An Improved Algorithm to Detect Credit Card Fraud

ABSTRACT
Due to a rapid advancement in the electronic commerce technology, the use of credit cards has dramatically increased. 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 survey on the present techniques available for detecting fraud in credit card is presented as a review paper. Fraud detection involves identifying Fraud as quickly as possible once it has been done. Fraud detection methods are continuously developed to defend criminals in adapting to their strategies. The transaction is classified as normal, abnormal or suspicious depending on this initial belief. Once a transaction is found to be suspicious, belief is further strengthened or weakened according to its similarity with fraudulent or genuine transaction history using Bayesian learning.

Keywords
Fraud Detection Techniques, Credit cards, Supervised Techniques, Unsupervised Techniques

1. INTRODUCTION
The popularity of online shopping is growing day by day. According to an ACNielsen study conducted in 2005, one-tenth of the worlds population is shopping online. Credit card is the most popular mode of payment. As the number of credit card users rises world-wide, the opportunities for attackers to steal credit card details and subsequently, commit fraud are also increasing. Credit-card-based purchases can be categorized into two types: 1) physical card and 2) virtual card. In a physical-card based purchase, the cardholder presents the 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 the second kind of purchase, only some important information about a card such as card number, expiration date, secure code etc. is required to make the payment. Such purchases are normally done on the Internet or over the telephone. To commit fraud in these types of purchases, a fraudster simply needs to know the card details [2]. The most common method of payment for online purchase is credit card. Now days, fraud in credit cards is increasing day by day. There has been a growing amount of financial losses due to credit card frauds as the usage of the credit cards become more and more common. Security means to use credit card safely and avoid the occurrence of fraud. The purpose of security is to avoid fraudulent usage of credit cards. Fraud cases due to lost cards, stolen cards, application fraud, counterfeit fraud, mail-order fraud and non- received issue (NRI) fraud are found. For decreasing these frauds, security with credit cards is needed. The credit card details should be kept private. If credit card details are leaked then it is difficult to secure credit card. There are several ways to steal credit card details such as phishing websites, steal/lost credit cards, counterfeit credit cards, theft of card details, intercepted cards
1

etc [26]. For security purpose, the above things should be avoided. The credit card security is needed for the detection of valid and invalid number of transactions. For this purpose credit card validation check detects errors in sequence of numbers and detects valid and invalid no. easily. Most fraudulent transactions result from stolen card numbers rather than the actual theft of card. So, keep credit card safely. Frauds committed over Internet such as online credit card frauds become the most popular ones because of their nature. In online fraud, the transaction is made remotely and only the cards details are needed. A manual signature, a PIN or a card imprint are not required at the time of purchase. Most of the time, the genuine cardholder is not aware that someone else has seen or stolen the card information. The only way to detect this kind of fraud is to analyze the spending patterns on every card and to figure out any Inconsistency with respect to the usual spending patterns. Fraud detection based on the analysis of existing purchase data of cardholder is a promising way to reduce the rate of successful credit card frauds. Transactions made with payment cards such as credit cards, prepaid cards, debit cards and smart phones can be considered as fraud. Credit card fraud can be defined as Unauthorized account activity by a person for whom the account was not intended. Operationally, this is an event for which action can be taken to stop the abuse in progress and incorporate risk management practices to protect against similar actions in the future [1]. In simple terms, Credit Card Fraud is defined as when an individual uses another individuals credit card for personal reasons while the owner of the card and the card issuer are not aware of the fact that the card is being used. And the persons using the card has not at all having the connection with the cardholder or the issuer and has no intention of making the repayments for the purchase they done. Fraud detection methods are continuously developed to defend criminals in adapting to their strategies. The development of new fraud detection methods is made more difficult due to the severe limitation of the exchange of ideas in fraud detection. Data sets are not made available and results are often not disclosed to the public. The fraud cases have to be detected from the available huge data sets such as the logged data and user behavior. At present, fraud detection has been implemented by a number of methods such as data mining, statistics, and artificial intelligence. Fraud is discovered from anomalies in data and patterns. Fraud cases discovered from anomalies in data and patterns. Fraud cases due to lost cards, stolen cards, application fraud, counterfeit fraud, mail-order fraud and non received issue (NRI) fraud are found. 2. Types of Frauds: Various types of frauds in this paper include credit card frauds, telecommunication frauds, and computer intrusions, Bankruptcy fraud, Theft fraud/counterfeit fraud, Application fraud, Behavioral fraud . 2.1 Credit Card Fraud: Credit card fraud has been divided into two types: 1) Offline fraud: Offline fraud is committed by using a stolen physical card at call center or any other place. 2) On-line fraud: On-line fraud is committed via internet, phone, shopping and web or in absence of card holder.
2

