Professional Documents
Culture Documents
Abstract— as the rapid growth of electronic commerce technologies, the use of credit cards gets increased. As the
Credit Cards become the popular mode to pay the payment for regular purchase and online purchase, cases of Credit
Card fraud also get increased. Financial fraud gets increasing similarly with the development of the modern
technologies and the global superhighway of the communication, which results in the loss of money/economy. The
financial fraud (fraudulent transactions) is scattered with the genuine transactions and with simple pattern matching
techniques which are not often sufficient to find the frauds accurately. The fraud detection systems are implemented
for all credit cards providing banks to minimize their losses. Many techniques such as artificial intelligence, neural
networks, k-nearest neighbour algorithm, artificial immune system, data mining, fuzzy logic, decision tree, marching,
learning, support vector machine, genetic algorithms etc., are evolved in detecting the various fraudulent transactions
with credit cards. Our paper presents a survey of various types of techniques which are used in the credit card fraud
detection.
Keywords— Credit card Fraud Detection method, Credit card fraud, Electronic Commerce.
I. INTRODUCTION
Data Mining has grown as a field of basic and employed research in computer science in general and e-
commerce in detail. In today’s electronic society, e-commerce has become a necessity sales carry for global business.
Due to the rapid progress of e-commerce and use of credit cards for the purchases has dramatically enhanced.
Unfortunately, the fraudulent use of the credit cards has also become an attractive source of the revenue for the criminals.
Occurrence of the credit card fraud is the increasing dramatically because of vulnerability of security weaknesses in the
traditional credit card process systems, leading to the loss of billions of dollars every year. [1]
Fraud detection is known as illegal activities in commercial and industrial establishments such as banks,
insurance companies, and cell phone companies and so on (Chandola et al., 2009). Fraud is a dangerous ethical trouble
with credit card companies in banking systems. The main purposes of credit card fraud detection are to detect credit card
fraud transactions and cut down the losses on account of such illegal activities. [3]
Credit card fraud is a variable term for theft and fraud committed using a credit card as a fraudulent source of
the funds in a transaction. The aim may be receiving goods without giving, or to obtain the unauthorized funds from an
account. The Credit Card Fraud, one of the largest menaces of business establishments today. However, fighting the
fraud in effect, it is crucial to first realize the mechanisms of executing a fraud. Credit card fraudsters hire a prominent
count of modus operandi to devote fraud. In simple terms, Credit Card Fraud is defined as:
When an individual users uses another individual. The credit card for personal reasons, while the owner of the
card and the card issuer are not aware of the fact that the card constitutes Used. Further, then someone using the card has
no connection with the cardholder or issuer, and has no purpose of either reaching the owner of the card or making
repayments for the purchases made.
Credit card frauds are devoted in the next ways:
An act of criminal deception (mislead with intent) by use of unauthorized account
And/or personal information
Illegal or unauthorized usage of account for personal benefit
Misrepresentation of account information to obtain goods and/or services. [7]
There are lots ways to devote credit card fraud. Fraudsters are very inventive, quick going people. Primarily
there are two common classes of credit card fraud:
1) Traditional Techniques
2) Modern Techniques
© 2014, IJARCSSE All Rights Reserved Page | 849
The first type of the credit card fraud to be named by this report is Application Fraud, where the individual will
spoof an application to develop a credit card. Application fraud can divide into assumed identity, where an individual
guesses to be someone else; and financial fraud, where someone feeds false information about his or her financial
position to get credit. This probe then becomes on to look at Intercept Fraud, where a card is enforced for legitimately,
but is stolen from the mail service before it arrives at its final destination. There is also the illegal habit of Lost and
Stolen Cards, which causes up a substantial area of credit card fraud. On that point are more sophisticated credit card
fraudsters, beginning with those who develop Fake and Doctored Cards, there are as well those who apply Skimming to
devote fraud. This is where the data applied on either the magnetic strip on the back of the credit card, or the information
stored on the smart chip is imitated from one card to some other. Site Cloning and False Merchant Sites on the Internet
are getting a popular method of fraud for lots any criminals with a skilled ability for hacking. Such sites are planned to
acquire people to hand over their credit card items without making they have been scammed. Triangulation is also a new
development.
