Data Warehousing &

Business Intel
DS-308
Course Instructor: Hamza Ali
2 Lecture 3
Types & Typical Applications of DWH

Outline:
Different types of DWH
 Financial
 Telecommunication
 Insurance
Typical Application of DWH
 Fraud Detection etc.…
3 Types of Data Warehouse:

 Financial
 Telecommunication
 Insurance
 Human Resource
 Global
 Exploratory
4 Types of Data Warehouse:
Financial:
 First data warehouse that an organization builds. This is appealing because:

 Nerve center, easy to get attention.

 In most organizations, the smallest data set.

 Touches all aspects of an organization, with a common denomination i.e. money.

 Inherent structure of data directly influenced by the day-to-day activities of

financial processing.
 Types of Data Warehouse:
5
Telecommunication
Dominated by sheer volume of data.

Many ways to accommodate call level detail:

 Only a few months of call level detail,
 Storing lots of call level detail scattered over different
storage media,

 Storing only selective call level detail, etc.

 Unfortunately, for many kinds of processing, working at

an aggregate level is simply not possible.
 Types of Data Warehouse:
6
Insurance
Insurance data warehouses are similar to other
data warehouses BUT with a few exceptions.
 Stored data that is very, very old, used for actuarial
processing.

 Typical business may change dramatically over last 40-

50 years, but not insurance.

 In retailing or telecom there are a few important dates,

but in the insurance environment there are many dates
of many kinds.
 Types of Data Warehouse:
7
Insurance
Insurance data warehouses are similar to other
data warehouses BUT with a few exceptions.
 Long operational business cycles, in years. Processing
time in months. Thus, the operating speed is different.

 Transactions are not gathered and processed, but are in

kind of “frozen”.

 Thus, a unique approach to design and implementation.

 Typical Applications:
8
Impact on organization’s core business is to streamline and maximize
profitability.

 Fraud detection.
 Profitability analysis.
 Direct mail/database marketing.
 Credit risk prediction.
 Customer retention modeling.
 Yield management.
 Inventory management.

.
 Typical Applications:
9

Fraud detection
 By observing data usage patterns.
 People have typical purchase patterns.
 Deviation from patterns.
 Certain cities notorious for fraud.
 Certain items bought by stolen cards.
 Similar behavior for stolen phone cards.
 Typical Applications:
10
Profitability Analysis
 Banks know if they are profitable or not.
 Don’t know which customers are profitable.
 Typically, more than 50% are NOT profitable.
 Don’t know which one?
 Balance is not enough; transactional behavior is the key.
 Restructure products and pricing strategies.
 Life-time profitability models (next 3-5 years).
 Typical Applications:
11

Direct mail marketing

 Targeted marketing.
 Offering high bandwidth package NOT to all users.
 Know from call detail records of web surfing.
 Saves marketing expense, saving pennies.
 Knowing your customers better.
12 Typical Applications:
Credit risk prediction
 Who should get a loan?
 Customer segregation i.e. stable vs. rolling.
 Qualitative decision making NOT subjective.
 Different interest rates for different customers.
 Do not subsidize bad customer on the basis of good.
 Typical Applications:
13
Yield Management
 Works for fixed inventory businesses.
 The price of item suddenly goes to zero.
 Item prices vary for varying customers.
 Example: Air Lines, Hotels etc.
 Price of (say) Air Ticket depends on:
 How much in advance ticket was bought?
 How many vacant seats were present?
 How profitable is the customer?
 Ticket is one-way or return?
 Recent Application
14
Agriculture Systems
 Agri and related data collected for decades.
 Metrological data consists of 50+ attributes.
 Decision making based on expert judgment. Recent Application
 Lack of integration results in underutilization.
 What is required, in which amount and when?
Questions?????