Telecommunication Fraud: The use of telecommunication services to commit other forms of fraud. Consumers, businesses and communication service provider are the victims. Computer Intrusion: Intrusion is defined as the act of entering without warrant or invitation; that means potential possibility of unauthorized attempt to access Information, Manipulate Information Purposefully. Intruders may be from any environment, an outsider (Or Hacker) and an insider who knows the layout of the system [24]. Bankruptcy Fraud: This column focuses on bankruptcy fraud. Bankruptcy fraud means using a credit card while being absent. Bankruptcy fraud is one of the most complicated types of fraud to predict [24]. Theft Fraud/ Counterfeit Fraud: In this section, the focus is on theft and counterfeit fraud, which are related to one other. Theft fraud refers to the other person who is not the owner of the card. As soon as the owner give some feedback and contact the bank, the bank will take measures to check the thief as early as possible. Likewise, counterfeit fraud occurs when the credit card is used remotely; where only the credit card details are needed [25]. Application Fraud: When someone applies for a credit card with false information that is termed as application fraud. For detecting application fraud, two different situations have to be classified. When applications come from a same user with the same details, that is called duplicates, and when applications come from different individuals with similar details, that is termed as identity fraudsters. Phua et al [7] describes application fraud as demonstration of identity crime, occurs when application forms contain possible, and synthetic (identity fraud), or real but also stolen identity information (identity theft). Internal Fraud: Banking sector allows their employees to access customer data. The data is the same information needed to access online banking to customer accounts. So the fraud can be done easily by an employee. Instead of this, financial institutions should require a password or PIN for net banking, and the password or PIN should be stored in the format of encrypted [6].

3. INTRODUCTION TO TYPES OF SOLUTIONS FOR THE FRAUD

Fraud and identity theft can be serious personal and financial challenges. The law limits financial liability to just $50, and many credit card companies will even waive that amount. Still, fraud and especially identity theft can cause a lot of frustration. If the fraud is suspected, heres how the problem is identify and help can be found. How to Deal with credit card fraud? Fraud generally refers to the unauthorized use of credit card accounts. Usually fraud is discovered when a credit card is lost or stolen, when unfamiliar charges on the billing statement are found, when calls or letters about transactions that have not been made, when Contacted by the credit card Companys fraud department to question a charge. If the fraud is suspected on the account, then one should contact the credit card company immediately. The credit card company will be able to help in verifying the fraud, remove the charges which have not been used by the card holder or any authorized person, close
3

down the account to prevent more fraudulent transactions and issue a new account number and new card, and transfer old information to the new account. Its also a good idea to check credit report to be sure theres nothing else that looks suspicious. In most cases, the involvement of law enforcement will be coordinated with the financial institution. How to Deal with identity theft? Identity theft is a particular type of fraud in which a thief uses the personal information to set up new accounts or get other benefits in the name of cardholder. Though its not as common as other types of fraud, it can be more challenging and costly to correct. Some signs of identity theft are: cardholder is not receiving the bills or other mail, receives credit card, being denied credit for no apparent reason, getting calls or letters about things that were not transaction by credit cardholder, being served court papers or arrest warrants for things in which there is no involvement of cardholder. Never assume that such unexplained occurrences are just a mistake always look into the details to find out for sure.

4. LITEATURE SURVEY
The detection of fraud is a complex computational task and still there is no system that surely predicts any transaction as fraudulent. They just predict the likelihood of the transaction to be a fraudulent. The properties of a good fraud detection system are: 1. It should identify the frauds accurately 2. It should detecting the frauds quickly 3. It should not classify a genuine transaction as fraud Outlier detection is a critical task in many safety critical environments as outliers indicate abnormal running conditions from which significant performance degradation may result. We can categorize and analyze a broad range of outlier detection methodologies as either supervised or unsupervised approaches.

