In recent days most of the small and large
originations have moved their daily businesses to online and provide the
services to customers using internet. Credit Card (CC) fraud is one of the
major issues in online transaction. In recent year CC fraud or no card frauds
are increased in day to day activity. The main reason is most of the customers
are using CC for their all kind of payment. So the aim of this paper is to
identify the different types of CC fraud and to review the alternative
techniques to detect the CC frauds. Therefore secured transaction is required
for CC holders when consuming their CC to make electronic payment for purchasing
goods. This study different types of fraudsters that commit CC fraud and the
type of techniques used by these cyber fraudsters to commit fraud on the
internet is discussed.
Card Fraud, Credit Card Fraudsters, K-means
CC fraud can be defined as the fraudulent use of a credit card
account through the theft of the account holder’s card number, card details and
personal information, through a wide variety of methods in order to perform
illegal transactions from the compromised account. Fraud can also define as
criminal activity or adjunct to identity theft. Due to the expansion of modern
technology, the payment mode of individual has been changed significantly. The
use of Online payment mode such as Online Banking, Debit Card, Credit Card etc.
Now a day tremendous volume and value increase in credit card transactions. CC
fraud begins with the theft of the physical card or with the compromise of data
linked with the account, including the card account number or other details
that would regularly and essentially be available to a merchant during a
genuine transaction. Fraud detection and software that analyzes the patterns or
blue print and unusual behaviour as well as individual transactions in order to
flag likely fraud. Profiles include such information as IP address.
MINING: AN OVERVIEW
Data mining is defined as the
extraction of hidden predictive information from large databases. In general,
data mining (sometimes called data or knowledge discovery) is the process of
analyzing data from different views and summarizing it into useful information
– information that can be used to increase profits, scores prices, or both.
Data mining software is one of a number of analytical tools for analyzing data.
It allows users to analyze data from different dimensions or angles or
categorize it, and summarize the relationships identified. Technically, data
mining is the process of finding correlations or patterns among dozens of
fields in large relational databases.
3.1 K-Means Clustering Algorithm
K-means clustering algorithm is
unsupervised liner clustering algorithm it solve the well known clustering
problem. K-Means clustering intends to partition n objects into k clusters in
which each object belongs to the cluster with the nearest mean. This method produce
exactly k different clusters of greatest possible characteristic. The best
number of clusters k leading to the greatest separation (distance) is not known
as a priori and should be compute from the data. The objective of K-Means
clustering is to minimize total intra-cluster variance, or, the squared error
Clusters the data into k groups
where k is predefined.
Select k points at random as
Assign objects to their closest
cluster center according to the Euclidean distance function.
Calculate the centroid or mean of
all objects in each cluster.
Repeat steps 2, 3 and 4 until the
same points are assigned to each cluster in consecutive rounds.
K-Means is relatively an professional
method. However, we need to specify the numeral of clusters, in advance and the
final results are sensitive to initialization and often terminates at a local
optimum. Unfortunately there is no worldwide hypothetical method to find the
optimal number of clusters. A practical approach is to compare the outcomes of
multiple runs with different k and choose the best one based on a predefined condition.
In general, a large k probably decrease the error but increase the risk of over
A Fraud Detection System (FDS)
runs at a credit card issuing bank. Each incoming transaction is submitted to
the FDS for proof. FDS receives the card details and validate the transaction
is genuine or not. If the FDS confirms the transaction to be hateful, it raises
an alarm, and the issuing bank decline the transaction. The concerned card
holder may then be contacted and alerted about the possibility that the card is
A customer is an individual or
business that purchases the goods or services produced by a business.
Attracting customers is the primary goal of most public-facing businesses,
because it is the customer who creates demand for goods and services. Since the evolution of the internet, many small and large companies have
moved their businesses to the internet to provide services to customers
worldwide. Now a day most of the customers are interested to purchase the goods
via online, so online customer are increasing rapidly.
A transaction is a business event
that has a monetary impact on an entity’s financial statements, and is recorded
as an entry in its accounting records. Examples of transactions are as follows:
Paying a supplier for services rendered or goods delivered. Online
transaction, also known as a PIN-debit transaction, is a password-protected
payment method that authorizes a transfer of funds over an electronic funds
transfer. When processed as an online transaction, the exchange
of funds is completed using an EFT network. Depending
on which EFT system your bank is associated with as a member bank. The cost of
the transaction typically amounts to an interchange fee which is charged to the
3.2.3 Acquiring Bank
An acquiring bank is a bank that
mainly works with businesses and organizations and is not focused on the
individual. When an organization wishes to store their money from a sale (is it
a church, non-profit, school or corporation), they would utilize an acquiring
bank. Within the acquiring bank, the business is required to set up a merchant
3.2.4 Merchant account
All merchants, be it
brick-and-mortar or e-commerce, are required to set up a merchant account if
they wish to accept a payment. All the funds from a merchant’s sales are held
for a merchant in the merchant account, and the liability of the merchant
account is underwritten by the acquiring bank. Merchants who are unable to pay
off debts in their merchant account due to excessive refunds or charge backs
will have those debts covered by the acquiring bank with whom they have their
3.2.5 Issuing Bank
A bank that issues accounts and
debit/credit cards directly to consumers, not businesses, is an issuing bank.
