business intelligence technologies

8/14/2019 Business Intelligence Technologies

1/29

First page

Business IntelligenceTechnologies

Donato MalerbaDipartimento di Informatica

Universit degli Studi, Bari, Italy

[email protected]

http://www.di.uniba.it/malerba


2/29

First page

Business Intelligence

s Business Intelligence is a global term

for all the processes, techniques and

tools that support business decision-

making based on informationtechnology.

s The approaches can range from a

simple spreadsheet to a major

competitive undertaking.s Data mining is an important new

component of business undertaking.


3/29

Business IntelligenceTechnologies

Data Sources

Paper, Files, Information Providers, Database Systems, OLTP

Data Warehouses / Data Marts

Data Exploration

OLAP, DSS, EIS, Querying and Reporting

Data Mining

Information discovery

Data Presentation

Visualization Techniques

Decision

Making

Increasing potentialto support

business decisions

End User

Business

Analyst

Data

Analyst

DB

Admin


4/29

First page

Business Processess

Data for support decision making

decisional

processes(agreement with a credit card)

managementprocesses

(grant a loan)

operational

processes(transaction on

bank account

(Ex.: Banking)

DSS o EIS

MIS

TPS

s Different information systems support the different processes


5/29

First page

DSS vs. EISs Decision Support Systems (DSS) and

Executive Information Systems (EIS):

information systems designed to help

managers in making choices.

s Different, yet interrelated applicationss A DSS focuses on a particular decision,

whereas an EIS provides a much wider

range of information (e.g., information on

financials, on production history, and on

external events).

s DSSs appeared in the 1970s

s EISs appeared in the 1980s.


6/29

First page

DSS vs. EISs The original EISs did not have the

analytical capabilities of a DSS

s An EIS is used by senior managers to

find problems; the DSS is used by the

staff people to study them and to offeralternatives (Rockart and Delong, 1988)

EIS DSS


7/29

First page

Where do Data ComeFrom?

s The EISs and DSSs often lacked a strong

database component.

s Most organizational information gathering

was (and is) directed to maintaining

current (preferably on-line) informationabout individual transactions and

customers.

s Managerial decision making requires

consideration of the past and the future,not just the present.

s New databases, called data warehouses,

were created specifically for analytic use


8/29

First page

A Data Warehouse is ...

A data warehouse is ax subject-oriented,

x integrated,

x time-variant, and

x nonvolatile

collection of data in support of managementsdecisionsInmon, W.H.

Building the Data WarehouseWellesley, MA: QED Tech. Pub. Group,1992


9/29

First page

subject-oriented ...

s The data in the warehouse is definedand organized in business terms, and is

grouped underbusiness-orientedsubject headings, such as

x customers

x products

x sales

rather than individual transactions.

s Normalization is not relevant.


10/29

First page

integrated ...s The data warehouse contents are defined such that

they are valid across the enterprise and its operationaland external data sources

Operational systems

Data warehouse

s The data in the warehouse should be

x clean

x validated

x properly integrated


11/29

First page

time-variant ...

s All data in the data warehouse is time-

stamped at time of entry into the

warehouse or when it is summarized

within the warehouse.s This chronological recording of data

provides historical and trend analysis

possibilities.

s On the contrary, operational data is

overwritten, since past values are not of

interests.


12/29

First page

nonvolatile ...

s Once loaded into the data warehouse, the

data is not updated.

s Data acts as a stable resource for

consistent reporting and comparativeanalysis.

s On the contrary, operational data is

updated (inserted, deleted, modified).


13/29

First page

Which Data in theWarehouse?

s

A data warehouse contains five types ofdata:

x Current detail data

x Old detail data

x Lightly summarized data

x Highly summarized data

x Metadata

s Granularityof the data: a key designissue


14/29

First page

Flow of Data

Operational

Environment

Clean thedata

Reside in

warehouse

Purge

Summarize

Archive


15/29

An Example of DataIntegration

Checking Account System

Jane Doe (name)

Female (gender)

Bounced check #145 on 1/5/95

Opened account 1994

Checking Account System

Jane Doe (name)Female (gender)

Bounced check #145 on 1/5/95

Opened account 1994

Savings Account System

Jane Doe

F (gender)

Opened account 1992

Savings Account System

Jane Doe

F (gender)

Opened account 1992

Investment Account System

Jane Doe

Owns 25 Shares Exxon

Opened account 1995

Investment Account System

Jane Doe

Owns 25 Shares Exxon

Opened account 1995

Customer

Jane Doe

Female

Bounced check #145

Married

Owns 25 Shares ExxonCustomer since 1992

Customer

Jane DoeFemale

Bounced check #145

Married

Owns 25 Shares ExxonCustomer since 1992

Operational

data

datawarehouse


16/29

First page

Cost and Size of a DataWarehouse

s Data warehouses are expensive

undertakings (mean cost: $2.2 million).

