provides data of a statistical nature,
such as counts and averages. The term statistical database is used in two
contexts:
•
Pure statistical database: This type of database only stores statistical
data. An example is a census database. Typically, access control for a
pure SDB is straightforward: Certain users are authorized to access
the entire database.
•
Ordinary database with statistical access: This type of database
contains individual entries; this is the type of database discussed so far
in this chapter. The database supports a population of non-statistical
users who are allowed access to selected portions of the database
using DAC, RBAC, or MAC. In addition, the database supports a set
of statistical users who are only permitted statistical queries. For these
latter users, aggregate statistics based on the underlying raw data are
generated in response to a user query, or may be pre-calculated and
stored as part of the database.
It is essentially important to know that, design of a statistical database
should utilize a statistical security management facility to enforce the
security constraints at the conceptual model level. Information revealed to
users is well defined in the sense that it can at most be reduced to nondecomposable information involving a group of individuals. In addition, the
design also takes into consideration means of storing the query information
for auditing purposes, changes in the database, users' knowledge, and some
security measures.
For the purposes of this section, we are concerned only with the latter type
of database and, for convenience; refer to this as an SDB. The access
control objective for an SDB system is to provide users with the aggregate
information without compromising the confidentiality of any individual
entity represented in the database. The security problem is one of inference.
The database administrator must prevent, or at least detect, the statistical
user who attempts to gain individual information through one or a series of
statistical queries.
For this discussion, we use the abstract model of a relational database table
shown as Figure 5.7. There are N individuals, or entities, in the table and M
attributes. Each attribute Aj has |Aj| possible values, with xij denoting the
value of attribute j for entity i. Table 5.3, taken from an example that we use
in the next few paragraphs. The example is a database containing 13
confidential records of students in a university that has 50 departments.
Table 6: Ordinary Database with Statistical Access
Figure 38. Abstract Model of a Relational Database
Statistics are derived from a database by means of a characteristic
formula, C, which is a logical formula over the values of attributes. A
characteristic formula uses the operators OR, AND, and NOT (+, ·, ~),
written here in order of increasing priority. A characteristic formula
specifies a subset of the records in the database. For example, the formula
(Sex = Male) · ((Major = CS) + (Major = EE))
specifies all male students majoring in either CS or EE. For numerical
attributes, relational operators may be used. For example, (GP> 3.7)
specifies all students whose grade point average exceeds 3.7. For simplicity,
we omit attribute names when they are clear from context. Thus, the
preceding formula becomes Male · (CS + EE).
The query set of characteristic formula C, denoted as X(C), is the set of
records matching that characteristic. For example, for C = Female · CS,
X(C) consists of records 1 and 4, the records for Allen and Davis.
A statistical query is a query that produces a value calculated over a query
set. Table 5.4 lists some simple statistics that can be derived from a query
set. Examples: count(Female · CS) = 2; sum(Female · CS, SAT) = 1400.
Table 7: Some Queries of a Statistical Database
3.1
Inference from a Statistical Database
A statistical user of an underlying database of individual records is
restricted to obtaining only aggregate, or statistical, data from the database
and is prohibited access to individual records. The inference problem in this
context is that a user may infer confidential information about individual
entities represented in the SDB. Such an inference is called a compromise.
The compromise is positive if the user deduces the value of an attribute
associated with an individual entity and is negative if the user deduces that
a particular value of an attribute is not associated with an individual entity.
For example, the statistic sum (EE· Female, GP) = 2.5 compromises the
database if the user knows that Baker is the only female EE student.
In some cases, a sequence of queries may reveal information. For example,
suppose a questioner knows that Baker is a female EE student but does not
know if she is the only one. Consider the following sequence of two
queries:
count (EE · Female) = 1
sum (EE · Female, GP) = 2.5
This sequence reveals the sensitive information.