4.1 Unsupervised Outlier Detection Technique

An unsupervised outlier detection technique does not make any assumption about the availability of labeled data. This method simply seek those accounts, customer etc, whose behavior is unusual [4]. Unsupervised methods are useful in applications where there is no prior knowledge as to the particular class of observations in a data set. An advantage of using unsupervised methods over supervised methods is that previously undiscovered types of fraud may be detected. There are some techniques which were used now a day they are as follows: 4.1.1 Peer Group Analysis [4] - Peer Group Analysis (PGA) is an unsupervised method for monitoring behavior over time in data mining [3].
4

The most important task of PGA method is to identify peer groups for all the target observations (objects). The tool detects individual objects that begin to behave in a way distinct from objects to which they had previously been similar. Each object is selected as a target object and is compared with all other objects in the database, using either external comparison criteria or internal criteria summarizing earlier behavior patterns of each object. Based on this comparison, a peer group of objects most similar to the target object is chosen. The tool is intended to be part of the data mining process, involving cycling between the detection of objects that behave in anomalous ways and the detailed examination of those objects. PGA method was used in detecting credit card fraud by changing the length of the time windows subsequent to that used initially to determine the peer group. 4.1.2 Break Point Analysis [4] - Break Point Analysis is another unsupervised outlier detection tool that we are developing for behavioral fraud detection. A break point is an observation or time where anomalous behavior is detected. Break point analysis operates on the account level, comparing sequences of transactions (their amount or frequency) to detect a change in behavior for a particular account. In break point analysis, we have a fixed length moving window of transactions: as a transaction occurs so it enters the window and the oldest transaction in the window is removed. An advantage of break point analysis is that there is no requirement of balanced data (i.e. data summarized at fixed time points, e.g. weekly), as the transactions between different accounts are not compared; anomalous sequences of events that may indicate fraudulent behavior can be identified. 4.1.3 K-Means Clustering technique [6] - K-Means clustering is a simple and efficient method to cluster the data. Initially, the numbers of cluster K, and Centroid values are determined. Any random objects as the initial Centroid or the first K objects can also serve as the initial Centroid. This technique is a nonhierarchical method initially takes the number of objects equal to the final required number of clusters. Iterate until stable (= no object move group): 1. Place K points into the space represented by the objects that are being clustered. These points represent initial group centroids. 2. Assign each object to the group that has the closest centroid. 3. When all objects have been assigned, recalculate the positions of the K centroids. 4. Repeat Steps 2 and 3 until the centroids no longer move. This produces a separation of the objects into groups from which the metric to be minimized can be calculated.

Fig 1: Block diagram of K- means algorithm Process [6]

4.2

Supervised Outlier Detection Technique

Supervised outlier detection techniques assume the availability of a data set which has labeled instances for the normal as well as the outlier class. Supervised methods to detect fraudulent transactions can be used to discriminate between those accounts or transactions known to be fraudulent and those known (or at least presumed) to be legitimate. Classification techniques such as statistical discriminant analysis and neural networks can be used to discriminate between fraudulent and non-fraudulent transactions to give transactions a suspicion score. Supervised methods are only trained to discriminate between legitimate transactions and previously known fraud [4]. On doing the literature survey of various methods for fraud detection to detect credit card fraud there are multiple approaches like Gass algorithm, Bayesian Networks, Hidden Markov Model, Genetic Algorithm, A Fusion approach using Dempster-Shafer Theory and Bayesian learning, Decision tree, Neural Network, Logistic Regression 4.2.1 Gass Algorithm [1] - This algorithm is a combination of genetic algorithm and scatter search. Genetic algorithms are inspired from natural evolution. The basic idea is that the survival chance of stronger members of a population is larger than that of the weaker members and as the generations evolve the average fitness of the population gets better. The less fit members of this generation are eliminated and the fitter members are selected as the parents for the next generation. This procedure is repeated until a pre-specified number of generations have passed, and the best solution found until then is selected. 4.2.2 Bayesian Networks [1] - For the purpose of fraud detection, two Bayesian networks to describe the behavior of user are constructed. First, a Bayesian network is constructed to model
6