s Since a data warehouse is designed for

the enterprise it has a typical storage

size running from 50 Gb to over aTerabite.

s Parallel computingto speed up dataretrieval

WAREHOUSE SIZE SERVER REQUIREMENTS

5-50 GB Pentium PC > 100MHz

50-500 GB SMP machine

> 500 GB SMP or MPP machine


17/29

First page

The Data Mart

s

A lower-cost, scaled-down version of thedata warehouse designed for the

strategic business unit (SBU) or

department level.

s An excellent first step for manyorganizations.

s Main problem: data marts often differ

from department to department.s Two approaches:

x data marts enterprise-wide systemx data warehousedata marts

A A hit t f D t


18/29

An Architecture for DataWarehousing

operational

databases

external sources

data

warehouse

extraction

cleaning

validation

summariz.

data mart

metadata

usedby

EIS

DSS

OLAP

data

mining

query

used

by


19/29

First page

On-Line AnalyticalProcessing (OLAP)

s Term introduced by E.F. Codd (1993) in

contrast to On-Line Transaction

Processing (OLTP)

s

The OLAP Councils definition:A category of software technology that

enables analysts, managers and executives

to gain insight into data through fast,

consistent, interactive access to a widevariety of possible views of information that

have been transformed from raw data to

reflect the real dimensionality of the

enterprise as understood by the user


20/29

First page

On-Line AnalyticalProcessing (OLAP)

s Basic idea: users should be able tomanipulate enterprise data models

across many dimensions to understand

changes that are occurring.

s Data used in OLAP should be in the

form of a multi-dimensional cube.

Time

Marke

t

Product


21/29

First page

DimensionalHierarchies

s

Each dimension can be hierarchicallystructured

Item

Product

Type of product

Day

Week

Month

Year

Store

City

State

Country


22/29

First page

OLAP Operations

s

Rollup: decreasing the level of details Drill-down: increasing the level of detail

s Slice-and-dice: selection and projection

s Pivot: re-orienting the multidimensional view of data


23/29

First page

Implementing Multi-dimensionality

s

Multi-dimensional databases (MDDB)s To make relational databases handle

multidimensionality, two kinds of tables are

introduced:

x Fact table: contains numerical facts. It islong and thin.

x Dimension tables: contain pointers to the

fact table. They show where the

information can be found. A separatetable is provided for each dimension.

Dimension tables are small, short, and

wide.


24/29

First page

Star Schema

STORE KEYPRODUCT KEY

PERIOD KEY

Dollars

Units

Price

STORE KEY

Store Desc.

City

State

District ID

District Desc.

Region ID

Region Desc.

Regional Mgr.Level

PERIOD KEY

Period Desc.

Year

QuarterMonth

Day

PRODUCT KEY

Product Desc.Brand

Color

Size

Manufacturer

Fact Table

Time Dimension

Product Dimension

Market Dimension


25/29

First page

MOLAP, ROLAP, DSS

s

The OLAP technology is considered anextension of the original DSS technology.s DSS applications are tools that access and

analyze data in relational database (RDB)

tables.s OLAP tools access and analyze

multidimensional data (typically three, up to

ten-dimensional data).

s OLAP technology is called MOLAP/ROLAP(multidimensional/relational OLAP) if it uses

an MDDB/RDB.


26/29

First page

OLAP/DSS

s

OLAP tools focus on providing multi-dimensional data analysis, that is superior to

SQL in computing summaries and

breakdowns along many dimensions.

s

OLAP tools require strong interaction fromthe users to identify interesting patterns in

data.

s An OLAP tool evaluates a precise query that

the user formulates.s OLAP users are farmers.

D W h


27/29

First page

Data Warehouse Data Mining

The rational to move from the datawarehouse to data mine arises from the

need to increase the leverage that an

organization can get from its existing

warehouse approach.

After implementing a data mining solution, an

organization could decide to integrate the solution

in a broader data-driven approach to businessdecision making. The data warehouse will provide

an excellent vehicle for such an integration.

r ca uccess ac ors


28/29

First page

r ca uccess ac orsfor BusinessApplications

s Peoplex Find a sponsor for the application

x Select the right user group

x Involve a business analyst with domain

knowledgex Collaborate with experienced data analysts

s Data

x Select relatively clean sources of data

x Select a limited set of data sources (e.g., thedata warehouse)

r ca uccess ac ors


29/29

First page

r ca uccess ac orsfor BusinessApplications (cont.)

s Applicationx Understand business objectives.

x Analyze cost-benefits and significance

of the impact on business problem.

x Consider legal or social issues in

collecting input data

business intelligence technologies

Documents