The preceding example shows how some knowledge of a single individual
in the database can be combined with queries to reveal protected
information. For a large database, there may be few or no opportunities to
single out a specific record that has a unique set of characteristics, such as
being the only female student in a department. Another angle of attack is
available to a user aware of an incremental change to the database. For
example, consider a personnel database in which the sum of salaries of
employees may be queried. Suppose a questioner knows the following
information:
•
Salary range for a new systems analyst with a BS degree is $[50K,
60K]
•
Salary range for a new systems analyst with a MS degree is $[60K,
70K]
Suppose two new systems analysts are added to the payroll and the change
in the sum of the salaries is $130K. Then the questioner knows that both
new employees have an MS degree.
In general terms, the inference problem for an SDB can be stated as
follows. A characteristic function C defines a subset of records (rows)
within the database. A query using C provides statistics on the selected
subset. If the subset is small enough, perhaps even a single record, the
questioner may be able to infer characteristics of a single individual or a
small group. Even for larger subsets, the nature or structure of the data may
be such that unauthorized information may be released.
3.2
Query Restriction
SDB implementers have developed two distinct approaches to protection of
an SDB from inference attacks (Figure 5.8):
Figure 39: Approaches to Statistical Database Security
•
Query restriction: Rejects a query that can lead to a compromise.
The answers provided are accurate.
•
Perturbation: Provides answers to all queries, but the answers are
approximate.
We examine query restriction in this section and perturbation in the next.
Query restriction techniques defend against inference by restricting
statistical queries so that they do not reveal user confidential information.
Restriction in this context simply means that some queries are denied.
Query Size Restriction
The simplest form of query restriction is query size restriction. For a
database of size N (number of rows, or records), a query q(C) is permitted
only if the number of records that match C satisfies
where k is a fixed integer greater than 1. Thus, the user may not access any
query set of less than k records. Note that the upper bound is also needed.
Designate All as the set of all records in the database. If q(C) is disallowed
because |X(C)| k, and there is no upper bound, then a user can compute q(C)
= q(All) − q(˜C). The upper bound of N - k guarantees that the user does not
have access to statistics on query sets of less than k records. In practice,
queries of the form q(All) are allowed, enabling users to easily access
statistics calculated on the entire database.
Query size restriction counters attacks based on very small query sets. For
example, suppose a user knows that a certain individual I satisfies a given
characteristic formula C (e.g., Allen is a female CS major). If the query
count(C) returns 1, then the user has uniquely identified I. Then the user
can test whether I has a particular characteristic D with the query count(C·
D). Similarly, the user can learn the value of a numerical attribute A for I
with the query sum(C, A).
Although query size restriction can prevent trivial attacks, it is vulnerable to
more sophisticated attacks, such as the use of a tracker. In essence, the
questioner divides his or her knowledge of an individual into parts, such
that queries can be made based on the parts without violating the query size
restriction. The combination of parts is called a tracker, because it can be
used to track down characteristics of an individual. We can describe a
tracker in general terms using the case from the preceding paragraph. The
formula C·D corresponds to zero or one record, so that the query
count(C·&) is not permitted. But suppose that the formula C can be
decomposed into two parts C = C1·C2, such that the query sets for both C1
and T = (C1·˜C2) satisfy the query size restriction. Figure 5.9 illustrates
this situation; in the figure, the size of the circle corresponds to the number
of records in the query set. If it is not known if I is uniquely identified by C,
the following formula can be used to determine if count(C) = 1:
count(C)=count(C1)−count(T)(5.2)
That is, you count the number of records in C1 and then subtract the
number of records that are in C1 but not in C2. The result is the number of
records that are in both C1 and C2, which is equal to the number of records
in C. By a similar reasoning, it can be shown that we can determine whether
I has attribute D with
count(C · D) = count(T +; C1 · D)−count(T)
For example, in Table 5.3, Evans is identified by C = Male·Bio·1979. Let k= 3 in
Equation 501. We can use T = (C1· ˜C2) = Male· ˜ (Bio· 1979). Both C1 and C2 satisfy
the query size restriction. Using Equations (5.2) and (5.3), we determine that Evans is
uniquely identified by C and whether his SAT score is at least 600:
Figure 40: Example of Tracker
In a large database, the use of just a few queries will typically be inadequate
to compromise the database. However, it can be shown that more
sophisticated tracker attacks may succeed even against large databases in
which the threshold k is set at a relatively high level [DENN79].