behavior under the assumption that the user is fraudulent (F) and another model under the assumption the user is a legitimate (NF). The fraud net is set up by using expert knowledge. The user net is set up by using data from non fraudulent users. During operation user net is adapted to a specific user based on emerging data. By inserting evidence in these networks and propagating it through the network, the probability of the measurement less than two is obtained. This means, it gives judgments to what degree observed user behavior meets typical fraudulent or non fraudulent behavior. On the one hand, Bayesian networks allow the integration of expert knowledge, which we used to initially set up the models. On the other hand, the user model is retrained in an unsupervised way using data. Thus our Bayesian approach incorporates both, expert knowledge and learning. 4.2.3 Hidden Markov Model [1] - A Hidden Markov Model is a double embedded stochastic process which used to model much more complicated stochastic processes as compared to a traditional Markov model. If an incoming credit card transaction is not accepted by the trained Hidden Markov Model with sufficiently high probability, it is considered to be fraudulent transactions [5]. Baum Welch algorithm is used for training purpose and K-means algorithm for clustering. HMM stores data in the form of clusters depending on three price value ranges low, medium and high. The probabilities of initial set of transaction have chosen and FDS checks whether transaction is genuine or fraudulent. Since HMM maintains a log for transactions it reduces tedious work of employee but produces high false alarm as well as high false positive. The initial choice of parameters affects the performance of this algorithm and, hence, they should be chosen carefully. 4.2.4 Genetic Algorithm [1] - Genetic algorithms, inspired from natural evolution was first introduced by Holland (1975). Genetic algorithms are evolutionary algorithms which aim at obtaining better solutions as time progresses . Fraud detection problem is classification problem, in which some of statistical methods many data mining algorithms have proposed to solve it. Among decision trees are more popular. Fraud detection has been usually in domain of Ecommerce, data mining [21]. GA is used in data mining mainly for variable selection [22] and is mostly coupled with other DM algorithms [23]. And their combination with other techniques has a very good performance. GA has been used in credit card fraud detection for minimizing the wrongly classified number of transactions [23]. And is easily accessible for computer programming language implementation, thus, make it strong in credit card fraud detection. But this method has high performance and is quite expensive. 4.2.5 A Fusion Approach Using Dempster-Shafer Theory and Bayesian Learning [1] Dempster-Shafer Theory basically proposes Fraud Detection System using information fusion and Bayesian learning in which evidences from current as well as past behavior are combined together and depending on certain type shopping behavior establishes an activity profile for every cardholder. It has advantages like: - high accuracy, processing speed, reduces false alarm, improves detection rate, applicable in E-commerce. But one disadvantage of this approach is that it is
7

highly expensive. DempsterShafer theory and Bayesian learning is a hybrid approach for credit card fraud detection which combines evidences from current as well as past behavior. Every cardholder has a certain type of shopping behavior, which establishes an activity profile for them. This approach proposes a fraud detection system using information fusion and Bayesian learning of so as to counter credit card fraud. The FDS system consists of four components, namely, rule-based filter, DempsterShafer adder, transaction history database and Bayesian learner. The transaction is classified as suspicious or suspicious depending on this initial belief. Once a transaction is found to be suspicious, belief is further strengthened or weakened according to its similarity with fraudulent or genuine transaction. 4.2.6 Decision Tree [1] - Decision trees are statistical data mining technique that express independent attributes and a dependent attributes logically AND in a tree shaped structure. Classification rules, extracted from decision trees, are IF-THEN expressions and all the tests have to succeed if each rule is to be generated [19]. Decision tree usually separates the complex problem into many simple ones and resolves the sub problems through repeatedly using [19] [20]. Decision trees are predictive decision support tools that create mapping from observations to possible consequences. Decision tree methods are C5.0, C&RT and CHAID. The advantage of applying the data mining techniques including decision trees and SVMs to the credit card fraud detection problem for the purpose of reducing the banks risk. 4.2.7 Neural Network [1] - Fraud detection methods based on neural network are the most popular ones. An artificial neural network 16] [17] consists of an interconnected group of artificial neurons .The principle of neural network is motivated by the functions of the brain especially pattern recognition and associative memory. The neural network recognizes similar patterns, predicts future values or events based upon the associative memory of the patterns it was learned. It is widely applied in classification and clustering. The advantages of neural networks over other techniques are that these models are able to learn from the past and thus, improve results as time passes. They can also extract rules and predict future activity based on the current situation. There are two phases in neural network [18] training and recognition. Learning in a neural network is called training. There are two types of NN training methods supervised and unsupervised. In supervised training, samples of both fraudulent and non fraudulent records are used to create models. In contrast, unsupervised training simply seeks those transactions, which are most dissimilar from the norm. On other hand, the unsupervised techniques do not need the previous knowledge of fraudulent and non fraudulent transactions in database. NNs can produce best result for only large transaction dataset. And they need a long training dataset. 4.2.8 Logistic Regression [1] - The two advanced data mining approaches, support vector machines and random forests, together with the well known logistic regression, [14] as part of an attempt to better detect (and thus control and prosecute) credit card fraud. The study is based on real-life data of transactions from an international credit card operation. It is wellunderstood, easy to use, and remains one of the most commonly used for data-mining in practice. It thus provides a useful baseline for comparing performance of newer methods. Supervised learning methods for fraud detection face two challenges.
8

