Transcript Data Mining in Bank

Data Mining in Banking
CS548 Xiufeng Chen
Sources

K. Chitra, B.Subashini, Customer Retention in Banking Sector using Predictive Data
Mining Technique, International Conference on Information Technology, Alzaytoonah
University, Amman, Jordan, www.zuj.edu.jo/conferences/icit11/paperlist/Papers/

Dr. B. Subashini Data Mining Techniques and its Applications in Banking Sector. Website:
www.ijetae.com

Boris Kovalerchuk, Evgenii Vityaev, DATA MINING FOR FINANCIAL APPLICATIONS Petra
Hunziker, Andreas Maier, Alex Nippe, Markus Tresch, Douglas Weers, and Peter Zemp,
Data Mining at a major bank: Lessons from a large marketing application
http://homepage.sunrise.ch/homepage/pzemp/info/pkdd98.pdf

Rene T. Domingo, APPLYING DATA MINING TO BANKING
http://www.rtdonline.com/BMA/BSM/4.html

Predicting Returns from the Use of Data Mining to Support CRM
http://insight.nau.edu/downloads/CRM%20Mining%20Returns%20Paper.pdf
Purposes of Data Mining in Banking

As banking competition becomes more and more global and intense, banks
have to fight more creatively and proactively to gain or even maintain market
shares.

1. Discover new customers.
Clustering different customers into some clusters.

2. Remain customers. Especially the VIP customers.
In general, 20% of customers bring 80% of revenues. Using
association rules can find association between services.

3. Risk Management.
Using decision tree to classify high risk people.
Bank of America

Bank of America identified savings of $4.8 million in
two years (a 400 percent return on investment)
from use of data mining analytics. (source: Bank of
America)

This analyzing method was used to allow Bank of
America to detect fraud and find eligible lowincome and minority customers to ensure B of A’s
compliance with the Fair Housing Act.
source: Bank of America
Flow of data mining technique
Problem Understanding
Data Understanding
Data Filtering
System Modeling
System Evaluation
Analyzing Results
Source: Customer Retention in Banking Sector using Predictive Data Mining
Technique
Preprocessing the data

Customer relationship
management (CRM):

is a strategy that can
help bank to build longlasting relationships with
their customers and
increase their revenues
and profits.
Source: Predicting Returns from the Use of Data Mining to Support
CRM
CRM
Source: Predicting Returns from the Use of Data Mining to Support
CRM
Discover new customers
 k-Means: k-Means is a distance-based clustering algorithm that
partitions the data into a predetermined number of clusters. Each cluster has a
centroid (center of gravity). Cases (individuals within the population) that are in
a cluster are close to the centroid. For example, segment customer profession
data into clusters and rank the probability that an individual will belong to a
given cluster, and give them banking services they might need.
Remain the number of customers

1) measurement of customer retention;

2) identification of root causes of defection and related key service issues;

3) development of corrective action to improve retention.
 Apriori: Apriori performs market basket analysis by discovering co-occurring
items (frequent itemsets) within a set. For example, find the items or attributes
which comes from the lost customers and specify their association rules.
Therefore, the bank can take much care of those customers.
Risk Management

In this approach, risk levels are organized into categories based on past
default history.

Decision Tree technique can be used to build models that can predict default
risk levels of new loan applications.

1. Credit Cards
2. Deposits – Savings A/C

3. Internet Banking
4. Housing Loans

5. Term Loans
6. Cheque / Demand Drafts

7. Cash Transactions
8. Cash Credit A/c(Types of Overdraft A/C]

9. Advances
10. ATM / Debit Cards
Conclusion

Data Mining techniques are very useful to the banking sector for

(1) better targeting and acquiring new customers,

(2) most valuable customer retention,

(3) automatic credit approval which is used for fraud prevention, fraud
detection in real time,

(4) providing segment based products,

(5) analysis of the customers,

(6) transaction patterns over time for better retention and relationship,

(7) risk management and marketing.
The End
Xiufeng Chen