Applications of Data Mining in

levelsordΔιαχείριση Δεδομένων

20 Νοε 2013 (πριν από 3 χρόνια και 8 μήνες)

85 εμφανίσεις

1



Applications of Data Mining in
Banking


Maria Luisa Barja (Maria.Barja@ubs.com)

Jesús Cerquides (Jesus.Cerquides@ubs.com)

Ubilab IT Laboratory

UBS AG

Zurich, Switzerland










5/11/98


©1998 M.L.Barja, J.Cerquides Ubilab IT Laboratory, UBS AG




2

Outline


Data Mining in Banking


Application Areas


Pitfalls in the Development of Data Mining Projects


An Alternative: A Data Mining Framework


Open Projects


Summary

5/11/98


©1998 M.L.Barja, J.Cerquides Ubilab IT Laboratory, UBS AG




3

Data Mining in Banking


Banks have many and huge databases


Valuable business information can be extracted
from these data stores


Unfeasible to support analysis and decision
making using traditional query languages


Human analysis breaks down with volume and
dimensionality


Traditional statistical methods do not scale and
require significant analysis expertise

5/11/98


©1998 M.L.Barja, J.Cerquides Ubilab IT Laboratory, UBS AG




4

Application Areas

Four main areas


Marketing


Credit Risk


Operational Risk


Data Cleansing

5/11/98


©1998 M.L.Barja, J.Cerquides Ubilab IT Laboratory, UBS AG




5

Applications: Marketing

Objective:

Improve marketing techniques and target customers

Traditional applications:


Customer segmentation


Identify most likely respondents based on previous campaigns


Cross selling



Develop profile of profitable customers for a product


Predictive life cycle management:


Develop profile of profitable customers X years ago


Attrition analysis:


Alert in case of deviation from normal behaviour






Techniques
:

5/11/98


©1998 M.L.Barja, J.Cerquides Ubilab IT Laboratory, UBS AG




6

Applications: Credit Risk

Objective:

Reduce risk in credit portfolio

Traditional applications:



Default prediction


Reduce loan loses by predicting bad loans


High risk detection


Tune loan parameters ( e. g. interest rates, fees) in order to maximize profits


Profile of highly profitable loans


Understand characteristics of most profitable mortgage loans

5/11/98


©1998 M.L.Barja, J.Cerquides Ubilab IT Laboratory, UBS AG




8

Applications: Data Cleansing

Objective:

Detect outliers, duplicates, missing values,...

Traditional applications:


Data quality control


Detect data values which do not follow the pattern


Missing values prediction


Predict values of fields based on previous values

5/11/98


©1998 M.L.Barja, J.Cerquides Ubilab IT Laboratory, UBS AG




9

Pitfalls in the Development of
Data Mining Projects


Data Mining is a process, not a package!


Expensive, difficult to justify in first instance


Having substantial parts in common, most data
mining projects provide
custom solutions

that:


Are more expensive


Take more time to develop


Have a higher risk of not being finished


Ideally, use more than one technique to get a full
view of the data


5/11/98


©1998 M.L.Barja, J.Cerquides Ubilab IT Laboratory, UBS AG




10

Proposed Alternative


Identify the common functionality used for
the development of data mining solutions


Implement and pack this functionality in a
way that it can be:


Reused in many projects.


Customized to meet the needs of each project.


Extended, so it grows with its usage.

5/11/98


©1998 M.L.Barja, J.Cerquides Ubilab IT Laboratory, UBS AG




11

Object Oriented Frameworks


A framework is a reusable, “semi
-
complete”
application that can be specialized to
produce custom applications.

Framework

Ensemble

Framework design expertise

Programming language expertise

OO expertise

Domain expertise

OO expertise

Programming language expertise

Framework usage expertise

Coding expertise

5/11/98


©1998 M.L.Barja, J.Cerquides Ubilab IT Laboratory, UBS AG




12

Data Mining Framework:
Benefits


Reduces design and development efforts for building
concrete applications.


Lowers threshold for “proof of concept” data mining
applications to be developed.


Allows comparison of results across various methods.


Facilitates selection of best method(s) for particular
domains and business objectives.


Eases extensibility to new types of methods and
algorithms.




5/11/98


©1998 M.L.Barja, J.Cerquides Ubilab IT Laboratory, UBS AG




13

Data Mining Framework:

General Architecture

Project Management

Technique

implementation

Component structure

5/11/98


©1998 M.L.Barja, J.Cerquides Ubilab IT Laboratory, UBS AG




14

Data Mining Framework:

Component Structure

Project Management

Technique

implementation

Data

Process

Visualization

Metadata

Component

5/11/98


©1998 M.L.Barja, J.Cerquides Ubilab IT Laboratory, UBS AG




15

Data Mining Framework: Method
Implementation

Database Access

Data Understanding

Data Preparation

Modeling

Learning Data

Project Management

Component structure

5/11/98


©1998 M.L.Barja, J.Cerquides Ubilab IT Laboratory, UBS AG




16

Data Mining Framework:
Modeling


Classification

Clustering

Regression

Prediction

Description

Learning Data

Modeling roles

5/11/98


©1998 M.L.Barja, J.Cerquides Ubilab IT Laboratory, UBS AG




17

Data Mining Framework:

Open Projects


Design and development of:


A graphical user interface.


The prediction/description component (based
on bayesian networks).


The clustering component.


The project management component.


The preprocessing component.

5/11/98


©1998 M.L.Barja, J.Cerquides Ubilab IT Laboratory, UBS AG




18

Summary


Data Mining has emerged as an strategic
technology for a large bank


Several business areas where it can be
applied


Application development difficulties


Proposed a solution based on OO
framework technology