Download Credit Card Fraud Detection Based on User Profile and Previous

Volume : 2 | Issue : 3 | March 2013
ISSN - 2250-1991
Research Paper
Engineering
Credit Card Fraud Detection Based on User
Profile and Previous Transaction
* Nitika Kadam, **Suryakant Soni, ***Devendra
Puntambekar, ****Rahul Kaul
* Computer Science Engineering, Mandsaur Institute of Technology, Indore
** Computer Science Engineering, Mandsaur Institute of Technology ,Indore
*** Information Technology, Mandsaur Institute of Technology, Indore
**** Computer Science Engineering, BM College,Indore
ABSTRACT
This paper reports on the findings of a research project that had the objective to prevent the customer from the fraud of credit
card. In today scenario we see that most of the people uses transaction through credit card either virtually or physically. 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. As credit card becomes the
most popular mode of payment for both online as well as regular purchase, cases of fraud associated with it are also rising. In
this paper, the concept of data mining and Hidden Markov Model is used to detect the credit card fraud during transactions.
Hidden Markov Model is the statistical tools for engineer and scientists to solve various problems. An Hidden Markov Model
is initially trained with the normal behavior of a customer who hold the credit card. With the concept of data mining, previous
information about the transaction is extracted then on the basis of extracted information, Hidden Markov Model consider
whether the current transaction is either fraudlent or genuine. This paper shows how advanced data mining techniques and
HMM can be combined successfully to detect the credit card fraud
Keywords: Credit Card , Internet,Data Mining, HMM, Online Shopping, Offline Shopping
INTRODUCTION
In day to day life credit cards are used for purchasing goods
and services. The most accepted payment mode is credit
card for both online and offline. For online transaction it uses
virtual card and for offline transaction it uses physical card.
In today’s world, credit card provides cashless shopping at
every shop.It will be the most convenient way to do online
shopping, paying bills etc. Therefore, risks of fraud transaction using credit card has also been increasing.To purchase
goods in physical transaction,Credit cards will insert into payment machine at merchant shop. In a physical-card based
purchase, the cardholder presents his card physically to
a merchant for making a payment. To carry out fraudulent
transactions in this kind of purchase, an attacker has to steal
the credit card. In online payment mode, attackers need only
little information for doing fraudulent transaction. In this purchase method ,transactions will be done through Internet or
telephone. To commit fraud in these types of purchases, an
attacker simply needs to know the card details.The only way
to detect this kind of fraud is to analyze the spending patterns
and to figure out any inconsistency.
HIDDEN MARKOV MODEL
In a Hidden Markov model, the state is not directly visible,
but output, dependent on the state, is visible. Each state has
a probability distribution over the possible output tokens.
Therefore the sequence of tokens generated by an HMM
gives some information about the sequence of states.A Hidden Markov Model is a finite set of states; each state is linked
with a probability distribution. Transitions among these states
are governed by a set of probabilities called transition probabilities. Credit card fraud detection based on Hidden Markov
Model, which does not require fraud signatures and still it is
capable to detect frauds just by bearing in mind a cardhold-
er’s spending habit. With the help of Hidden Markov Model
we will be able to find out the fraudulent transaction by using
spending profiles of user. This works on the user spending
profiles which can be divided into major three categories:
· Lower profile
· Middle profile
· Higher profile
It keeps record of spending profile of the card holder by both way, either offline or online. Hence, Hidden Markov Model is a perfect solution for addressing detection of fraud transaction through credit card.
DATA MINING
Data Mining is used to detect fraud for its effectiveness. Data
mining is a process that uses a variety of data analysis tools to
discover patterns and relationships in data that may be used
to make a valid prediction. The six basic steps of data mining
process are defining the problem, preparing data, exploring
data, building models, exploring and validating models, deploying and updating models. Data mining is a field at the
intersection of computer science and statistics, is the process
that attempts to discover patterns in large sets. The overall
goal of the data mining process is to extract information from
a data set and transform it into an understandable structure
for further use. Aside from the raw analysis step, it involves
database and data management aspects, data preprocessing, model and inference considerations,interestingness,met
rics,complexity considerations,post-processing of discovered
structures, visualization, and online updating.
TECHNIQUE AND ALGORITHM USED
Technique that is used for the detection of credit card fraud
based on the concept of data mining is Knowledge Data Discovery (KDD).
PARIPEX - INDIAN JOURNAL OF RESEARCH X 1
Volume : 2 | Issue : 3 | March 2013
KDD:
Many people treat data mining as a synonym.for popularly
used term KDD. The process involves in KDD are:
· Data Cleaning
· Data Integration
· Data Selection
· Data Transformation
· Data Mi ning
· Pattern Evaluation
· Knowledge Presentation
ISSN - 2250-1991
then it will direct to give permission for transaction. If the detected transaction is fraudulent then the Security information
form will arise. It has a set of question where the user has
to answer them correctly to do the transaction. These forms
have information such as personal, professional, address;
dates of birth, etc are available in the database. If user entered information will be matched with database information,
then transaction will be done securely, else user transaction
will be terminated.