We have looked at query size restriction in some detail because it is easy to
grasp both the mechanism and its vulnerabilities. A number of other query
restriction approaches have been studied, all of which have their own
vulnerabilities. However, several of these techniques in combination do
reduce vulnerability.
Query Set Overlap Control
A query size restriction is defeated by issuing queries in which there is
considerable overlap in the query sets. For example, in one of the preceding
examples the query sets Male and Male· ˜ (Bio· 1979) overlap significantly,
allowing an inference. To counter this, the query set overlap control
provides the following limitation.
A query q(C) is permitted only if the number of records that match C
satisfies
for all q(D) that have been answered for this user, and where r is a fixed
integer greater than 0.
This technique has a number of problems, including the following
[ADAM89]:
1. This control mechanism is ineffective for preventing the cooperation
of several users to compromise the database.
2. Statistics for both a set and its subset (e.g., all patients and all patients
undergoing a given treatment) cannot be released, thus limiting the
usefulness of the database.
3. For each user, a user profile has to be kept up to date.
3.3
Partitioning
Partitioning can be viewed as taking query set overlap control to its logical
extreme, by not allowing overlapping queries at all. With partitioning, the
records in the database are clustered into a number of mutually exclusive
groups. The user may only query the statistical properties of each group as a
whole. That is, the user may not select a subset of a group. Thus, with
multiple queries, there must either be complete overlap (two different
queries of all the records in a group) or zero overlap (two queries from
different groups).
The rules for partitioning the database are as follows:
1. Each group G has g = |G| records, where g = 0 or g≥ n, and g even,
where n is a fixed integer parameter.
2. Records are added or deleted from G in pairs.
3. Query sets must include entire groups. A query set may be a single
group or multiple groups.
A group of a single record is forbidden, for obvious reasons. The insertion
or deletion of a single record enables a user to gain information about that
record by taking before and after statistics. As an example, the database of
Table 5.3a can be partitioned as shown in Table 5.5. Because the database
has an odd number of records, the record for Kline has been omitted. The
database is partitioned by year and sex, except that for 1978, it is necessary
to merge the Female and Male records to satisfy the design requirement.
Table 8 Partitioned Database
Partitioning solves some security problems but has some drawbacks. The
user’s ability to extract useful statistics is reduced, and there is a design
effort in constructing and maintaining the partitions.
3.4
Query Denial and Information Leakage
A general problem with query restriction techniques is that the denial of a
query may provide sufficient clues that an attacker can deduce underlying
information. This is generally described by saying that query denial can
leak information.
Here is a simple example from [KENT05]. Suppose that the underlying
database consists of real-valued entries and that a query is denied only if it
would enable the requestor to deduce a value. Now suppose the requester
poses the query sum(x1, x2, x3) and the response is 15. Then the requester
queries max(x1, x2, x3) and the query is denied. What can the requester
deduce from this? We know that the max(x1, x2, x3) cannot be less than 5
because then the sum would be less than 15. But if max(x1, x2, x3) > 5, the
query would not be denied because the answer would not reveal a specific
value. Therefore, it must be the case that max(x1, x2, x3) = 5, which
enables the requester to deduce that x1 = x2 = x3 = 5.
[KENT05] describes an approach to counter this threat, referred to as
simulatable auditing. The details of this approach are beyond the scope of
this chapter. In essence, the system monitors all of the queries from a given
source and decides on the basis of the queries so far posed whether to deny
a new query. The decision is based solely on the history of queries and
answers and the specific new query. In deciding whether to deny the query,
the system does not consider the actual values of database elements that will
contribute to generating the answer and therefore does not consider the
actual value of the answer.
Thus, the system makes the denial decision solely on the basis of
information that is already available to the requester (the history of prior
requests). Hence the decision to deny a query cannot leak any information.
