Lesson Business intelligence systems

Lesson Business intelligence systems present the content: organizations need business intelligence; business intelligence systems are available; typical reporting applications; typical data-mining applications; purpose of data warehouses and data marts; business intelligence applications delivered

Nga.lethiquynh@ueh.edu.vn

Q3 What are typical reporting applications?

Q4 What are typical data-mining applications?

Q5 What is the purpose of data warehouses

and data marts?

Q6 How Are Business Intelligence

Applications Delivered?

Q1 Why do organizations

need business intelligence?

Why do organizations need BI?

► Data communications and data storage

are essentially free, and enormous

amounts of data are created and

stored every day:

► 12 000 gigabytes per person of data,

worldwide, in 2009

► Businesses use BI systems to:

► Process data (from operational DB, Social

Data, purchased data, etc.)

► produce patterns, relationships, and other

forms of information;

► deliver that information on a timely basis to

users who need it

► Example:

► Identifying changes in purchasing patterns

► BI for entertainment: Netflix has data on

watching, listening, and rental habits

determines what people actually want

Q2 What business

intelligence systems are

available?

► Organization needs

OPERATIONAL DATA INFROMATION TO SUPPORT

BI systems (tools)

Business intelligence (BI) tools

BI systems provide valuable information for

decision-making

Three primary BI systems

1. Reporting tools

► integrate data from multiple systems

► sorting, grouping, summing, averaging, comparing data.

2. Data-mining tools

► use sophisticated statistical techniques, regression analysis

and decision tree analysis

► used to discover hidden patterns and relationships

►

3. Knowledge-management tool

► creates value by collecting and sharing human

knowledge about products, product uses, best

practices, other critical knowledge

► used by employees, managers, customers,

suppliers, others who need access to company

knowledge

Tools versus applications

versus systems

► BI tool is one or more computer programs BI

tools implement the logic of a particular

procedure or process

► BI application is the use of a tool on a

particular type of data for a particular

purpose

► BI system is an information system having all

five components that delivers results of a BI

application to users who need those results

Q3 What are typical

reporting applications?

Operations commonly used by reporting

tools

Raw Data

groupingfilteringcalculating

formatting

sorting

Basic reporting operations

List of sales data

Data sorted by

customer name

Sales data filtered to show

repeat customers, and formatted

for easier understanding

► A reporting application is a BI

application that inputs data from one

or more sources and applies a reporting

tool to that data to produce

information.

► Important reporting applications:

► RFM analysis

► OLAP

RFM analysis

RFM analysis allows you to analyse and rank

customers according to their purchasing

► M = how much money a customer typically

spends on your products

Divides customers into five groups and assigns a

score from 1 to 5:

• R score 1 = top 20 per cent of 'most recent orders'

• R score 5 = bottom 20 per cent (longest since last

• M score 1 = top 20 per cent of 'most money spent'

• M score 5 = bottom 20 per cent 'who spent least'.

Example of RFM score data

► Ajax has ordered recently and orders

frequently M score of 3 indicates it does not

order most expensive goods:

► a good and regular customer but need to attempt to

up-sell more expensive goods to Ajax.

► Bloominghams has not ordered in some time,

but when it did, ordered frequently and

orders were of highest monetary value:

► may have taken its business to another vendor

► sales team should contact this customer

immediately.

Interpreting RFM score results

► Caruthers has not ordered for some

time, did not order frequently and did

not spend much:

► sales team should not waste any time on

this customer

► Davidson in middle

► set up on automated contact system or use

the Davidson account as a training

exercise

► OLAP : more generic than RFM

► OLAP provides the ability to sum,

count, average and perform other

simple arithmetic operations on groups

Features of OLAP reports

OLAP reports

► Have:

► measures: the data item of interest

► Example: Total sales, average sales, and average cost

► Dimension: a characteristic of a measure

► Example: Purchase date, customer type, customer location, and sales region

► A presentation like above is Also called OLAP cube:

► presentation of measure with associated dimensions.

► Users can alter format

► Users can drill down into data, i.e divide data into

more detail

► May require substantial computing power

Source: textbook[1], pg 292

OLAP product family and

store location by store type

Source: textbook[1], pg 293

location by store type, drilled down

to show stores in California

► stores results in OLAP database.

► Third-party vendors provide software

for more extensive graphical displays.

Source: textbook[1], pg 296

Q4 What are typical

data-mining applications?

► Data mining is the application of statistical

techniques to find patterns and relationships

among data for classification and prediction

Data mining application

► Data Mining = Knowledge Discovery in

Database (KDD)

► Categories:

► Unsupervised Data Mining

► supervised Data Mining

► Analysts do not create model before running analysis.

► Apply data-mining technique and observe results

► Analysts create hypotheses after analysis to explain

patterns found

▪ No prior model about the patterns and

relationships that might exist

► Common statistical technique used:

▪ Cluster analysis to find groups of similar customers

from customer order and demographic data

Supervised Data Mining

► Model developed before analysis

► Statistical techniques used to estimate

parameters

► Examples:

▪ Regression analysis—measures impact of set

of variables on one another

▪ Used for making predictions

► Market-basket analysis is a data-mining technique

for determining sales patterns

▪ Uses statistical methods to identify sales

patterns in large volumes of data

▪ Shows which products customers tend to buy

together

▪ Used to estimate probability of customer

purchase

▪ Helps identify cross-selling opportunities

► "Customers who bought book X also bought book Y”

Market-Basket Analysis

► Terms:

and B will be purchased together

customer will buy B if he/she bought A

or decreases when other products are purchased

Lift= 0.926/0.7=

1.322

Mask and Fins

Market-Basket Analysis

Market-Basket Analysis

Q5 What is the purpose of

data warehouses and data

marts?

Warehouses and Data Marts?

► Purpose:

▪ To extract, clean and prepare data from

various operational systems and other

sources

▪ To store and catalog data for BI processing

▪ Stored in data-warehouse DBMS

Components of a Data Warehouse

Source: textbook[1], pg 304

► Internal operations systems

outside sources

user-generated content applications

data-warehouse meta database

Data Warehouses vs Data Marts

Data mart is a collection of data

▪ Created to address particular needs

▪ Smaller than data warehouse

▪ Users may not have data management

Q6 How Are Business

Intelligence Applications

Delivered?

Applications Delivered?

► the authorized allocation of BI results

to users

► BI servers can be:

► Website from which users can download, or

pull, BI application results

interface

► A BI application server: to support user

subscriptions to particular BI application

results (e.g alerts via email or phone

whenever a particular event occurs)

► use metadata to determine what results

to send to which users and, on which

schedule

► BI results can be delivered to “any”

device .

► exception alert

► a dramatic fall in a stock price or

exceptionally high sales volume

Q3 What are typical reporting applications?

Q4 What are typical data-mining applications?

Q5 What is the purpose of data warehouses

and data marts?

Q6 How Are Business Intelligence

Applications Delivered?

► What is Business Intelligence?

► RFM analysis for customer segmentation

and loyalty marketing

► 5 Techniques that make RFM analysis

work for you