Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd
Upload
15 views

Lecture Slide

wwd

Uploaded by

ayk0227
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
15 views

Lecture Slide

wwd

Uploaded by

ayk0227
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 12

Lecture 9

Smart Data and Discovery


Outline:

3. Datawarehouse Architecture
Data Mining Applications

Data mining has been named differently such as


knowledge discovery, business intelligence, predictive
modelling, predictive analytics and so on.

Data Mining is a business process for exploring large


amount of data to discover meaningful patterns and
rules.
Data Mining Applications

● Financial Data Analysis


● Retail Industry
● Telecommunication Industry
● Biological Data Analysis
● Other Scientific Applications
● Intrusion Detection
Retail Data Analytics

The main purpose of data mining techniques in retail


industry is to understand markets and customers.

Data mining can also be used to optimise sales


operation ( For example, choosing the right mix of
products and optimal price)

Having insights about markets and customers could


enhance competitive advantages.
Retail Data Analytics

Understanding customers is also known


as customer relationship management.

Customer relationship management identifies


five different stages or phases of customers:
1. Prospects
2. Responders
3. New customers
4. Established customers
5. Former customers.
Data Mining for Customer Acquisition

Customer Acquisition is the process of attracting


prospects and turning them into new customers.

Data mining techniques can be applied for response


modelling to predict how likely a customer responds to
different channels such as email, direct mail and
telemarketing.

Response modelling helps retailers to perform targeted


marketing by focusing only on those prospects who are
more likely to respond.
Decision Tree for Response Modelling

What is decision tree?

Decision tree is a supervised learning method used for


classification and regression.

The goal is to create a model that predicts the value of a


target variable by learning simple decision rules inferred
from the data features.
Decision Tree for Response Modelling

The first step to apply any data mining technique is to


prepare a data set.

The following is a sample of data set used to build


response model using decision tree data mining
technique.
Decision Tree for Response Modelling

A sample data set.

The first column is the target variable that shows whether the
prospect customer responds or not to an email marketing
campaign.

The other columns are features which contains “age”, “lifestyle”,


“zipcode”, “familystatus”, “car_value”, “sports_type” and
“earnings per year”.
Decision Tree for Response Modelling

Output of Decision tree:

We can use this


response model to
predict whether a
given prospect
customer will respond
to the email
marketing campaign
Data Warehouse Architecture

1.Operational data sources


2.Data staging area
3.Data presentation area
4.Data access tools

You might also like