Triangulation is when a merchant bids a product at a very cheap cost of a website. When a customer looks
forward to buy the product the merchant tells customer to give via email once the item is delivered. The merchant applies
a fraudulent card number to buy the product from a Web site and ships the product to the consumer, who then ship the
merchant his or her credit card items via email. The merchant becomes on running in this way utilizing the credit card
numbers that have been sent from the consumers to buy products, expecting for a abruptly time to be a legitimate
merchant ahead he or she closes the Web site and starts a new one. Their fields substantially the more advanced
fraudsters, who use Credit Card Generators; computer emulation software that creates valid credit card numbers and
expiry dates. These authors are extremely reliable at arriving at valid credit card details and are available for free
download off the internet. Making them available to lots someone who run fraudulent operations. [5]
Credit-card purchases can be classified into 2 types:
1) Virtual card and,
2) Physical card.
The physical-card based purchase; the cardholder gives his card physically to a merchant for arriving at a
payment. To convey out fraudulent deals in this form of purchase, an attacker has to stealthy credit card. If the cardholder
does not make the loss of the card, it can lead to a significant financial loss for Credit Card Company. In the second form
of purchase, only some crucial information around a card (card number, expiration date, security code) is called in to
build the payment. These purchases are commonly caused on the Internet. To denote the fraud in various types of
purchases, a fraudster merely needs to know the card items. Most of the time, genuine cardholder where not aware that
somebody else has considered or stolen his card data. The only path to notice this form of fraud is to examine the
expenditure patterns on all cards and to figure out the, any inconsistency with the respect to “usual” expenditure patterns.
Fraud-detection is based on an analysis of the existing purchase date of the cardholder is promising way to
decrease the rate of the successful credit card frauds. Since humans incline to show particular behaviouristic profiles, all
cardholder can be constituted by a set of patterns holding information about the distinctive purchase category, the time
ago the final purchase, the amount of money spent. [9]
The Credit card fraud, divided into 2 types: The offline fraud and the online fraud. Offline fraud is devotedly
using a stolen physical card at call canter or storefront. In almost, the institution publishing the card, cans locket earlier, it
is used in a fraudulent mode. Online fraud is the committed via web, and phone shopping (or) the cardholder not
introduce. Only the card‘s details are wanted, and a manual signature and card imprint were not expected at the time of
purchase. [5]
2. Blast-Ssaha Hybridization
Blast -Ssaha Hybridization [25] is explained as hybridization of Blast and SSAHA algorithms. These algorithms
are basically the efficient and two-stage sequence alignment way of algorithm which is used for the analysing spending
behaviour of the customers. The algorithm performance is good, and also accuracy is high. This is very useful in the
telecommunication and the banking fraud detection with high processing speed. [26]
Merits: accuracy, high performance, and high speed.
Demerits: This algorithm does not detect the cloning of the credit cards.
4. Genetic Algorithm
Genetic algorithm was first introduced by Holland (1975). The main aim of genetic algorithm was obtained
better solution as the time progresses. The fraud detection problem is one of the classifications of problem, in which
some of the statistical methods with of many data-mining algorithms that have been proposed to solve it. The decision
tree is the most popular among all other methods. The fraud detection is usually used in the domain of banking, data
mining, and E-commerce.[27]
Genetic algorithms [28] are the evolutionary algorithms; aim at obtaining the better solutions as time progresses.
Since their first introduction by Holland, they have been successfully applied to many problem domains from astronomy
to sports, from optimization to computer science, etc. They have also been used in the data mining mainly for the
variable selection and mostly coupled with the other data mining algorithms. In this study, we try to solve our
classification problem by using only a genetic algorithm solution.
6. Decision Trees
In this study, the performance of classifier models built by using the well-known decision tree methods C5.0,
C&RT and CHAID. A decision tree is a tree structure which attempts to separate the given records into mutually
exclusive subgroups. To do this, starting from the root node, each node is split into child nodes in a binary or a multi split
fashion related to the method used based on the value of the attribute (input variable) which separates the given records
best. Records in a node is recursively separated into child nodes until there is no split that makes statistical difference on
the distribution of the records in the node or the number of records in a node is too small. Each decision tree method uses
its own splitting algorithms and splitting metrics. [22][31]
VIII. CONCLUSION
Currently the credit card risk monitor system is one of the key factor for the merchants banks, the organization to
improve the merchants risk managements level in automatic, adequate and scientific way. We have many ways of
detection of the credit card fraud detection. If one of the method or the combination of the algorithms is applied for the
credit card fraud-detection system and the probability of the fraud transactions can be predicted soon after the credit card
transactions is carried by the banks. A series of anti fraud methods and strategies can be adopted for the prevention of
bank from great losses and reduces the risks. In this paper we had given contribution of various effective ways of the
credit card fraudulent detection.
