Credit Card Fraud Prediction

Running head: Credit card fraud prediction
Predicting the credit card fraud
Name of Student
Name of the University
Author note

1CREDIT CARD FRAUD PREDICTION
Abstract
In current days all the financial industries are facing potential challenges while
performing all type of transaction, which are fraudulent. However, the credit card fraud is
vital among all. Therefore, detecting the credit card fraud is necessary. This report intended
to discuss the detection of the same and check whether predictive model using big data will
beneficial for the detection or not. Numerous methods such as data mining, real time data
driven approaches and the one class classification techniques are majorly used for predicting
and detecting the hazardous theft activity in the credit card transactions. This article provides
an overview of different recent approaches for determining the fraud in the credit card
transaction.

2CREDIT CARD FRAUD PREDICTION
Table of Contents
Introduction................................................................................................................................3
Literature review........................................................................................................................3
Syntax matrix.........................................................................................................................3
Methodology..............................................................................................................................6
Conclusion..................................................................................................................................7
Reference....................................................................................................................................8

3CREDIT CARD FRAUD PREDICTION
Introduction
Fraud in the credit card mainly occurs when the customer give their credit card
number to any unfamiliar individuals, when their cards get stolen or lost, when any mail
diverted from the intended recipient has taken by the criminals. This type of fraud has
become a worldwide problem in current days (Pawar, Kalavadekar & Tambe, 2014).
Numerous fraudsters are widely developing different analytics for affecting the actual
working procedure of the credit card. Due to the enhancement in the internet technology, the
application of the credit card has dramatically increased with the increasing credit card fraud
(Dal Pozzolo et al., 2015). A single class approach for classification can be resorted in the big
data paradigm. Along with that, a well-developed hybrid architecture of the Particle Swarm
Optimization with a neutral network can be used. Moreover, one framework can also be
developed that will be beneficial in analytical interfacing with Hadoop and that will to read
the data in an efficient manner by providing efficient fraud detection.
Literature review
Syntax matrix
Article Main idea A Main idea B Main idea C
Credit card fraud
detection using big
data analytics: Use
of psoaann based
one-class
classification.
The vital intention of
this article is to
adopt the one class
classification in the
paradigm of big data
for detecting the
credit card fraud
This article focussed
on implementation
of a hybrid
architecture of the
Auto-associative
Neural Network as
well as Optimization
In the OCC, there
exist a series of
samples that are
accessible to a
specific class,
whereas the samples
for other classes are

4CREDIT CARD FRAUD PREDICTION
(Kamaruddin &
Ravi, 2016). In
recent years, the
banking and finance
are being the major
victim of the cyber
frauds. However,
this particular
adoption can be
beneficial in
combating the
fraudsters and
creating a secure
transaction
environment.
of the Particle
Swarm, for
explaining the nature
and working
methodology of the
one-class
classification. There
exist different
methods of
classification such
as, multiple, binary
and One-Class
Classification
(OCC). However,
among all, the OCC
is the most efficient
one in detecting the
credit card fraud.
either less or
completely absent.
Apart from the
particle swarm
optimization, Auto
associative neutral
network and the
PSOAANN
architecture, the
machine learning
methodology can
also be used to
enhance the fraud
detection process
while undertaking
credit card
transactions.
Real time data-
driven approaches
for credit card fraud
detection
The main idea
behind this article is
to purpose real time
data-driven
approaches by
utilizing the optimal
anomaly detection
The secondary idea
of the article
explains the
necessity of fraud
detection and
explaining the two
important real time
The one class
support vector
mechanism
(OCSVM) with the
help of the selection
of the optimal kernel
is beneficial in