1. The first is of unbalanced class sizes of legitimate and fraudulent transactions, with legitimate transactions far outnumbering fraudulent ones. 2. The second problem in developing supervised models for fraud can arise from potentially undetected fraud transactions, leading to mislabeled cases in the data to be used for building the model. For the purpose of this study, fraudulent transactions are those specifically identified by the institutional auditors as those that caused an unlawful transfer of funds from the bank sponsoring the credit cards. These transactions were observed to be fraudulent expose. Our study is based on real-life data of transactions from an international credit card operation. The transaction data is aggregated to create various derived attributes.

5. ANALYSIS OF EXISTING TECHNIQUES

Esakkiraj et al [6] has design a model with the sequence of operations in online transaction by using HIDDEN MARKOV MODEL (HMM) and decides whether the user act as a normal user or fraud user. This paper predicts the fraudulent during the transaction time and prevents the money transfer. The main objective is to ensure that the genuine transactions should not be rejected. In this, clustering method is used which is one of the data mining technique. Srivastava et .al. [2] has implemented a model to show the sequence of credit card transaction process and presents the experimental results which shows the experimental results which shows the effectiveness of the system and demonstrate the usefulness of learning the spending profile of cardholders. Comparative studies reveal that the Accuracy of the system is close to 80 percent over a wide variation in the input data. The system is also scalable for handling large volumes of transactions. In the paper, Patidar [8] et al has tried to detect fraudulent transaction through the neural network using genetic algorithm. Artificial neural network work as a human brain, though it is impossible for the artificial neural network to imitate the human brain, yet neural network and brain, depend on the neurons for working, which is the small functional unit in brain as well as ANN. Genetic algorithm are used for making the decision about the network topology, number of hidden layers, number of nodes that will be used in the design of neural network for the credit card fraud detection problem. In this paper feed forward back propagation algorithm is used for future results. In this paper different technique that is being used to execute credit card fraud how credit card fraud impact on the financial institution as well as merchant and customer, fraud detection technique used by VISA and MasterCard. Neural network is a latest technique that is being used in different areas due to its powerful capabilities of learning and predicting that Back propagation Network is the most popular learning algorithm to train the neural network so in this paper BPN is used for training purpose and then in order to choose those parameter to perform neural network as accurately as possible, we use genetic algorithm, and using this combined Genetic Algorithm and Neural Network (GANN) try to detect the credit card fraud successfully. The research paper by Subashini et al [8], aims to enhance and evaluate the fraudulence in credit card approval process using the classification models based on decision trees (C5.0 & CART), Support Vector Machine (SVM) using SMO, Bayes Net and Logistic Regression. Five
9

