Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

USING DATA MINING FOR BANK DIRECT MARKETING

No description
by

LI Icy

on 12 December 2013

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of USING DATA MINING FOR BANK DIRECT MARKETING

USING DATA MINING FOR BANK DIRECT MARKETING
Data Understanding
Data Mining
Pattern Evaluation
The purpose of the predictive model is applying it to new data.
To evaluate the regression model, I use the model to test 2nd data set -
selected.csv
with
60
examples, randomly selected from full.csv.
Compare the computing result with the realistic result.
Summary
The best model achieved a high predictive performance.
Apply the best DM model in a real setting, with a tighter interaction with marketing managers, in order to gain a valuable feedback and improve bank marketing campaigns.
Data Set Information
Two data sets:
1)
full.csv
with all
45211
examples, ordered by date (05/2008-11/2010).
2)
selected.csv
with
60
examples, randomly selected from full.csv.

Objectives:
Build models to predict if the client will subscribe a term deposit.
Choose the best model to anticipate
and maximize profit.
Name: LI Bingjie
SID: 1155036208

Related with direct marketing campaigns of a Portuguese banking institution.
The marketing campaigns were based on phone calls.
# about basic information of the client:
1 - 4: age(n) ; job; marital ; education
5 - default: has credit in default?
6 - balance: average yearly balance(n)
7 - housing: has housing loan?
8 - loan: has personal loan?
# related with the last contact of the current campaign:
9 - 12: contact ; day(n) ; month ; duration(n)
# about campaigns before:
13 - 16: campaign(n) ; pdays(n) ; previous(n) ; poutcome

Output variable (desired target):
17 - y - has the client subscribed a term deposit? (binary: 'yes','no')
Attribute Information
Business background
Promotion approaches:
1) Mass campaigns
responses are typically very low, less than 1%.

2) Directed marketing
focus on targets that assumable will be interested, more attractive due to its efficiency.
Data Preparation
• Data cleaning:
No missing value.

• Data reduction and transformation:
Stepwise Logistic Regression Model
result ->
Classification and Prediction
– Finding models that describe and distinguish classes or concepts for future prediction.

Trend and evolution analysis

Modeling:
Decision-tree;
Regression.
<- Decision-tree Model
result
Assessing the Models
In terms of profit, regression line is always above tree line. Regression model performs better than tree model to maximize the profit.
model assessment measure
= Average Profit
Evaluation
Among 60 examples, 58 have been correctly predicted, just 2 deviations.
Accuracy rate = 96.7%
Accept the regression model.
set
X: EP_Y_ - contains the expected profit values
Y: D_Y_ - assigns either the achieve or miss decision status to a client in the score data set.
Allocate 60% of the input data to the training data set and 40% of the input data to the validation data set.
original variables
transformed variables
(mostly Maximize Normality)
age ( 5 buckets)
selected variables
Profit Matrix
achieve a client who finally subscribes a term deposit. - earn 5 profit
achieve a client who finally doesn't subscribe a term deposit. - loss 1 profit
Full transcript