For this approach, the system determines whether any collection of database
values might lead to information leakage and denies the query if leakage is
possible. In practice, a number of queries will be denied even if leakage is
not possible. In the example of the preceding paragraph, this strategy would
deny the max query whether or not the three underlying values were equal.
Thus, this approach is more conservative in that it issues more denials than
an approach that considers the actual values in the database.
3.5
Perturbation
Query restriction techniques can be costly and are difficult to implement in
such a way as to completely thwart inference attacks, especially if a user
has supplementary knowledge. For larger databases, a simpler and more
effective technique is to, in effect, add noise to the statistics generated from
the original data. This can be done in one of two ways (Figure 5.8): The
data in the SDB can be modified (perturbed) so as to produce statistics that
cannot be used to infer values for individual records; we refer to this as
data perturbation. Alternatively, when a statistical query is made, the
system can generate statistics that are modified from those that the original
database would provide, again thwarting attempts to gain knowledge of
individual records; this is referred to as output perturbation.
Regardless of the specific perturbation technique, the designer must attempt
to produce statistics that accurately reflect the underlying database. Because
of the perturbation, there will be differences between perturbed results and
ordinary results from the database. However, the goal is to minimize the
differences and to provide users with consistent results. As with query
restriction, there are a number of perturbation techniques. In this section, we
highlight a few of these.
Data Perturbation Techniques
We look at two techniques that consider the SDB to be a sample from a
given population that has a given population distribution. Two methods fit
into this category. The first transforms the database by substituting values
that conform to the same assumed underlying probability distribution. The
second method is, in effect, to generate statistics from the assumed
underlying probability distribution.
The first method is referred to as data swapping. In this method, attribute
values are exchanged (swapped) between records in sufficient quantity so
that nothing can be deduced from the disclosure of individual records. The
swapping is done in such a way that the accuracy of at least low-order
statistics is preserved. Table 5.6, from [DENN82], shows a simple example,
transforming the database D into the database D. The transformed database
D has the same statistics as D for statistics derived from one or two
attributes. However, three-attribute statistics are not preserved. For
example, count(Female· CS· 3.0) has the value 1 in D but the value 0 in D.
Table 9
Example of Data Swapping
Another method is to generate a modified database using the estimated
underlying probability distribution of attribute values. The following steps
are used:
1. For each confidential or sensitive attribute, determine the probability
distribution function that best matches the data and estimate the
parameters of the distribution function.
2. Generate a sample series of data from the estimated density function
for each sensitive attribute.
3. Substitute the generated data of the confidential attribute for the
original data in the same rank order. That is, the smallest value of the
new sample should replace the smallest value in the original data, and
so on.
Output Perturbation Techniques
A simple output perturbation technique is known as random-sample
query. This technique is suitable for large databases and is similar to a
technique employed by the U.S. Census Bureau. The technique works as
follows:
1. A user issues a query q(C) that is to return a statistical value. The
query set so defined is X(C).
2. The system replaces X(C) with a sampled query set, which is a
properly selected subset of X(C).
3. The system calculates the requested statistic on the sampled query set
and returns the value.
Other approaches to output perturbation involve calculating the statistic on
the requested query set and then adjusting the answer up or down by a given
amount in some systematic or randomized fashion. All of these techniques
are designed to thwart tracker attacks and other attacks that can be made
against query restriction techniques.
With all of the perturbation techniques, there is a potential loss of accuracy
as well as the potential for a systematic bias in the results.
Limitations of Perturbation Techniques
The main challenge in the use of perturbation techniques is to determine the
average size of the error to be used. If there is too little error, a user can
infer close approximations to protected values. If the error is, on average,
too great, the resulting statistics may be unusable. For a small database, it is
difficult to add sufficient perturbation to hide data without badly distorting
the results. Fortunately, as the size of the database grows, the effectiveness
of perturbation techniques increases.
