What is BI?

Business Intelligence (BI) is about getting

the right information, to the right decision
makers, at the right time.

BI is an enterprise-wide platform that

supports reporting, analysis and decision
making.

BI leads to:
❖ fact-based decision making
❖ “single version of the truth”

BI includes reporting and analytics.

Business Intelligence by Other Names

Reporting

Decision Support
Competitive
System
Intelligence

Business
Business
Investment
Analytics

Business
Insight
BI Defined

Howard Dresner, of the Gartner Group, in 1989 coined the term BI. He
defined BI as
“a set of concepts and methodologies to improve decision making in
business through use of facts and fact-based systems”.
BI Defined
BI is…

• Fact-based decision making

• Single version of truth

• 360 degrees perspective on your business

• Virtual team members on the same page

• Visibility into enterprise performance

Why BI?

What happened?
Know the past
Why did it happen?

What is happening?
Analyze the Present
Why is it happening?

What will happen? Predict the Future

How BI?
IT Department
Business Users
Source Systems

ORDERS Data Warehouse

BI Tools

INVOICES •Report
•Extract
Relational Tables •Query
•Transform •Analyze
•Load •Alert
SHIPPING

Web Clicks OLAP

External
Data
Analytic Applications
(Product Profitability, Workforce Planning, CRM)
Challenges in Retail Industry

RETAIL
SHOP

Manage
Cost to Serve ? Inventory?
Promotional Attract
Offers? Customers?

30%
Off
 Retail Industry Leveraging BI

Data Warehouse
Daily Sales
Transactions

Customers Transactions

Customer,
Demographics Products
or products/price
details

Online Shopping
Optimization
Reporting Tools Analytics Tools Best Practices
Techniques

Optimized
Optimized Costs Inventory
New Identifying
and
Promotions Customer
Allocation
Needs

30%
Off
Data Mining in Retail

The Diapers-Beer Example

• A (hypothetical) pattern learned from transaction data: “On Friday evenings,

shoppers who buy diapers also buy beer.”

• Highlights new, surprising correlations that can be acted on by the store.

– To promote more users to display this behavior, consider dynamic store
layout decisions that might alter locations of products based on co-
purchase
– Consider couponing strategies that can be used to cross promote related
products in some cases.
Data Mining for Marketing

Online Product Recommendations

• Amazon.com pioneered the use of collaborative filtering based
approaches to recommend products to users online.
BI Applications

BI applications can be divided into:

• Technology solutions
– DSS- decision support system
– EIS-Executive information system
– OLAP- online analytical processing
– Managed Query and Reporting
– Data Mining
• Business Solutions
– Performance Analysis
– Customer Analysis
– Market Place Analysis
– Productivity Analysis
– Sales Channel Analysis
– Behavioral Analysis
– Supply Chain Analysis
Open Source BI Tools

RDBMS MySQL, Firebird

Pentaho Data Integration (formerly

ETL Tools
called Kettle), SpagoBI

Analysis Tools Weka, RapidMiner, SpagoBI

Reporting Tools/Ad Hoc

Pentaho, BIRT, Actuate, Jaspersoft
Querying/Visualization
Popular BI Tools

AB INITIO 3.0.2 SPOTFIRE (TIBCO) 3.2.x

INFORMATICA 9 SAS 9.2

NCR TERADATA 13 MICROSTRATEGY 9

MICROSOFT
SQL SERVER 2008 SSIS 2008 SSRS 2008 SSAS 2008

SAP SAP
BUSINESS
SYBASE IQ
OBJECTS 5.x

ORACLE
ORACLE 11G R2 ORACLE WAREHOUSE BUILDER SIEBEL 8.1 HYPERION 11.1.3

IBM
NETEZZA 4.6, DB2 DATASTAGE 8.5 COGNOS v10 SPSS 9

BIRT

PENTAHO RAPIDMINER

MYSQL WEKA

ANALYTICS,
ETL, DATA OLAP, REPORTING,
DBMS ANALYSIS VISUALIZATION,
INTEGRATION DATA WAREHOUSING AD HOC QUERYING MINING

Back End BI Functions Front End

BI Framework

Business Requirement
BI Architecture

Data Resource Administration Data Warehousing

Data Sources

BI & DW Operations
Program Management

Data Acquisition, Cleaning & Integration

Development
Data Stores

Information Services
Information Delivery Business Analytics

Business Applications

Business Value

Source: TDWI
Business Layer

BUSINESS
BUSINESS VALUE
REQUIREMENTS

BUSINESS
LAYER

PROGRAM
DEVELOPMENT
MANAGEMENT
Business Layer
Business requirements: The requirements are a product of three steps of a
process that includes:

➢ Business drivers (the impulses that initiate the need to act).

Examples: changing workforce, changing labor laws, changing
economy, changing technology, etc.
➢ Business goals (the targets to be achieved in response to the business
drivers).
Examples: increased productivity, improved market share, improved
profit margins, improved customer satisfaction, cost reduction, etc.
➢ Business strategies (the planned course of action that will help achieve
the set goals).
Examples: outsourcing, global delivery model, partnerships, customer
retention programs, employee retention programs, competitive pricing,
etc.
Business Layer

Business Value: Business value can be measured in terms of ROI (Return on

Investment), ROA (Return on Assets), TCO (Total Cost of Ownership),
TVO (Total Value of Ownership), etc.

