Electronic Banking Fraud Detection Using Data Mining Techniques Report

This report, submitted by a student for ITECH 5500, focuses on electronic banking fraud detection using various data mining techniques. The study explores the challenges of online banking security, including the congestion and insecurity of internet communication. The research delves into the application of unsupervised machine learning methods, such as principal component analysis and K-mean clustering, for outlier detection. The report emphasizes the importance of integrated systems for enhanced performance and the need for clustering-based classification models. The synthesis matrix summarizes key articles, highlighting the effectiveness of data mining in detecting sophisticated online banking fraud, the application of decision tree models, and the use of data mining in risk management within the banking sector. Furthermore, the report discusses the application of data mining in customer relationship management and the use of sequence pattern mining algorithms in e-banking services. The references include articles on data mining techniques in the banking sector, fraud detection, and risk management. The report aims to provide a comprehensive overview of how data mining can be utilized to improve fraud detection and enhance security within the electronic banking environment.

Running head: ELECTRONIC BANKING FRAUD DETECTION USING DATA MINING TECHNIQUE
ELECTROONIC BANKING FRAUD DETECTION TECHNIQUE USING DATA MINING
TECHNIQUES
Name of the Student
Name of the University
Author Note

ELECTRONIC BANKING FRAUD DETECTION USING DATA MINING TECHNIQUES 1
Proposal:
In spite of the challenging economy, the utilization of e channel related platforms
such as ATM, Internet Banking, Web, POS has continued for experiencing the significant
growth. The major problems for that are, the communication that has been done over internet
in not secure as well as it is often congested, the finance related institution should also have
to content with the challenges related to internet including the quality of services, insecurity
as well as some aberrations in the electronic banking. The research will deal with the
processes for the computing to the presence of outliers by utilizing difference measures of
distance as well as natural performance related to detection for the machine learning that are
unsupervised like principle component analysis and K-mean Clustering Analysis. For a
comprehensive evaluation of the techniques related to data mining, predictive modelling and
machine learning for the algorithms of Anomaly Detection on the transaction dataset of
electronic banking. It can be concluded that the integrated technique systems will be able to
provide better performance efficiency than the systems that are singular. Along with that the
clustering based is required for the classification model.
Synthesis matrix:
Article Main idea 1 Main idea 2 Main idea 3
Effective detection
of sophisticated
online banking
fraud on extremely
imbalanced data
An online banking
fraud detection
system will be a
typical use case of
the Web of things
methodology that
In the paper the
authors have
summarise the
primary
characteristics of the
frauds related to
The authors have
implemented a
system of risk
management of
online banking. The
system integrates

2ELECTRONIC BANKING FRAUD DETECTION USING DATA MINING TECHNIQUES
are broad based. The
online banking fraud
detection requires to
be instant as this is
so much difficult to
recover the loss if
during the period of
detection, a fraud
will remain as
undetected.
sector of banking.
As most of the
works that are
published are related
to the computer
intrusion, domain of
fraud related to
credit card. The
authors have
discussed each of the
limitations for better
detection of online
fraud in the banking
sector.
several features as
the models that are
related to data
mining as well as
aims for
consolidating
various sources of
the resources for
solving the
problems.
Data Mining
Techniques and its
Applications in
Banking Sector
The authors have
used decision tree
model for solving
the prediction
problems and
classification where
the examples are
classified in the
classes. Usually the
decision trees are
becoming so large as
The authors have
discussed about the
value prediction
methods instead of
classification of loan
application that are
new. This attempts
for predicting the
default amount that
are expected for the
new loan
Most of the common
methods for the data
mining that are used
for the customer
profiling are,
clustering,
regression and
classification and
lastly the association
rule discovery.

3ELECTRONIC BANKING FRAUD DETECTION USING DATA MINING TECHNIQUES
the procedure is
involving the
removing and
identifying the
branches which
contain the rate of
error.
application. The
values that are
predicted are
numeric as well as it
needs the techniques
which are able to
take the numerical
data or the variables
that are predicted.
A REVIEW ON
DATA MINING IN
BANKING
SECTOR
In their paper the
authors have
discussed about risk
management in data
mining for
identifying the risk
related factors in
each of the
departments of the
sector of banking.
The credit related
risks can be
quantified by the
alterations in the
value of the credit
products or the
As said by the
authors the data
mining will be
useful in all of the
three phases of the
relationship cycle of
the customer. Within
the CRM context,
the data mining will
be seen as the
business-driven
procedure that aimed
at the consistent
utilization of the
profitable
knowledge and
The data mining is
able to improve the
electronic marketing
and telemarketing by
the identification of
the potential
consumers who are
adhere to the
technologies that are
modern such as
smartphone, internet.
In the areas of e
banking as well as
the other internet
services that are
used for banking

4ELECTRONIC BANKING FRAUD DETECTION USING DATA MINING TECHNIQUES
credit consumer
portfolios that are
based in the
alteration in the
tendency of high
risk, instrument
rating, recovery rate
and default
probability.
discovery from the
data of the
organisation.
sector are able to use
another algorithm
that are known as
sequence pattern
mining algorithm.
The algorithm can
be used by the
management for
finding associations
between various
items or events in
the data.

5ELECTRONIC BANKING FRAUD DETECTION USING DATA MINING TECHNIQUES
References:
Chitra, K., & Subashini, B. (2013). Data mining techniques and its applications in banking
sector. International Journal of Emerging Technology and Advanced
Engineering, 3(8), 219-226.
Jayasree, V., & Balan, R. V. S. (2013). A review on data mining in banking sector. American
Journal of Applied Sciences, 10(10), 1160.
Patel, R. D., & Singh, D. K. (2013). Credit card fraud detection & prevention of fraud using
genetic algorithm. International Journal of Soft Computing and Engineering, 2(6),
292-294.
Wei, W., Li, J., Cao, L., Ou, Y., & Chen, J. (2013). Effective detection of sophisticated
online banking fraud on extremely imbalanced data. World Wide Web, 16(4), 449-
475.