The last-mentioned reference reported the following result. Assume the size
of the database, in terms of the number of data items or records, is n. If the
number of queries from a given source is linear to the size of the database
(i.e., on the order of n), then a substantial amount of noise must be added to
the system in terms of perturbation, to preserve confidentiality. Specifically,
suppose the perturbation is imposed on the system by adding a random
amount of perturbation ≤x . Then, if the query magnitude is linear, the
perturbation must be at least of order √n . This amount of noise may be
sufficient to make the database effectively unusable. However, if the
number of queries is sublinear (e.g., of order √n), then much less noise must
be added to the system to maintain privacy. For a large database, limiting
queries to a sublinear number may be reasonable.
4.0
Conclusion
As we all know that Computer security and Statistical database management
system is the life wire of any existing organization. With the introduction to
basic concepts Statistics database system and Computer security, students
must have acquired the basic sense of the central importance of statistical
databases in today’s information systems environment. And with this, they
should be able to design and model various Statistical database system to
safeguard crackers in real life situations.
5.0
Summary
In this unit we have learnt that:
A general problem with query restriction techniques is that the denial
of a query may provide sufficient clues that an attacker can deduce
underlying information.
The main challenge in the use of perturbation techniques is to
determine the average size of the error to be used. If there is too little
error, a user can infer close approximations to protected values.
Fortunately, as the size of the database grows the effectiveness of
perturbation techniques increases.
Partitioning can be viewed as taking query set overlap control to its
logical extreme, by not allowing overlapping queries at all.
A statistical user of an underlying database of individual records is
restricted to obtaining only aggregate, or statistical, data from the
database and is prohibited access to individual records
6.0
1.
Tutor Marked Assignment
(a) Differentiate between the following:
(i) Query Denial and Information leakage (ii) Statistical Database and
Query restriction management.
(b) What did you understand by the term ‘Partitioning’?
2. (a) What are the characteristics of Perturbation approach?
(b) Mention the advantages of using Perturbation technique over other
methods.
7.0
Further Reading and Other Resources
Dorothy E. Denning, Secure statistical databases with random sample
queries, ACM Transactions on Database Systems (TODS), Volume 5, Issue
3 (September 1980), Pages: 291 – 315.
Wiebren de Jonge, Compromising statistical databases responding to
queries about means, ACM Transactions on Database Systems, Volume 8,
Issue 1 (March 1983), Pages: 60 – 80.
Dorothy E. Denning, Jan Schlörer, A fast procedure for finding a tracker in
a statistical database, ACM Transactions on Database Systems, Volume 5,
Issue 1 (March 1980) . Pages: 88 – 102.
CCSP Self-Study: Advanced AAA Security for Cisco Router Networks By
John Roland
Unwitting Collaborators: Series Introduction By Frank Fiore, Jean Francois
Intrusion Detection Systems by Earl Carter
Module 3 Application of Statistical Database System
Unit 1 SPEA SMART Airport Statistical Data
Management System (SMART STAT)
1.0
Introduction
2.0
Objective
3.0
Uses of the Airport Statistical Data Management System
3.1
Benefits of the Airport Statistical Data Management
System
3.2
Outline of the Airport Statistical Data Management
System
3.3
Features and Functionalities of the STAT
4.0
Conclusion
5.0
Summary
6.0
Tutor Marked Assignment
7.0
Further Reading and Other Resources
1.0
Introduction
The SPEA Smart Airport Statistical Data Management System
(SMART STAT) is an intelligent system that transforms raw flight and
Airport operational data into management information. STAT performs
statistical analysis on flight and Airport operational data and makes data
available to other SMART systems.
Figure 41: Diagram of SPEA Smart Airport Statistical Data Management System
2.0
Objective
At the end of this unit, you should be able to:
Appreciate the uses of the Airport Statistical Data Management
System.
State the major concepts in Airport Statistical Database System.
Draw the major components of Airport Statistical Management
Diagrams, Naming Conventions and Design Issues in data
management system
Explain what Airport Statistical database model is all about
Mention benefits of Airport Statistical Management System.
3.0
Uses of the Airport Statistical Data
Management System
•
Evaluate the current status of ground handling operations.
•
Optimize the deployment of existing Airport resources.
•
Determine future need for additional Airport resources.
•
Consolidate data for billing.
3.1
Benefits of the Airport Statistical Data
Management System
•
Improved tactical and strategic planning.
