Ijarcce 41

ISSN (Online) 2278-1021
ISSN (Print) 2319-5940
International Journal of Advanced Research in Computer and Communication Engineering

Vol. 4, Issue 3, March 2015
Credit Card Fraud Detection using Hidden

Markov Model
AashleshaBhingarde1, AvnishBangar2, Krutika Gupta3, SnigdhaKarambe4
Department of Information Technology P.V.P.P College of EngineeringSion, Mumbai1,2,3,4
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.
Keywords: Internet, online shopping, credit card, e-commerce security, fraud detection, Hidden Markov Model.
I. INTRODUCTION
In day to day life credit cards are used for purchasing visible to an external observer and therefore states are
goods and services with the help of virtual card for online ``hidden'' the outside; hence the name Hidden Markov
transaction or physical card for offline transaction. In a Model. Hence, Hidden Markov Model is a perfect solution
physical-card based purchase, the cardholder presents his for addressing detection of fraud transaction through credit
card physically to a merchant for making a payment. To card. One more important benefit of the HMM-based
carry out fraudulent transactions in this kind of purchase, approach is an extreme decrease in the number of False
an attacker has to steal the credit card. If the cardholder Positives transactions recognized as malicious by a fraud
does not realize the loss of card, it can lead to a substantial detection system even though they are really genuine. In
financial loss to the credit card company. In online this prediction process, HMM consider mainly three price
payment mode, attackers need only little information for value ranges such as:
doing fraudulent transaction (secure code, card number,
expiration date etc.). 1) Low (l),
In this purchase method, mainly transactions will be done 2) Medium (m) and,
through Internet or telephone. To commit fraud in these 3) High (h).
types of purchases, a fraudster simply needs to know the First, it will be required to find out transaction amount
card details. belongs to a particular category either it will be in low,
Most of the time, the genuine cardholder is not aware that medium, or high ranges.
someone else has seen or stolen his card information. The
only way to detect this kind of fraud is to analyze the III. APPLICATION DESCRIPTION
spending patterns on every card and to figure out any In existing models, the bank is verified credit card
inconsistency with respect to the “usual” spending information, CVV number, Date of expiry etc., but all
patterns. Fraud detection based on the analysis of existing these information are available on the card itself.
purchase data of cardholder is a promising way to reduce Nowadays, bank is also requesting to register your credit
the rate of successful credit card frauds. Since humans card for online secure password. In this new model, after
tend to exhibit specific behaviouristic profiles, every feeding details of card at merchant site, then it will transfer
cardholder can be represented by a set of patterns to a secure gateway which is established at bank’s own
containing information about the typical purchase server. But, it is not verifying that the transaction is
category, the time since the last purchase, the amount of fraudulent or not. If hackers will get secure code of credit
money spent, etc. Deviation from such patterns is a card by phishing sites or any other source, then it is very
potential threat to the system. difficult to trace fraudulent transaction. In proposed model
based on HMM will help to verify fraudulent of
II. HIDDEN MARKOV MODEL transaction during transaction will be going to happen. It
A Hidden Markov Model is a finite set of states; each state includes two modules are as follow
is linked with a probability distribution. Transitions among
these states are governed by set probabilities called IV. ONLINE SHOPPING
transition probabilities. In a particular state a possible It comprises with many steps, first is to login into a
outcome or observation can be generated which is particular site to purchase goods or services, then choose
associated symbol of observation of probability an item and next step is to go to payment mode where
distribution. It is only the outcome, not the state that is credit card information will be required. After filling all
Copyright to IJARCCE DOI 10.17148/IJARCCE.2015.4341 169
ISSN (Online) 2278-1021
ISSN (Print) 2319-5940
International Journal of Advanced Research in Computer and Communication Engineering

Vol. 4, Issue 3, March 2015
these information, now the page will be directed to

proposed fraud detection system which will be installed at
bank’s server or merchant site.
V. FRAUD DETECTION SYSTEM

