Prediction of Credit Card Fraud Research Question 2022

Running head: PREDICTION OF CREDIT CARD FRAUD
PREDICTION OF CREDIT CARD FRAUD
Name of the Student
Name of the Organization
Author Note

1
PREDICTION OF CREDIT CARD FRAUD
Abstract
Banking and all the financial industries are seen to be facing a number of several huge
challenges in the specific form of a number of transactions which are fraudulent. It has been
observed that the fraud of credit card is one of them. Hence, for the main purpose of detecting
the credit card fraud detection it is to be checked if predictive model utilising big data will be
helping in credit card fraud detection or not. The specific study has a proper introduction at
first which will be outlining the thesis and after that the literature review will be comprising
of a synthesis matrix which will be able to demonstrate the ability of finding all the several
sources which possess with the familiarity of the existing knowledge body. Lastly there is a
conclusion which will be answering the research question or rather the defined thesis.

2
PREDICTION OF CREDIT CARD FRAUD
Table of Contents
Introduction................................................................................................................................3
Literature Review.......................................................................................................................3
Methodology..............................................................................................................................8
Conclusion..................................................................................................................................8
References..................................................................................................................................9
Bibliography.............................................................................................................................10

3
PREDICTION OF CREDIT CARD FRAUD
Introduction
A large number of fraud cases of the credit card as well as the net banking on the
internet are considered to be as the largest international problem within the specific domain
of the banking. Particularly in the year 2014, it has been observed that the worldwide fraud
has been fully accounted for a great loss of nearly about sixteen dollars and this specific
figure has been observed to be hugely increasing day by day. This particularly because all the
several fraudsters are hugely developing a large number of new analytics for changing the
actual normal behaviour of the specific working of the detection system of the fraud of the
credit cards. It is to be understood if there can be a development of any kind of predictive
model for specially detecting the credit card frauds. A one-class approach of classification
can be resorted in the specific paradigm of big data. An architecture (hybrid) of the Particle
Swarm Optimization and also the Neural Network which is Auto-Associative can also be
utilised. Due to the huge advancement in the technology of internet, the utilisation of the
several credit cards has been seen to have dramatically incremented and it has finally lead to
an increment in the number of several frauds of credit card. A framework can also be
developed which will be particularly analytical of interface with the Hadoop which will be
able to read the data or rather information in a much efficient manner and it can also provide
the specific server which is analytical or the fraud detection.
Literature Review
Synthesis Matrix
Articles Main Idea A Main Idea B Main Idea C Main Idea D
Credit card
fraud detection
using big data
analytics: Use
The main aim
of the paper is
to resort to one-
class approach
The study will
be presenting
the particular
implementation
In particularly
the OCC one,
there are a
number of
The study
mostly shows
the employment
of OCC for the

4
PREDICTION OF CREDIT CARD FRAUD
of psoaann
based one-class
classification
of classification
in the particular
paradigm of the
big data. It has
been observed
that all the
several financial
as well as the
banking
industries are
hugely facing a
number of very
much severe
challenges in
the specific
form different
cyber frauds.
of the hybrid
architecture of
both the Auto-
associative
Neural Network
and the
Optimization of
the Particle
Swarm, which
will be
proposed
elsewhere, for
one-class
classification in
the specific
computational
structure of
Spark. It has
been known that
classification
can be easily
performed in a
number of
different ways
like the binary,
multi or rather
the One-Class
Classification
(OCC) one.
several samples
which are seen
to be accessible
for one-specific
class, whereas a
number of
samples for
some other
classes are
either very
much few in
number or
rather totally
absent. Hence,
whenever
anyone will be
coming across
either the
regular or rather
the normal
samples in huge
abundance and
it has been
noticed that the
samples of the
interest class are
very much
scarce, there
can be the
employment of
the particular
one-class
approach of the
classification
for detecting the
credit card
frauds and some
other rare
events as well.
fraud detection
of the credit
card in the
particular
structure of the
big data. Big
data is known to
be well
characterized by
utilizing 4V’s
viz., the
velocity,
veracity and the
volume.
Predictive
Modelling For
Credit Card
Fraud Detection
Using Data
This particular
paper mainly
aims at the
analytical
framework of
The specific
sector which is
the banking and
the finance
sector is
Such systems
are known to be
very much
vulnerable with
a huge number
As data is seen
to be
incrementing in
the particular
terms of Peta