methods to detect fraud are presented here. The result shows that Logistic Regression method is provides 73.1% success rate as compared to other algorithms (Bayes Net success rate is 72.4%, Decision tree (CART) gets the highest success rate 74.1%). Hence from the results, it is understood that while making decisions on classifying customers combination of different classification models need to be used to make correct decision about a customer. Phua et al. [10] determines the comparative study of technical and review articles on credit card fraud detection. The neural network and Bayesian network comparison study conducted by Maes uses the STAGE algorithm for Bayesian networks and back propagation algorithm for neural networks in credit transactional fraud detection. Comparative results show that Bayesian networks were more accurate and much faster to train, but Bayesian networks are slower when applied to new instances. Ezawa and Norton [12] developed Bayesian network models in four stages with two parameters. They argue that regression, nearest-neighbor, and neural networks are too slow and decision trees have difficulties with certain discrete variables. Bagheri et. al. [11] evaluated the performance of an ensemble of three classifiers, each trained on different feature sets. Three ancient shape description methods, including shape, signature, Zernike moments, and generic Fourier descriptors, were used to extract informative features from logo images and each set of features was fed into an individual classifier. In order to reduce recognition error, the Dempster-Shafer combination theory was employed to fuse the three classifiers trained on different sources of information. The classification results of the individual classifiers were compared with those obtained from fusing the classifiers by the Dempster-Shafer combination method. Using ensemble methods for the classification of logo images is effective, though different combination methods would show different performances, and even some combination of base classifiers and ensemble methods would deteriorate the performance of the best single classifier. However, as demonstrated by the following experiments, by using the DS fusion method, the classification performance was significantly increased compared with single classifiers trained by a specific set of features. 6. CONCLUSION AND FUTURE WORK Credit card fraud has become more and more rampant in recent years. Fraud detection methods are continuously developed to defend criminals in adapting to their strategies. In Fraud detection, identifying Fraud as quickly as possible once it has been done through fraud detection techniques, is now becoming easier and faster. The techniques which were studied here, through which credit card fraud can be detected quickly and fast and the crime can be stopped. The Future work is to design an improved technique which will be much better than the available techniques. 7. REFERENCES [1] Suman and Nutan Review Paper on Credit Card Fraud Detection , International Journal of Computer Trends and Technology (IJCTT) volume 4 Issue 7July 2013.

10

[2]

A. Srivastava, A. Kundu, S. Sural, and A. K. Majumdar, Credit Card Fraud Detection Using Hidden Markov Model, IEEE transactions on dependable and secure computing, vol. 5, no. 1, january-march 2008.

[3] Z. Ferdousi and A. Maeda Unsupervised Outlier Detection in Time Series Data, Proceedings of the 22nd International Conference on Data Engineering Workshops (ICDEW'06) 2006 IEEE. [4] [5] R.J. Bolton and D.J. Hand Unsupervised Profiling Methods for Fraud Detection, Department of Mathematics Imperial College London {r.bolton, d.j.hand}@ic.ac.uk J.S. Mishra, S. Panda, and A. Kumar Mishra, A Novel Approach for Credit Card Fraud Detection Targeting the Indian Market IJCSI International Journal of Computer Science Issues, Vol. 10, Issue 3, No 2, May 2013 ISSN (Print): 1694-0814 | ISSN (Online): 1694-0784 www.IJCSI.org

[6] S. Esakkiraj and S. Chidambaram, A Predictive Approach for Fraud Detection Using Hidden Markov Model International Journal of Engineering Research & Technology (IJERT) Vol. 2 Issue 1, January- 2013 C. [7] Phua, D. Alahakoon and V. Lee, Minority Report in Fraud Detection: Classification of Skewed Data, ACM SIGKDD Explorations Newsletter, vol. 6, no. 1, pp. 50-59, 2004. [8] R. Patidar and L. Sharma , Credit Card Fraud Detection Using Neural Network NCAI2011, 13-14 May 2011, Jaipur, India International Journal of Soft Computing and Engineering (IJSCE) ISSN: 2231-2307, Volume-1, Issue-NCAI2011, June 2011. [9] B. Subashini and Dr. K. Chitra Enhanced System for Revealing Fraudulence in Credit Card Approval, International Journal of Engineering Research & Technology (IJERT) Vol. 2 Issue 8, August 2013 ISSN: 2278-0181.