REFERENCES
[1] Amlan Kundu ,Suvasini Panigrahi , A.K. Majumdar, Shamik Sural , “The Credit card fraud detection: A fusion
approach using Dempster–Shafer theoryand Bayesian learning”, International journal on Information Fusion 10
(2009)
[2] Kavita Rawat ,Jyoti Hazrati, “Credit Card Fraud Detection Using HiddenMarkov Model”, International Journal of
Latest Research in Science and Technology Vol.1,Issue 4,November-December (2012)
[3] Maryamsadat Hejazi, Yashwant Prasad Singh, “The Credit Card Fraud-Detection Using the Kernel Methods with
Support Vector Machine”, Journal of Advanced Computer Science and Technology Research 2 (2012)
[4] Chiyuan LI, ZhiqiangYAO,“The Validation of Credit Card Number on Wired and Wireless Internet”, Journal Of
Networks, Vol. 6, No. 3, March 2011
[5] Venkata Ratnam Ganji, Siva Naga Prasad Mannem, “Credit card fraud detection using anti-k nearest neighbor
algorithm”, (IJCSE) ISSN : 0975-3397 Vol. 4 No. 06 June 2012
[6] Manoel Fernando Alonso Gadi, Alair Pereira do Lago, Xidi Wang, “A comparasion of classification methods applied
on credit card fraud detection”, Article presented to Brazil, Sao Paulo, april 2008
[7] Tej Paul Bhatla, Vikram Prabhu & Amit Dua,“Understanding Credit Card Frauds”, Cards Business Review#2003–01
June 2003
[8] Clifton Phua,Ross Gayler, Vincent Lee, Kate Smith-Miles, “On the communal analysis suspicion scoring for identity
crime in streaming credit applications”, European Journal of Operational Research 195 (2009)
[9] Abhinav Srivastava, Amlan Kundu, Shamik Sural, “Credit Card Fraud-Detection Using the Hidden Markov Model”
[10] Dr. R.Dhanapal, V.Dheepa, “The Analysis of Credit Card Fraud Detection”, International Journal of Recent Trends
in the Engineering, Volume 2, No. 3, November 2009
Bibliography
Ms. Rajeswari. K was born on September 20th, 1989, in Pondicherry, India. She received the
B.Tech degree in Information Technology from the Christ College of Engineering and Technology,
Pondicherry University and she is currently pursuing the first year M.Tech under the stream of
Networking from Sri Manakula Vinayagar Engineering college , Pondicherry University.
Ms. Durga.M as born on February 5th 1991, in Pondicherry, India. She received the B. Tech degree
in Information Technology from the Indira Gandhi College of Engineering and Technology for
women, Anna University and she is currently pursuing the final year M.Tech under the stream of
Networking from Sri Manakula Vinayagar Engineering college , Pondicherry University.
Mr. Sakthivel Punniyakodi was born on March 16th 1991, in Pondicherry, India. He received the
B. Tech degree in Information Technology from the Christ College of Engineering and
Technology, Pondicherry University and he is currently pursuing the first year M.Tech under the
stream of Computer Science from Sri Manakula Vinayagar Engineering college , Pondicherry
University.
Mr. Venkatesh @ Saravanan.R was born on 17th June on 1989, in Karaikal, Pondicherry, India. He
received the Diploma in Computer science from the Karaikal Polytechnic college, Pondicherry
university in 2007,then B.Tech degree in computer science from the Bharathiyar College of
engineering and technology, karaikal, Pondicherry university,2010 ,he was worked in Iwiz Hitec
Solutions (P) Ltd, Chennai as JAVA Programming in Software development around 2years and he
is currently doing his Final year M.Tech in Networking from Sri Manakula Vinayagar Engineering
college , Pondicherry University.