RESULT
In this section, it is shown that fraud detection will be checked
on previous transactions and also calculate percentage of each
spending profile (low, middle and high) based on total number
of transactions. In Table, list of all transactions are shown:
No. of Transaction
1st
2nd
3rd
4th
5th
6th
7th
8th
9th
10th
An algorithm called as HMM is used for the detection of credit
card fraud.This algorithm primarily focus on three categories:
· Lower Profile
· Middle Profile
· High Profile
This algorithm performs prediction analysis for the fraud detection. Prediction symbol for this model is V { l, m,h},where
l stands for lower profile,m stands for middle profile and h
stands for higher profile.E.g. If card holder perform a transaction as Rs 2000 and card holders profile groups as l (low) =
(Rs 0,Rs 1500], m (middle) = (Rs 1500, Rs 5000], and h (high)
= (Rs 5000, up to credit card limit], then transaction which
card holder want to do will come in middle profile group.So
the corresponding profile group or symbol is M and V (2) will
be used.Credit card holder purchase various product with different amount in various period of time. It uses the deviation
in a purchasing amount of latest 5 transaction sequence (and
adding one new transaction in that sequence) which is one of
the possibilities related to the probability calculation.If model
does not have data of last 5 transactions in the initial stage,
in that case, model will ask to the cardholder to feed basic information during transaction about the cardholder such as address, annual income,date of birth,email id etc.On the basis of
this information HMM model obtained relative data of transaction for further verification on spending profile of cardholder.
Amount
140
125
15
5
10
125
15
120
10
280
No. of transaction
11th
12th
13th
14th
15th
16th
17th
18th
19th
20th
Amount
210
550
800
110
135
118
20
148
141
6
The most recent transaction is placed at the first position and
correspondingly first transaction is placed at the last position
in the table .The pattern of spending profile of the card holder
is shown in below Figure based on all transactions done.
The percentage calculation of each spending profile (low, medium and high) of the card holder based on price distribution
range as mentioned earlier is shown in below Figure
VERIFICATION OF FRAUD TRANSACTION:
All the information about credit card such as Credit card number, CVV number, credit card Expiry month and year, name
of credit card holder etc will be checked with credit card database. If cardholder 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. If user
credit card has less than 5 transactions then it will directly ask
to provide personal information to do the transaction. Once
database of 5 transactions will be developed, then fraud detection system will start to work. By using this observation, determine users spending profile. The purchase amount will be
checked with spending profile of user. By transition probabilistic calculation based on HMM, it concludes whether the transaction is real or fraud. If transaction may be concluded as
fraudulent transaction then user must enter security information. This information is related with credit card (like account
number, security question and answer which are provided at
the time of registration). If transaction will not be fraudulent
2 X PARIPEX - INDIAN JOURNAL OF RESEARCH
CONCLUSION
In this paper it has been discussed that how Hidden Markov
Model will facilitate to stop fraudulent transaction through
credit card. With the help of HMM, based on three categories such as low profile, middle profile, high profile and on
the previous transactions of credit card holder our proposed
model determine whether the current transaction is genuine
or fraud. If fraud transaction occurs then it terminate the current transaction else continue with the transaction. Our proposed model suggest that if the previous transaction of a card
holder is not present then information related with credit card
like account number, security question and answer must provided by the cardholder. The Hidden Markov Model makes
the processing of detection very easy and tries to remove the
complexity.
Volume : 2 | Issue : 3 | March 2013
ISSN - 2250-1991
REFERENCES
1. Ghosh, S., and Reilly, D.L., 1994. Credit Card Fraud Detection with a Neural-Network, 27th Hawaii International l Conference on Information Systems, vol. 3 (2003),
pp. 621- 630. | 2. Aleskerov, E., Freisleben, B., and Rao, B., 1997. CARDWATCH: A Neural Network Based Database Mining System for Credit Card Fraud Detection,
Proceedings of IEEE/IAFE: Computational Intelligence for Financial Eng.(1997), pp. 220-226. | 3. Syeda, M., Zhang, Y. Q., and Pan, Y., 2002 Parallel Granular Networks
for Fast Credit Card Fraud Detection, Proceedings of IEEE International Conference on Fuzzy Systems, pp. 572-577 (2002). | 4. Data Mining- Concepts And Techniques
by Han and Kamber. | 5. Hidden Markov Models, Theory and Applications by Przemyslaw Dymarski | 1. Ghosh, S., and Reilly, D.L., 1994. Credit Card Fraud Detection with a Neural-Network, 27th Hawaii International l Conference on Information Systems, vol. 3 (2003), pp. 621- 630. | 2. Aleskerov, E., Freisleben, B., and Rao, B.,
1997. CARDWATCH: A Neural Network Based Database Mining System for Credit Card Fraud Detection, Proceedings of IEEE/IAFE: Computational Intelligence for
Financial Eng.(1997), pp. 220-226. | 3. Syeda, M., Zhang, Y. Q., and Pan, Y., 2002 Parallel Granular Networks for Fast Credit Card Fraud Detection, Proceedings of
IEEE International Conference on Fuzzy Systems, pp. 572-577 (2002). | 4. Data Mining- Concepts And Techniques by Han and Kamber. | 5. Hidden Markov Models,
Theory and Applications by Przemyslaw Dymarski |
PARIPEX - INDIAN JOURNAL OF RESEARCH X 3