5CREDIT CARD FRAUD PREDICTION
methods for
detecting the credit
card fraud. Credit
card fraud is
responsible for
major financial
losses for both the
customer as well as
the organizations.
The real time data
driven techniques
will amplify the
detection process by
delivering minimal
percentage of the
false alarm rate.
data driven
approaches for
identifying the fault
(Van et al., 2015).
Fraudulent activities
will lead to several
major potential
challenges in
carrying out the
transactions by
creating higher risk
and cost loss.
Therefore, the credit
card fraud detection
has received an
attention. The fraud
detection methods
must be flexible
So that it can cope
up with the
continuous evolution
of fraud over the
time. Two major
data driven
approaches such as
separating the data
from its origin
Via searching one
hyper plane. In order
to separate the data
and to detect the
fraud it uses several
algorithm. On the
other hand, the main
intention of the one-
class classification
approach and the
control charts are
almost similar as in
both the cases only a
single class is being
shown in the training
data that is helpful in
explaining its
characteristics and
providing ways for
identifying abnormal
behaviour while
performing credit
card transactions.

6CREDIT CARD FRAUD PREDICTION
OCSVM and control
chart can be used in
detecting the
unwanted activity in
the transactions that
occur using the
credit card.
A novel approach
for automated credit
card transaction
fraud detection
using network-based
extensions
This article tends to
deliver an effective
review of the credit
card fraud detection
happen in the
financial industry
based of the
applications of the
data mining
technique (Sharma
& Panigrahi, 2013).
A major upsurge of
the fraud has been
experienced in the
current scenario
need to get immense
attention.
The vast failure of
particular internal
systems of auditing
of the organizations
in case of identifying
the accounting
frauds is looking
forward to the
different procedures
that are especially
designed for
identifying all the
accounting frauds,
which is known as
the forensic
accounting.
A series of different
techniques related to
the data mining
provides a large aid
in detecting the
financial frauds as
dealing with wide
amount of data
creates several
complexity and
potential challenge
in the forensic
accounting.

7CREDIT CARD FRAUD PREDICTION
Methodology
This study majorly relies on the necessary primary and all the secondary sources.
Primary data is gathered from different sources for example interviews and any kind of
survey. On the other hand, the secondary sources will include the information that are
collected from the authenticated appropriate internet resources.
Conclusion
Therefore, from the above, it can be concluded that, the different predictive model
through the use of the big data are vital in detecting the credit card fraud. The PSOAAN
based approach of the one-class classification is one of the important approach in identifying
the unethical credit card activities. This report has explained the use of the predictive
modelling for credit card fraud detection by using data analytics. Furthermore, the above
study has also explained that, the fraud detection approach of automated transaction can be
carried out by using several data mining and cloud approach. Apart from that, the two vital
approaches of the real time data driven approaches have also elaborated in this report to
identify the unauthorized and unethical credit card activities.

8CREDIT CARD FRAUD PREDICTION
Reference
Dal Pozzolo, A., Boracchi, G., Caelen, O., Alippi, C., & Bontempi, G. (2015, July). Credit
card fraud detection and concept-drift adaptation with delayed supervised
information. In 2015 international joint conference on Neural networks (IJCNN) (pp.
1-8). IEEE.
Kamaruddin, S., & Ravi, V. (2016, August). Credit card fraud detection using big data
analytics: Use of psoaann based one-class classification. In Proceedings of the
International Conference on Informatics and Analytics (p. 33). ACM.
Pawar, A. D., Kalavadekar, P. N., & Tambe, S. N. (2014). A survey on outlier detection
techniques for credit card fraud detection. IOSR Journal of Computer Engineering,
16(2), 44-48.
Sharma, A., & Panigrahi, P. K. (2013). A review of financial accounting fraud detection
based on data mining techniques. arXiv preprint arXiv:1309.3944.
Tran, P. H., Tran, K. P., Huong, T. T., Heuchenne, C., HienTran, P., & Le, T. M. H. (2018,
February). Real time data-driven approaches for credit card fraud detection. In
Proceedings of the 2018 International Conference on E-Business and Applications
(pp. 6-9). ACM.
Van Vlasselaer, V., Bravo, C., Caelen, O., Eliassi-Rad, T., Akoglu, L., Snoeck, M., &
Baesens, B. (2015). APATE: A novel approach for automated credit card transaction
fraud detection using network-based extensions. Decision Support Systems, 75, 38-48.

9CREDIT CARD FRAUD PREDICTION