All the information about credit card (Like Credit card
number, credit card CVV number, credit card Expiry
month and year, name on credit card etc.) will be checked
with credit card database. If User entered database is
correct then it will ask Personal Identity number (PIN).
After matching of Personal Identity number (PIN) with
database and account balance of user’s credit card is more
than the purchase amount, the fraud checking module will
be activated.
Fig(1):System Architecture
The verification of all data will be checked before the first
page load of credit card fraud detection system. If user VII. SUMMARY
credit card has less than 10 transactions then it will Established connection between the database and web
directly ask to provide personal information to do the application. We have effectively implemented the first
transaction. Once database of 10 transactions will be phase of our credit card fraud detection model. Admin
developed, then fraud detection system will start to work. login form is created successfully with the commencement
By using this observation, determine users spending of admin session and termination after logout
profile.
VIII. FUTURE SCOPE
The purchase amount will be checked with spending The future work on which we are focusing now is to
implement and measure the performance of our proposed
profile of user. By transition probabilistic calculation
based on HMM, it concludes whether the transaction is system so that we can justify that our proposed system is
real or fraud. If transaction may be concluded as better in credit card fraud detection then all the previous
fraudulent transaction then user must enter security proposed system.
information. This information is related with credit card
(like account number, security question and answer which REFERENCES
[1]. P. Chan and S. Stolfo, “Metalearning for Multistrategy and Parallel
are provided at the time of registration). Learning,” Proc. SecondInt’l Workshop Multistrategy Learning,
Center for Artificial Intelligence, George Mason Univ., Fairfax,Va., 1993, pp. 150–165.
If transaction will not be fraudulent then it will direct to [2]. S. Stolfo et al., “JAM: Java Agents for Metalearning over
Distributed Databases,” Proc.Third Int’l Conf. Knowledge Discovery
give permission for transaction. If the detected transaction
andData Mining, AAAI Press, Menlo Park, Calif., 1997, pp. 74–81.
is fraudulent then the Security information form will arise. [3]. D. Wolpert, “Stacked Generalization,” Neural Networks,Vol. 5,
It has a set of question where the user has to answer them 1992, pp. 241–259.
correctly to do the transaction. These forms have [4]. P. Chan and S. Stolfo, “Toward Scalable Learning with
Nonuniform Class and Cost Distributions: A Case Study in Credit
information such as personal, professional, address; dates Card Fraud Detection,” Proc. Fourth Int’l Conf.Knowledge
of birth, etc are available in the database. Discovery and Data Mining, AAAI Press, Menlo Park, Calif., 1998,
pp. 164–168.
If user entered information will be matched with database [5]. Lijun Cao, Xiyin Liu, TiejunZhou ,Zhongping Zhang Aiyong Liu;
Based on the flow of anti-k nearest neighbors algorithm for data
information, then transaction will be done securely. And mining outliers; In Proceedings of IC-BNMT2010
else user transaction will be terminated and transferred to [6]. Yufeng Kou, Chang-Tien Lu, SiriratSinvongwattana ,Yo-Ping
online shopping website. Huang; Survey of Fraud Detection Techniques; in Proceedings of
the 2004 IEEE International Conference on Networking, Sensing &
Control Taipei, Taiwan, March 21-23, 2004
VI. SYSTEM ARCHITECTURE [7]. Y. Dora Cai, David Clutter, Greg Pape, Jiawei Han. MAIDS;
The implemented architecture consists of two Mining alarming incidents from data streams; In SIGMOD,
subsystems: database interface and credit card fraud Paris,2004:919-920
(CCF) detection engine. The database interface subsystem [8]. Charu C. Aggarwal, Philip S. Yu; An effective and efficient
algorithm for high-dimensional outlier detection; In The
is the entry point through which the transactions are VLDBJournal (2005) 14: 211–221
read into the system. It is the system’s interface with [9]. [5] AleksandarLazarevic, Vipin Kumar; Feature Bagging for
the banking software. Visual Basic.Net was used for the Outlier Detection; In Proceedings of KDD’05, August 21–24, 2005,
design of CCF detection, that is, as a front-end while
Microsoft Access was used for the design of training
and test database, as back-end. In the CCF detection
subsystem, each transaction entering into the system was
passed to the host server where the corresponding
transaction profile is further checked using neural
networks and transactions business rules.
Copyright to IJARCCE DOI 10.17148/IJARCCE.2015.4341 170

Ijarcce 41

Uploaded by

Copyright:

Available Formats

Ijarcce 41

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ijarcce 41

Uploaded by

Copyright:

Available Formats

ISSN (Online) 2278-1021

ISSN (Print) 2319-5940

International Journal of Advanced Research in Computer and Communication Engineering

Credit Card Fraud Detection using Hidden

International Journal of Advanced Research in Computer and Communication Engineering

these information, now the page will be directed to

V. FRAUD DETECTION SYSTEM

You might also like