5
PREDICTION OF CREDIT CARD FRAUD
Analytics the Big Data for
processing a
huge volume of
data. It also
comprises of the
implementation
of several
algorithms for
the detection of
fraud and the
performance of
them have been
well observed
on the specific
dataset of the
benchmark for
detecting all the
frauds on the
basis of real
time and this
will be
providing very
much low risk
and much
higher
satisfaction of
the customer.
considered to be
very much
essential in the
generation of
the most current
days where
almost all of the
huge number of
human have to
hugely deal
with the bank
both on the
internet or
rather
physically. Both
the profitability
as well as the
productivity of
the private as
well as the
public sectors
has hugely
incremented in
a tremendous
way due to the
one and only
information
system of the
banking. In the
current days, it
has been
observed that
most of the
transactions of
the application
systems of E-
commerce are
properly done
with the help of
credit card as
well as the net
banking system.
of full new
attacks as well
as techniques at
a specific rate
of alarming.
Hence, the
detection of the
frauds in the
banking is
considered to be
one of the most
important or
rather the most
vital aspects in
the current days
because finance
is considered to
be a huge or
rather major
sector of the
life.
Bytes and
therefore for the
improvement of
the performance
of the server
which is
analytical in the
building of the
model, anyone
can be
possessing with
an interface
analytical
structure with
that of Hadoop
which can be
reading data in
an efficient
manner and also
can be
providing it to
the particular
analytical server
for the
prediction of the
several kinds of
fraud of credit
cards.
A novel
approach for
This specific
paper is seen to
While it has
been noticed
All these
network based
The results of
the study will

6
PREDICTION OF CREDIT CARD FRAUD
automated
credit card
transaction
fraud detection
using network-
based
extensions
be actually
proposing
APATE which
is basically a
novel approach
for detecting all
the several
transactions of
the credit cards
which are very
much fraudulent
and this is
mostly
conducted in all
the several
stores on the
internet. In the
last few decade,
the huge and the
specific ease of
the payment
upon the
internet has
been seen to be
opening up a
number of
several full new
opportunities
for the e-
commerce and
thereby
decrementing
all the several
boundaries for
retailing which
are
geographical.
that e-
commerce has
been gaining
popularity to a
great extent, it
is also seen to
be the huge
playground of a
number of
several
fraudsters as
well who are
mostly trying to
totally misuse
the specific
transparency of
all the
purchases done
online and the
transferring of
the records of
the credit card
as well. The
approach is
known to be
comprising of a
number of
several features
which are
intrinsic and
these are
particularly
derived from all
the several
characteristics
of all the
transactions
which are
incoming and
also the
particular
history of the
spending of the
customers
features are
seen to be
hugely utilized
by directly
exploiting the
specific
network of all
the several
holders of the
credit card and
merchants and
this will be
further deriving
a suspiciousness
score which will
be time
dependent of
each and every
object of the
network.
be particularly
showing that all
of the different
intrinsic as well
as the features
based upon the
network are
very much
strong sides
which are
intertwined of
the totally
similar kind of
picture. The
specific
combination of
such two kinds
of few features
will be leading
to the specific
best models of
performance
which will be
reaching all the
AUC scores far
and far higher
than that of
0.98.

7
PREDICTION OF CREDIT CARD FRAUD
utilising several
fundamentals of
the Regency-
Frequency-
Monetary or
PFM and all the
several features
based upon the
network.
A review of
financial
accounting
fraud detection
based on data
mining
techniques.
This specific
study will be
particularly
delivering a
particular
review of the
fraud detection
of the financial
accounting
which will be
based of the
techniques of
data mining. It
has been
observed that
with a huge
upsurge of the
financial
accounting
fraud in the
recent scenario
which has been
experienced, it
has been known
that FAFD or
rather financial
accounting
fraud detection
has totally
become a much
emerging
concern of the
huge
importance for
several
The huge failure
of the specific
internal system
of auditing of
any
organization in
the
identification of
several
accounting
frauds has been
leading towards
the utilisation of
all the several
procedures
which are
specialised for
detecting all the
financial
accounting
frauds and this
is known as the
forensic
accounting.
A number of
several
techniques of
data mining are
seen to be well
providing a
huge aid in the
detection of the
financial
accounting
frauds as
dealing with
huge volumes
of data as well
as a number of
several
complexities of
data related to
finance are very
much huge
challenges for
the specific
accounting of
the forensic.
This paper has
well presented a
study upon the
application of
all the several
techniques of
data mining for
detecting all the
frauds of
financial
accounting. A
framework has
also been
proposed for the
techniques of
data mining
based upon the
specific
detection of the
fraud which will
be accounting.

8
PREDICTION OF CREDIT CARD FRAUD
industries,
research as well
as academics.
Data mining
techniques and
its applications
in banking
sector
This specific
study has its
main aim on the
several data
mining
techniques and
all of the
applications of
it in the sector
of banking like
the detection
and the
prevention of
fraud, retention
of the customer,
marketing and
the management
of risk.
It has been well
said in this
specific study
that data mining
has become an
essential area
strategically for
a number of
organizations of
business
involving the
sector of
banking. It is
basically a
procedure of
properly
analysing the
data from
different kinds
of perspectives
and finally
summarizing it
into an
information
which will be
very much
valuable.
Data mining
will be assisting
all of the banks
in directly
looking for all
the patterns
which are
hidden in a
specific group
and then
discover a kind
of unknown
relation within
the data. All the
techniques of
early data
analysis have
been oriented
towards the
extraction of
both the
characteristics
of data which
will be both
quantitative as
well as
statistical. Such
kinds of
techniques will
be facilitating
very much
useful
interpretations
of data for the
particular sector
of banking for
avoiding the
attrition of the
customers and
also frauds as
Fraud has been
a significant
problem in the
sector of
banking. Hence,
this study
actually
specialises upon
all the
techniques of
data mining and
all of the
applications of
it within the
sector of
banking like
that of the
prevention and
the detection of
frauds.

9
PREDICTION OF CREDIT CARD FRAUD
well.
Application of
credit card fraud
detection:
Based on
bagging
ensemble
classifier
The study has
its main aim at
the evaluation
of the
performance
which has been
well performed
on the real life
transactions of
the credit card
for exactly
demonstrating
the particular
benefit of the
bagging
ensemble
algorithm.
Currently, it has
been well
observed that
credit card fraud
has been
increasing very
much
considerably
with that of the
development of
the fully
modern
technology and
the worldwide
superhighways
of sharing
information.
All the various
frauds of credit
cards highly
costs the
consumers and
the company of
finance a huge
billions of
dollar annually.
All the
fraudsters
continually try a
lot in searching
full new rules as
well as tactics
for committing
several actions
which will be
illegal. Hence,
the systems of
fraud detection
have become
very much
important for all
the banks as
well as the
institutions of
finance for
lowering all of
their losses.
The techniques
which have
been utilised for
the detection of
frauds are the
Support Vector
Machines, K-
Nearest
Neighbour
algorithms and
the NB. Such
kinds of
techniques can
be utilised alone
or rather in the
collaboration
utilising several
techniques of
ensemble or
meta-learning
for building
several
classifiers.
This study
specifically says
that among all
the methods
which have
been existing,
ensemble
methods of
learning have
been well
identified as a
popular
procedure. This
is not only
because of the
fact that it is an
implementation
which is quite
straightforward,
but also because
of its fully
exceptional
performance
which is highly
predictive on

10
PREDICTION OF CREDIT CARD FRAUD
various
practical
problems.
Hence, this
particular paper
has basically
trained a
number of
techniques of
data mining
utilised in the
detection of
credit card
fraud.
Effective
detection of
sophisticated
online banking
fraud on
extremely
imbalanced data
The study is
mostly based
upon the
particular
effective
detection of the
fully
sophisticated
fraud of online
banking on
greatly
imbalanced
data. It has been
known that
sophisticated
fraud of online
banking highly
However, it has
been well
known that
there is a very
much limited
data or
information
which has been
available for
distinguishing
the fraud which
is dynamic from
the customer
behaviour
which will be
genuine in such
a kind of data
This will be
actually making
the effective
detection much
more
challenging as
well as
essential. This
particular paper
will be
comprising of a
proposal for an
effective
framework of
the detection of
the fraud of
online banking
There has been
a specific kind
of algorithm
which has been
introduced
named the
ContrastMiner.
It has been
basically
introduced for

11
PREDICTION OF CREDIT CARD FRAUD
reflects the
integrative kind
of several
resources in all
the social as
well as the
physical worlds.
Its particular
detection is
basically a kind
of typical
utilisation case
of the widely-
based
methodology of
Wisdom Web of
Things.
environment
which is highly
sparse as well
as totally
imbalanced.
that will be
synthesizing all
the various
relevant
resources. It
will also be
involving
various
advanced
techniques of
data mining. By
specifically
building a
particular
contrast vector
for each and
every
transaction
based upon the
historical
sequence of the
behaviour of the
customers, there
will be a
profiling of the
rate of
differentiating
of each of the
recent
transaction
against the
preference of
the customer’s
behaviour.
an efficiently
mining various
patterns of
contrast and
also for
distinguishing
fraudulent from
the behaviour
which will be
genuine
followed by a
kind of effective
selection of
pattern as well
as scoring of
risk which will
be combining
several
predictions
from different
kinds of
models. Several
results from the
experiments on
the huge scale

12
PREDICTION OF CREDIT CARD FRAUD
real online data
of banking have
been clearly
demonstrating
that the system
can be highly
achieving
higher accuracy
and lower
volume of alert
than that of the
latest
benchmarking
detection
system of fraud
involving the
knowledge of
domain and all
the detection
methods of the
traditional
frauds.

13
PREDICTION OF CREDIT CARD FRAUD
Methodology
The particular study will be highly dependent upon both the primary as well as all the
secondary sources as well. Primary data is known to be obtained from all the several sources
like that of several interviews as well as any kind of survey. All the secondary sources are
seen to be involving different kinds of information which will be properly collected from the
internet resources.
Conclusion
It can be concluded that there can be a predictive model utilising big data analytics
can be developed. There are a number of models or rather frameworks which can be
developed for protecting from several frauds of the credit cards. There can be an utilisation of
the PSOAAN based one-class classification and some models as well. The study has been
able to provide a brief description of the predictive modelling for the detection of the fraud of
credit card by utilising data analytics and has proved that a predictive model can be
developed for the purpose of the fraud detection of the credit cards. There has also been a
novel approach for the fraud detection of the automated transaction of the card by utilising all
the several extensions based upon the cloud. Lastly, there is also a presence of the financial
accounting fraud detection which has been upon the several techniques of data mining.

14
PREDICTION OF CREDIT CARD FRAUD
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.
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.
Patil, S., Nemade, V., & Soni, P. K. (2018). Predictive Modelling For Credit Card Fraud
Detection Using Data Analytics. Procedia computer science, 132, 385-395.
Sharma, A., & Panigrahi, P. K. (2013). A review of financial accounting fraud detection
based on data mining techniques. arXiv preprint arXiv:1309.3944.
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.
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.
Zareapoor, M., & Shamsolmoali, P. (2015). Application of credit card fraud detection: Based
on bagging ensemble classifier. Procedia computer science, 48(2015), 679-685.

15
PREDICTION OF CREDIT CARD FRAUD
Bibliography
Patil, S., Nemade, V., & Soni, P. K. (2018). Predictive Modelling For Credit Card Fraud
Detection Using Data Analytics. Procedia computer science, 132, 385-395.
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.
Sharma, A., & Panigrahi, P. K. (2013). A review of financial accounting fraud detection
based on data mining techniques. arXiv preprint arXiv:1309.3944.
Zareapoor, M., & Shamsolmoali, P. (2015). Application of credit card fraud detection: Based
on bagging ensemble classifier. Procedia computer science, 48(2015), 679-685.
Veeramachaneni, K., Arnaldo, I., Korrapati, V., Bassias, C., & Li, K. (2016, April). AI^ 2:
training a big data machine to defend. In 2016 IEEE 2nd International Conference on
Big Data Security on Cloud (BigDataSecurity), IEEE International Conference on
High Performance and Smart Computing (HPSC), and IEEE International
Conference on Intelligent Data and Security (IDS) (pp. 49-54). IEEE.
Rushin, G., Stancil, C., Sun, M., Adams, S., & Beling, P. (2017, April). Horse race analysis
in credit card fraud—deep learning, logistic regression, and Gradient Boosted Tree.
In 2017 Systems and Information Engineering Design Symposium (SIEDS) (pp. 117-
121). IEEE.
Bahnsen, A. C., Aouada, D., Stojanovic, A., & Ottersten, B. (2016). Feature engineering
strategies for credit card fraud detection. Expert Systems with Applications, 51, 134-
142.

16
PREDICTION OF CREDIT CARD FRAUD
Baesens, B., Van Vlasselaer, V., & Verbeke, W. (2015). Fraud analytics using descriptive,
predictive, and social network techniques: a guide to data science for fraud detection.
John Wiley & Sons.
Albashrawi, M., & Lowell, M. (2016). Detecting financial fraud using data mining
techniques: a decade review from 2004 to 2015. Journal of Data Science, 14(3), 553-
569.
Baesens, B., Bapna, R., Marsden, J. R., Vanthienen, J., & Zhao, J. L. (2016).
Transformational Issues of Big Data and Analytics in Networked Business. MIS
quarterly, 40(4).
Jordan, M. I., & Mitchell, T. M. (2015). Machine learning: Trends, perspectives, and
prospects. Science, 349(6245), 255-260.
Malini, N., & Pushpa, M. (2017, February). Analysis on credit card fraud identification
techniques based on KNN and outlier detection. In 2017 Third International
Conference on Advances in Electrical, Electronics, Information, Communication and
Bio-Informatics (AEEICB) (pp. 255-258). IEEE.
Bello-Orgaz, G., Jung, J. J., & Camacho, D. (2016). Social big data: Recent achievements
and new challenges. Information Fusion, 28, 45-59.