For credit cards, the issuing bank (such as Wells Fargo or Bank of America)
will set up a line of credit and be responsible to ensure that the cards debts
are paid off. Any debts on a card that is ultimately not paid off by the
cardholder will have to be paid off by the bank that issued the card.
3.2.6 Payment Gateway
To accept a payment online, a
merchant will need to set up a payment gateway account. A payment gateway is a
service that authorizes and transmits transaction data on behalf of the
merchant to the payment processor used by the merchant’s acquiring bank.
Think of the
payment gateway as the online equivalent of an in-store point-of-sale system.
Brick-and-mortar merchants don’t need a payment gateway, as in-store
transactions are handled by their point-of-sale system, which collects the sale
and passes the payment details to the payment processor of the merchant’s
For online purchases, a customer
typically enters their payment information into the shopping cart or online
checkout used by the merchant’s e-commerce platform, which forwards the payment
details to the payment gateway, who in turn communicates with the payment
processor of the merchant’s acquiring bank to authorize the transaction.
gateway will charge the merchant a small flat fee per order.
3.3.Essentials for Fraud Detection:
Fraud detection in real time
Anything less is a window for
the criminals to get away. Even for
batch processes, the scoring engine should evaluate the transaction, and an
authorization or decline result should take place prior to funds movement.
Analytics is the only way to
actually detect fraudulent patterns of behavior efficiently. The production of the analytics should
include a buyer score, which determines how the activity corresponds to the
customers’ actual behavior, and a transaction fraud score, which determine the
fraudulent nature of the operation.
Understanding customer behavior is essential—it helps reduce the impact
on your customers and your fraud operation by reducing fake positives. Certain customers transact in ways that may
Workflow is critically important to solving
challenges with fraud resources. In our
last fraud survey, it was obvious that monetary institution have many hoops to
work through to correct and manage each customer’s fraud issues. Specific actions, data and processes are
required to handle each type of fraud case and to use as verification for
prosecution. Depending on the type of
payment defrauded, there are specific steps required to make the customer
whole, and every step to back out the transaction creates working costs. A cohesive and supple workflow engine allows
analysts to consolidate and, in many cases, automate the remediation process.
Efficient rules engine
The rules engine links the analytic scoring
to an action based on the currently available information. Rules are essential to react quickly to shut
down fraud, and allow management and documents of the processes used to define
and refine your actions, in a repeatable and auditable manner. Business rule is
the simplest approach is to use expert knowledge to implement business
knowledge of fraudulent behavior as part of a computer-based expert system. In
the business rules involves the history, frequency, location, goods, phone
number, e-mail address, shipping address, merchant’s website and dollar amounts
of previous transactions for that credit card from its database records.
Based on credit card details, different type of transaction business rule
created in the rule engine module to find the fraud transaction.
In this paper, a data mining computer program has
been model as a subsystem which can be used with software systems and
applications in banks to detect credit-illegal dishonesty/stealing (by lying)
in a transaction on the internet. This Data mining (online or paper form that
asks for a job, money, admission, etc.) accepts input formatted on a pattern on
which a transaction is being did/done/completed and matches it with the
credit-card holder’s patterns of its credit-card online consumption it’s been
trained with to classify a (happening or viewable immediately, without any
delay) transaction as real and true, suspicious illegal dishonesty/stealing (by
lying) or illegal transaction. In the case of the suspicious illegal
dishonesty/stealing (by lying) classification, the bank using the system can
(ask lots of questions about/try to find the truth about) further by calling
the credit-card owner (related to/looking at/thinking about) the suspicious
fake/illegal (because of lying and stealing) transaction.
T. F. Lunt, A. Tamaru, F. Gilham, R. Jagannathan, P. G. Neumann, H. S.
Javitz, A. Valdes, and T.
D. Garvey. AReal-Time Intrusion Detection Expert System (IDES) – Final
Technical Report. Technical report, SRI Computer Science Laboratory, SRI International,
Menlo Park, CA, Feb. 1992.
2 Dr. Yashpal Singh and Singh Chauhan,
Neural networks in data mining. Journal of Theoretical and Applied Information Technology
(2005-2009), vol, 5, no. 6. pp. 37-42.
3 Khyati Chaudhary, Jyoti Yadav and
Bhawna Mallick, A review of fraud detection techniques: credit-card, International
Journal of Computer Applications (2012), vol. 45, no. I, pp.39-44
4 Cybercrime: protecting against the
growing threat Global Economic Crime Survey – PWC Global Economic. ONLINE.
Available at: http://www.pwc.com/en_GX/gx/economic-crime-survey/assets/GECS_GLOBAL_REPORT.pdf. Accessed 12
5 S. Rosset, U. Murad., E. Neumann, Y.
Idan, and G. Pinkas. Discovery of fraud rules for telecommunicationslchallenges
and solutions. In Proceedings of the fifth ACM
SlGKDDinternational conference on Knowledge
discovery and data mining, pages
409-413. ACM Press, 1999.
Jiawei Han, MichelineKamber; Data Mining: Concepts and Techniques 1
M. Halkidi, Y.Batistakis, M. Vazirgiannis; Clustering algorithms and validity
measures : 0-7695-1218-6/01 2001 IEEE2