[10] L. Phua, V. lee , K. Smith and R. Gayler, A Comprehensive Survey of Data Miningbased Fraud Detection Research, School of Business Systems, Faculty of Information Technology, Monash University, Clayton campus, Wellington Road, Clayton, Victoria 3800, Australia. [11] M. A. bagheri, Q. Gao and S. Escalera Logo recognition based on the DempsterShafer Fusion of Multiple Classifi ers, Advances in Artificial Intelligence, Lecture Notes in Computer Science Volume 7884, 2013, pp 1-12. [12] K.J. Ezawa and S.W. Norton Constructing Bayesian Networks to Predict Uncollectible Telecommunications Accounts Intelligent Systems and Their Applications Volume 11 Issue 5, October 1996,IEEE Educational Activities Department Piscataway, NJ, USA

11

[13] Maes, S., Tuyls, K., Vanschoenwinkel, B. & Manderick, B. (2002), Credit Card Fraud Detection using Bayesian and Neural Networks, Proc. of the 1st International NAISO Congress on Neuro Fuzzy Technologies. [14] S. Bhattacharyya, S. Jha, K. Tharakunnel and J. Christopher Westland, Data mining for credit card fraud: A comparative study, Decision Support Systems 50 pp. 602 613,2011.

[15] C.C. Aggarwal, Outlier Analysis publication: @Springer, Part of springer science plus business media. [16] S. Benson Edwin Raj and A. A Portia Analysis on Credit Card Fraud Detection Methods, IEEE International Conference on Computer, Communication and Electrical Technology; (2011). (152-156). [17] Ray-I Chang, Liang-Bin Lai, Wen-De Su, Jen-Chieh Wang and Jen-Shiang Kouh Intrusion Detection by Back propagation Neural Networks with Sample-Query and Attribute-Query, Research India Publications; (2006). (6-10). [18] T. Guo and Gui-Yang Li Neural Data Mining For Credit Card Fraud Detection, IEEE, Proceedings of the Seventh International Conference on Machine Learning and Cybernetics; (2008). [19] S. Benson Edwin Raj and A. Annie Portia, Analysis on Credit Card Fraud Detection Methods, International Conference on Computer, Communication and Electrical Technology ICCCET2011, 18th & 19th March, 2011. [20] Y. Sahin and E. Duman, Detecting Credit Card Fraud by Decision Trees and Support Vector Machines, International Multiconference of Engineers and computer scientists March, 2011. [21] K. Rama Kalyani and D. Uma Devi , Fraud Detection of Credit Card Payment System by Genetic Algorithm, International Journal of Scientific & Engineering Research Volume 3, Issue 7, July-2012. Bidgoli, B. M., Kashy, D., Kortemeyer, G. and Punch, and W. F Predicting student performance: An Application of data mining methods with the educational web based system LON-CAPA, In Proceedings of ASEE/IEEE frontiers in education conference. . (2003).

[22]

[23] E. Duman and M. Hamdi Ozcelik Detecting credit card fraud by genetic algorithm and scatter search, Elsevier, Expert Systems with Applications, (2011). 38; (13057 13063). [24] L. Delamaire, H. Abdou and J. Poinon , Credit card fraud and detection techniques: a review, Banks and Bank Systems, Volume 4, Issue 2, 2009.
12

[25] K. Chaudhary, J. Yadav and B. Mallick, A review of Fraud Detection Techniques: Credit Card, International Journal of Computer Applications (0975 8887) Volume 45 No.1, May 2012. [26] European payment cards fraud report, Payments, Cards and Mobiles LLP & Author, (2010). [27] Richard J. Bolton and David J. Hand, Unsupervised Profiling Methods for Fraud Detection, Technical Report (Department of Mathematics, Imperial College, London), 2002. [28] Bolton, R. J. and D.J. Hand, "Unsupervised Profiling Credit Control VII, Edinburgh, UK, 5-7 Sept (2001). [29] Aleskerov, E., B. Freisleben, and B. Rao CARDWATCH: A Neural Network Based Database Mining System for Credit Card Fraud Detection, Computational Intelligence for Financial Engineering, Proceedings of the IEEE/IAFE, 220-226. [30] Blunt G. and D.J. Hand The UK credit card market. Technical Report, Department of Mathematics, Imperial College, London. [31] Bolton R.J. and D.J. Hand Peer Group Analysis. Technical Report, Department of Mathematics, Imperial College, London. [32] Brause, R., T. Langfsord and M. Hepp Neural Data Mining for Credit Card Fraud Detection, Proceedings 11th IEEE International Conference on Tools with Artificial Intelligence.

13