DATA MANAGEMENT

1
DATA MANAGEMENT
CONTENT 1. Data Collection 2. Sources of
External Data 3. Structures of Databases 4.
Data Warehouse 5. Data Visualisation
2

• 2-3 million data records are processed monthly
• How to manage?
• How to use for decision support?
• How to hold down costs?
• How to improve customer service?
• How to utilize resource effectively?
• How to improve service quality?
• Answers
• Develop a comprehensive database (data
  warehouse) and DSS approach
• Very effective

3
Data Warehousing and Online Analytical Processing
(OLAP)
4
Data Collection

• manually
• by instruments and sensors
• Methods for data collection
• surveys (e.g. questionnaires)
• observations (e.g. video cameras)
• from experts (e.g. interviews)
• Problems with data
• data are not correct
• data are not timely
• data are not measured or indexed properly
• needed data simply do not exist

5
Sources of External Data
1. Internet e.g. home pages of vendors,
clients, competitors view / download
information 2. Commercial data banks sell
access to specialised databases e.g. CompuServe,
Compustat, Dow Jones Information
Service Interactive Data Corporation, Lockheed
Information Systems, Mead Data Central
6
Structures of Databases

• 1. Relational Databases
• Two-dimensional tables columns fields
• rows records
• Data files may be related by means of a common
  data field

• Advantages
• simple to use
• can be easily expanded and altered

7

• 2. Hierarchical Databases
• Orders the data items in a top-down fashion,
  like a tree.

8

• 3. Network Databases
• Permits more complex links than hierarchical
  database.
• Enables sharing of some items and this save
  storage space.

9

• 4. Object-Oriented Databases
• Appropriate for complex data.
• Based on the principle of object-oriented
  programming language.
• Objects encapsulate data and procedural code.
• Enable nesting of objects.
• Hierarchy of classes.
• 5. Multimedia-Based Databases
• Manage data in a variety of formats in addition
  to the standard text or numeric fields,
  including images, video clips, sound.
• 6. Document-Based Databases
• Electronic Document Management System (EDM).
• Enables paper storage.

10
Data Warehouse

• Data warehouse establish a data repository that
  makes operational data accessible in a form that
  is acceptable for decision support systems and
  for executive support systems.
• Benefits
• Supports all decision maker's data requirements.
• Provides summary information.

11

• Suitability
• Appropriate for organizations where
• Data are in different systems,
• Same data have different representations in
  different systems.
• Characteristics
• 1. Data organized by detailed subject with
  information relevant for decision support
• 2. Integrated data 3. Time-variant data
• 4. Non-volatile data

12

• Online analytical processing
• Done by end-users of decision support systems
  and executive support systems
• Includes activities such as generating queries,
  requesting ad hoc reports, conducting
  statistical analysis
• Using SQL for Querying
• SQL - structured query language is a language
  which becomes astandard for data access and
  manipulation in relational databasemanagement
  systems
• Example
• "Identify the employees whose monthly salary is
  greater than 2000."
• SELECT Name, Salary
• FROM Employees
• WHERE Salary gt 2000

13

• Data Mining
• Term used to describe knowledge discovery
  in database.
• The activity of looking for very specific,
  detailed, but unknown, information in databases.

14
User queryShow Revenuesfor March 1991By
Salesperson
Data mining
Reports
SQL Query
Network
Data warehouse
Data mining
SalesDatabase
Network
ResultsBrown 20,000 Jones 30,000...
Data warehouse
15
Data Visualisation

• Data visualisation is a technology that
  supports visualisation of information including
  digital images, graphical user interface, multidi
  mensions, tables and graphs, geographic
  information systems, virtual reality, 3-D
  presentations and animation.
• Multidimensional Presentation
• Factors usually considered
• Dimensions products, salespeople, industry,
  country,...
• Measures money, sales volume,
• Time daily, weekly, monthly, quarterly, yearly
• Example. Manager wants to know the sales of a
  productin a certain geographical area, by a
  specific salesperson,during a specified month,
  in terms of specified units

16

• Limitations
• Multidimensional databases require more
  space (up to 40 or more) than a summarised
  relational database.
• Multidimensional product has higher
  costs (usually 50 or more) than standard
  relational product.
• Loading of multidimensional databases consume
  system resources and time.
• Interface and maintenance are more complex for
  multidimensional databases.

17

• Summary
• Data for IS must be collected frequently using
  one of several methods.
• Data are organised in either relational,
  hierarchical or network architectures. For many
  DSS the relational type is preferable.
• Object-oriented and multimedia databases are
  becoming increasingly more important for
  decision making applications.
• Data for DSS are often processes and stored in
  a data warehouse to faster accessibility.
• Online analytical processing (OLAP is a set of
  tools for timely data analysis.
• Data mining is the discovery of knowledge in
  databases.
• Data multidimensionality enables people to view
  data quickly by different dimensions.