Program management: It is the component that ensures people, projects and

priorities work in a manner in which individual processes are compatible
with each other; so as to ensure seamless integration and smooth
functioning of the entire program.

Development: The process of development consists of database/data-

warehouse development (consisting of ETL, data profiling, data cleansing
and database tools), data integration system development (consists of data
integration tools and data quality tools) and business analytics development
(about processes and various technologies used).
Administration and Operations Layer

BI Architecture
• Should follow design standards
DATA • Must have a logically apt data model
• Metadata should be of high standard
• Performed according to business semantics and rules
• During integration, certain processing standards have
INTEGRATION
to be followed
• Data must be consistent

• Information derived from data that has been

INFORMATION integrated should be usable, findable and as per the
requirements

• Technology used for deriving information must be

accessible
TECHNOLOGY • Also, it should have a good user-interface
• Should support analysis, decision support, data and
storage management

• Consists of different roles and responsibilities, like

ORGANIZATION
management, development, support and usage roles
 Data Warehousing - Overview

The term "Data Warehouse" was first coined

by Bill Inmon in 1990. According to Inmon, a
data warehouse is a subject oriented,
integrated, time-variant, and non-volatile
collection of data.
A data warehouses provides us generalized
and consolidated data in multidimensional
view. Along with generalized and
consolidated view of data
• A data warehouse is a database, which is kept separate
from the organization's operational database.
• There is no frequent updating done in a data warehouse.
• It possesses consolidated historical data, which helps
the organization to analyze its business.
• A data warehouse helps executives to organize,
understand, and use their data to take strategic
decisions.
• Data warehouse systems help in the integration of
diversity of application systems.
• A data warehouse system helps in consolidated
historical data analysis.
 Data Warehouse Features

• Subject Oriented − A data warehouse is subject oriented because it

provides information around a subject rather than the organization's ongoing
operations. These subjects can be product, customers, suppliers, sales,
revenue, etc. A data warehouse does not focus on the ongoing operations,
rather it focuses on modelling and analysis of data for decision making.
• Integrated − A data warehouse is constructed by integrating data from
heterogeneous sources such as relational databases, flat files, etc. This
integration enhances the effective analysis of data.
• Time Variant − The data collected in a data warehouse is identified with a
particular time period. The data in a data warehouse provides information
from the historical point of view.
• Non-volatile − Non-volatile means the previous data is not erased when
new data is added to it. A data warehouse is kept separate from the
operational database and therefore frequent changes in operational
database is not reflected in the data warehouse.
OLAP (for online analytical processing) is
software for performing multidimensional
analysis at high speeds on large volumes of data
from a data warehouse, data mart, or some
other unified, centralized data store
OLTP (online transaction processing) is a
class of software programs capable of
supporting transaction-oriented applications
on the Internet.
Sr.N Data Warehouse (OLAP-On Line Analytical Operational Database(OLTP On Line
o. Processing)
1 It involves historical processing of
information.
2 OLAP systems are used by knowledge
workers such as executives, managers, and database professionals.
analysts.
3 It is used to analyze the business.
4 It focuses on Information out.
5 It is based on Star Schema, Snowflake
Schema, and Fact Constellation Schema.
6 It focuses on Information out.
7 It contains historical data.
8 It provides summarized and consolidated
data.
9 It provides summarized and
multidimensional view of data.
10 The number of users is in hundreds.
11 The number of records accessed is in
millions.
Transaction Processing)
It involves day-to-day processing.
OLTP systems are used by clerks, DBAs, or
It is used to run the business.
It focuses on Data in.
It is based on Entity Relationship Model.
It is application oriented.
It contains current data.
It provides primitive and highly detailed data.
It provides detailed and flat relational view of
data.
The number of users is in thousands.
The number of records accessed is in tens.
CRISP-DM
stands for cross-industry process for data mining

Business
Understanding
Modeling

Data
Understanding CRISP- Evaluation
DM

Data Deployment
Preparation
CRISP-DM
stands for cross-industry process for data mining

Business
Understanding

Set the objectives

CRISP-
DM
CRISP-DM
stands for cross-industry process for data mining

Data
Understanding

CRISP-
DM
Data collection
Data exploration
Assessing data
quality
 CRISP-DM
stands for cross-industry process for data mining

Data
Preparation

Data integration (if the data is from different sources)

CRISP-
Data cleaning DM
Missing value analysis
Data imputation
Variable standardization
Feature selection/Feature engineering
Outliers detection and treatment
CRISP-DM
stands for cross-industry process for data mining

Modeling

Model selection and assumptions if

any
CRISP-
DM Model output

Model interpretation from business

point of view
CRISP-DM
stands for cross-industry process for data mining

CRISP- Evaluation
DM

Deployment