•
Improved management decision making.
•
Reduced operating cost and capital required for Airport expansion.
•
Availability of concise structured information from customized
reports.
•
Flexibility in accessing data using standard query languages.
•
Accurate and precise information for billing.
•
Authorized users can access information via workstations or the
Internet.
3.2
Outline of the Airport Statistical Data
Management System
The SPEA Smart Airport Statistical Data Management System (SMART
STAT) is an intelligent system that transforms raw flight and Airport
operational data into management information. The ability to manipulate
and perform statistical analysis on historical data from any source is a
powerful tool for Airport Management. It allows them to evaluate the
current status of ground handling operations and to make informed
decisions on current and future operations and resource requirements.
Simple selections allow the user to run queries and also create useful
customized reports on data within the SMART system and other Airport
databases. SMART STAT allows users to access the database via Standard
Query Language (SQL). Microsoft .Net, Open Database Connectivity
(ODBC), Java Database Connectivity (JDBC) and other.
The following flight data is available for statistical analysis:
•
Flight information - flight number, flight date and airline.
•
Complete flight routing - scheduled time of departure and arrival for
all Airports, including multiple city/leg flights.
•
Flight qualification - flight type, traffic type, loading type and service
type.
•Aircraft information - aircraft type, aircraft registration number and
aircraft configuration.
•
Passenger information - terminating/transfer/transit passengers,
passengers per weight category, passengers per class and special
requirements.
•
Baggage information – number and type bags, mishandling and
irregularities.
•Load information - mail, cargo, crew and passengers.
•
Times - scheduled times, estimated times, touch-down/take-off times,
block on/off times, red times, zone-in times and boarding/last
call/flight closed times.
•
Resources allocated to the flight - check-in counters, boarding gate,
baggage reclaim belt and parking stand.
•
Punctuality and regularity - flight delays and delay codes.
SMART STAT integrates fully with the Airport Operational DataBase
System (SMART AODB) and other SMART Systems and interfaces with
other Airport databases. Data is immediately available for analysis as it is
recorded in any database. The Airport Billing System (SMART BILL)
extracts data from SMART STAT for accurate billing. Statistical data can
also be exported to any other billing software operated at the Airpor
3.3
Features and Functionalities of the STAT
•
High performing and secure Oracle 10g database.
•
Access the database via SQL, Microsoft . Net, ODBC, JDBC and
other.
•
User-friendly and highly configurable Windows browser based
Graphical User Interface (GUI).
•
Comply with International Air Transport Association (IATA)
standards.
•
Full seamless integration with all SMART Systems.
•
Robust, reliable and scalable system.
•
Supports concurrent users and is accessible from any authorized
workstation.
•
Data can be automatically backed-up on different media types.
4.0
Conclusion
With the overview of the Airport database management system,
individuals and organizations can uncover hidden processes,
methodologies, as well as the benefits of managing their data, which they
can use to predict the behaviour of customers, products and processes.
5.0
Summary
In this unit we have learnt that:
How to use Statistical database system in an Airport.
How to design a model for an Airport Statistical Database Management
system, which is a collection of programs that manage the database
structure and that control shared access to the data in the database.
The various benefits that can be derived in using this system, like
controlling redundancy, restricting unauthorized access, saving time etc,
with flexibility, economies of scale, and potential for enforcing
standards, are some of the implications of database approach.
6.0
Tutor Marked Assignment
1. (a) What are the benefits of Airport database management system?
2. (a) What are the characteristics of Airport database approach?
(b) Mention the advantages of using Airport database management
approach over other methods
7.0
Further Reading and Other Resources
SPEA SMART Airport Passenger and Baggage Management Systems
SPEA SMART Airport Resource Utilization Systems
SPEA SMART Airport Technology Support Systems
SPEA SMART Airport Data Management Systems
Statistical Data Management System
Semantic Message Processor System
Airport Operational DataBase System
Airport Departure Control System
Airport Operational DataBase System
Airport Resource Management System
Airport Stand Allocation System
Airport Baggage Reconciliation System
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement