www.ijecs.

in
International Journal Of Engineering And Computer Science ISSN: 2319-7242
Volume 5 Issue 4 April 2016, Page No. 16183-16186

Credit Card Fraud Detection System

Suchita Anant Padvekar, Pragati Madan Kangane, Komal Vikas Jadhav

Atharva College of Engineering

Department of Computer Science
University of Mumbai
Mumbai, India
Email: jadhav.komal.4 @gmail.com
Email: [email protected]
Email: [email protected]
ABSTRACT
The most accepted payment mode is credit card for both online and offline in today’s world, it provides
cashless shopping at every shop in all countries. It will be the most convenient way to do online shopping,
paying bills etc. Hence, risks of fraud transaction using credit card has also been increasing. In the existing
credit card fraud detection business processing system, fraudulent transaction will be detected after
transaction is done. It is difficult to find out fraudulent and regarding loses will be barred by issuing
authorities. Hidden Markov Model is the statistical tools for engineer and scientists to solve various
problems. In this paper, it is shown that credit card fraud can be detected using Hidden Markov Model
during transactions. Hidden Markov Model helps to obtain a high fraud coverage combined with a low false
alarm rate.
existing purchase data of cardholder is a
1.1 INTRODUCTION promising way to reduce the rate of successful
2 In day to day life credit cards are used for credit card frauds. Since humans tend to
purchasing goods and services with the help of exhibit specific behaviorist profiles, every
virtual card for online transaction or physical cardholder can be represented by a set of
card for offline transaction. In a physical-card patterns containing information about the
based purchase, the cardholder presents his typical purchase category, the time since the
card physically to a merchant for making a last purchase, the amount of money spent, etc.
payment. To carry out fraudulent transactions Deviation from such patterns is a potential
in this kind of purchase, an attacker has to threat to the system
steal the credit card. If the cardholder does not
realize the loss of card, it can lead to a 1.2 BASIC CONCEPT
substantial financial loss to the credit card In the existing credit card fraud detection business
company. In online payment mode, attackers processing system, fraudulent transaction will be
need only little information for doing detected after transaction is done. It is difficult to
fraudulent transaction (secure code, card find out fraudulent and regarding loses will be
number, expiration date etc.). In this purchase barred by issuing authorities. Hidden Markov
method, mainly transactions will be done Model is the statistical tools for engineer and
through Internet or telephone. To commit scientists to solve various problems. In this paper,
fraud in these types of purchases, a fraudster it is shown that creditcard fraud can be detected
simply needs to know the card details. Most of using Hidden Markov Model during transactions.
the time, the genuine cardholder is not aware Hidden Markov Model helps to obtain a high
that someone else has seen or stolen his card fraud coverage combined with a low false alarm
information. The only way to detect this kind rate.
of fraud is to analyze the spending patterns on
every card and to figure out any inconsistency
with respect to the “usual” spending patterns. 1.3 APPLICATIONS
Fraud detection based on the analysis of

Suchita Anant Padvekar,IJECS Volume 05 Issue 4 April 2016 Page No.16183-16186 Page 16183
 DOI: 10.18535/Ijecs/v5i4.22

 The detection of the fraud use of the card 2. System for project
is foundmuch faster than the existing
The information form will be arise which has set
system.
of questions where the user has to answer that
 The Transactions of the account holder question correctly to do the transaction.These
never stoppedAs this system allows the form have informationlikePersonal,Professional
user to use the virtual cardUsing the virtual address,Dob,etc.are available in database.If user
ID and password,untill he gets thenew entered information will be matched with the
card. database information then the transaction will be
done securely.Else user transaction will bbe
 The user can easily block the card by terminated and transfer to online shopping
himself when hefinds that the card is being website.
stollen.
 In this system we have used the ONE
TIMEPASSWORD for the security to get
the virtual ID and Password securely.
 We can find the most accurate detection
using thistechnique.

1.4 Performance

 . It determines the length of time system

used by the system to process data. This
test is conducted prior to implementation
to transaction determine how long it takes
to get a response to an inquiry, make a
backup copy of a file, or send a
transmission and get a response.
Peak workload performance
Fig 1.0
 It determines whether the system will
handle the volume of activities that occur 3.1 Technologies Used:
when the system is at the peak of its MICROSOFT VISUAL STUDIO
processing demand. For example, test the
system by activating all terminals at the C# (MS Visual Studio 2010)
same time.
About Visual Studio:-
1.5 Maintainability
The Microsoft Visual Studio development system
is a suite of development tools designed to aid
 Maximizes the correct prediction and
maintain the incorrect prediction at an software developers—whether they are novices or
acceptable level seasoned professionals—face complex challenges
1.6 Environmental and create innovative solutions. Every day,
software developers break through tough
 The credit card fraud detection system shall problems to create software that makes a
not cause physical harm to users and non- difference in the lives of others. Visual Studio's
users. role is to improve the process of development to
 The credit card fraud detection system shall make the work of achieving those breakthroughs
not cause interference to external systems or easier and more satisfying.
any type of misbehave.
How Visual Studio improves the process of
development:
Suchita Anant Padvekar,IJECS Volume 05 Issue 4 April 2016 Page No.16183-16186 Page 16184
 DOI: 10.18535/Ijecs/v5i4.22

Productive:- today’s increasingly complex business

applications. SQL Server 2005 allows companies
Visual Studio-branded tools continually deliver
to gain greater insight from their business
better ways for software developers to do more
information and achieve faster results for a
with less energy wasted on repetition and
competitive advantage.
drudgery. From efficient code editors,
IntelliSense, Wizards, and multiple coding Key Capabilities:-
languages in one integrated development •High Availability: Ensure business continuity
environment (IDE) to high-end application life- with the highest levels of system availability
cycle management (ALM) products in Microsoft® through technologies that protect your data
Visual Studio® Team System. New versions of against costly human errors and minimize
Visual Studio keep bringing innovative tools to disaster recovery downtime.
help developers focus on solving problems, not •Performance and Scalability: Deliver an
waste time on minutiae. infrastructure that can grow with your business
and has a proven record in handling today's large
Integrated:-
amounts of data and most critical enterprise
With Visual Studio, software developers benefit workloads.
from an integrated product experience that spans •Security: Provide a secure environment to
tools, servers, and services. Visual Studio address privacy and compliance requirements
products work well together—not just with one with built-in features that protect your data
another, but also with other Microsoft software, against unauthorized access.
such as Microsoft server products and the
Microsoft Office system. •Manageability: Manage your infrastructure with
automated diagnostics, tuning, and configuration
Comprehensive:- to reduce operational costs while reducing
Visual Studio offers a choice of tools for all maintenance and easily managing very large
phases of software development—development, amounts of data.
testing, deployment, integration, and
management—and for every kind of developer— •Developer Productivity: Build and deploy critical
from the novice to the skilled professional. Visual business-ready applications more quickly by
Studio is also engineered to support development improving developer productivity and reducing
across all types of devices—PCs, servers, the project life cycle times.
Web, and mobile devices.
•Business Intelligence: Gain deeper insight into
Reliable your business with integrated, comprehensive
Visual Studio is engineered and tested to be analysis and reporting for enhanced decision
consistently dependable, secure, interoperable, making.
and compatible. Visual Studio offers an
unmatched combination of security features, 3.SystemRequirement
scalability, and interoperability. Although Visual
3.1 Minimum Hardware Interface:
Studio always incorporates forward-thinking
features, it is designed to ensure backward-  Interface between the software product and
compatibility wherever possible. hardware components
 The device types supported would be:
About SQL Server:-  600Mhz processor
Microsoft SQL Server 2005 is a comprehensive,  192 MB RAM
integrated data management and analysis software
that enables organizations to reliably manage
 Basic Webcam (Inbuilt Or External)
mission-critical information and confidently run
 Standard keyboard

Suchita Anant Padvekar,IJECS Volume 05 Issue 4 April 2016 Page No.16183-16186 Page 16185
 DOI: 10.18535/Ijecs/v5i4.22

 Microsoft mouse and compatible pointing  Video 800×600,256 color s

device

3.2 Minimum Software Interface:-The software required is

 Microsoft windows server 2003

 Microsoft windows server 2005
 Microsoft SQL server 2005

4. Conclusion providing us with this valuable opportunity of

pursuing such a challenging project.
We have proposed an application of HMM in
credit card fraud detection. The different steps in 6. Refrences
credit card transaction processing are represented
as the underlying stochastic process of an HMM. [1] Ghosh, S., and Reilly, D.L., 1994. Credit Card
We have used the ranges of transaction amount as Fraud Detection with a Neural-Network, 27th
the observation symbols, whereas the types of Hawaii International l Conference on Information
Systems, vol. 3 (2003), pp. 621- 630.
item have been considered to be states of the
[2] Syeda, M., Zhang, Y. Q., and Pan, Y., 2002
HMM. We have suggested a method for finding Parallel Granular Networks for Fast Credit Card
the spending profile of cardholders, as well as Fraud Detection, Proceedings of IEEE
application of this knowledge in deciding the International Conference on Fuzzy Systems, pp.
value of observation symbols and initial estimate 572-577 (2002).
of the model parameters. It has also been [3] Stolfo, S. J., Fan, D. W., Lee, W.,
explained how the HMM can detect whether an Prodromidis, A., and Chan, P. K., 2000. Cost-
incoming transaction is fraudulent or not. Based Modeling for Fraud and Intrusion
Comparative studies reveal that the Accuracy of Detection: Results from the JAM Project,
the system is close to 80 percent over a wide Proceedings of DARPA Information Survivability
variation in the input data. The system is also Conference and Exposition, vol. 2 (2000), pp.
scalable for handling large volumes of 130-144.
[4] Aleskerov, E., Freisleben, B., and Rao, B.,
transactions..
1997. CARDWATCH: A Neural Network Based
Database Mining System for Credit Card Fraud
5. Acknowledgement Detection, Proceedings of IEEE/IAFE:
Computational Intelligence for Financial Eng.
We would like to express our deep gratitude (1997), pp. 220-226.
towards all the people who have selflessly helped [5] M.J. Kim and T.S. Kim, “A Neural Classifier
us in our project. We would like to thank with Fraud Density Map for Effective Credit Card
everyone for always having faith in us and Fraud Detection,” Proc. Int’l Conf. Intelligent
Data Eng. and Automated Learning, pp. 378-383,
2002.

Suchita Anant Padvekar,IJECS Volume 05 Issue 4 April 2016 Page No.16183-16186 Page 16186